• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    method and time synchronization system of signals from respective measuring devices. the present invention relates to a phase time synchronization between measuring devices not sharing the same clock for their respective signal sampling is carried out by a time labeling of samples of the signals in time blocks followed by an adjustment of the phase values of components of interest to the signals in the time blocks regrouped so that they refer to common time references between the measuring devices. the labeling is carried out with a synchronization signal available on the measuring devices, supplemented by account values provided by a meter operated by a reference clock for each measuring device.
    公开号:BR112012024003B1
    申请号:R112012024003-0
    申请日:2011-03-16
    公开日:2020-03-10
    发明作者:Sylvain Riendeau；François Léonard；Patrick Picher；Michel Gauvin；Hugo Bertrand；Louis Dupont
    申请人:HYDRO-QUéBEC；
    IPC主号:
    专利说明:

    Invention Patent Descriptive Report for "METHOD AND SYSTEM FOR SYNCHRONIZATION OF SIGNAL PHASE TIME FROM RESPECTIVE MEASURING DEVICES".
    FIELD OF THE INVENTION
    [0001] The present invention relates to a method and system of time synchronization of signals from the respective measuring devices.
    TECHNICAL STATUS
    [0002] Several systems, processes and techniques need to synchronize phase measurements between different conversion devices that do not share the same clock signal for their respective sampling. Several existing systems perform sampling for a watch controlled by a common time reference. This approximation requires a material performing a feedback (closed phase locking loop or "phase-lock-loop") between the phase of the sampling clock and the time reference, which moves costs. In addition, this approach limits noise reduction strategies since the noise of the sampling clock and that of the reference clock are mixed with the servo clock control errors.
    SUMMARY
    [0003] An objective of the invention is to propose a method and a system of time synchronization of signals from the respective measuring devices that have a low cost compared to the existing techniques and which is potentially more accurate.
    [0004] Another objective of the invention is to propose a replacement of the usual equipment used for the time synchronization of phase measurements with a temporal labeling of samples of the measurement signals, followed by calculations correcting the sampling frequency, the temporal labeling at which the phase is referenced, as well as values of time and phase characteristic of each component of interest in the signals.
    [0005] In accordance with an aspect of the invention, a method of synchronizing signal phase time from respective measuring devices is proposed, comprising the steps of: for each measuring device: receiving a synchronization signal available on each measuring device measurement; producing a reference clock signal having a higher rate than the synchronization signal; operating a counter in response to the reference clock signal to produce account values; complete the synchronization signal by the account values provided by the meter; selecting at least one block of time having a finite number of samples in the signal from the measuring device; establish temporal locations of at least two samples from each time block with the complete synchronization signal; estimate a phase value and a time characteristic of at least one component of the signal from the measuring device in each time block; assign to each time block a time label derived from the completed synchronization signal; and producing data representative of at least one component, the phase value, the time characteristic, the time locations and the time label for each time block; and for the set of measuring devices: regroup the data relating to the blocks of time having close time labels as the same time labels serving as common time references; and calculating new phase values of at least one component in the time blocks according to the respective common time references and corresponding time locations for the phase time synchronization of the signals from the measuring devices.
    [0006] In accordance with another aspect of the invention, a system of time synchronization of signals from respective measuring devices is proposed, comprising: for each measuring device, a phase unit comprising: a receiver for receiving a synchronization signal available in each phase measurement unit; a clock for producing a reference clock signal having a higher rate than the synchronization signal; and a processing unit; and for the set of measuring devices, a phase processing unit comprising a processing unit; the processing unit of each phase measurement unit being configured to receive the signal from the corresponding measuring device, receive the synchronization signal, receive the reference clock signal, provide a counter operating in response to the reference clock signal to produce account values, complete the synchronization signal with the account values provided by the accountant, select at least one block of time having a finite number of samples in the signal from the measuring device, establish temporal locations of at least two samples of each time block with the completed synchronization signal, estimate a phase value and a time characteristic of at least one component of the signal from the measuring device in each time block, and produce data representative of at least one component, the value phase, time characteristic, and time locations; the processing unit of one of each phase measurement unit and the phase processing unit that is configured to assign to each time block a time label derived from the completed synchronization signal, the time label being part of the data for each time block; and the processing unit of the phase processing unit being configured to regroup the data relating to the time blocks having close time labels with the same time labels serving as common time references, and to calculate new phase values of at least one component in them time blocks according to the respective common time references and the corresponding time locations for the phase time synchronization of the signals coming from the measurement units.
    [0007] The following sets out some possibly preferred features of the invention that should be considered in a non-restrictive manner.
    [0008] The present invention aims at a phase time synchronization between two or several measuring devices that do not share the same clock for their respective sampling of signals to be measured. The measuring devices digitize one or more analog signals by blocks of time or without interruption. Phase time synchronization consists of adjusting the phase values so that they refer to a common time reference between the different devices. Phase synchronization refers to one or more spectral components. Alternatively, a correction of the frequency of each component can also be carried out. A spectral component can be the result of a Fourier transform, small wave analysis or any other process leading to assigning a phase value to a signal component. The conversion units associated with the measuring devices can be an element of a permanent, portable or mobile system.
    [0009] The synchronization signal that represents the common time reference preferably comes from a GPS receiver, but it can also come from a carrier wave generated locally and transmitted by radio, by electrical conduction or by any other means (for example IEEE 1588) and converted numerically if necessary. [00010] The invention aims especially at systems, processes and techniques that exploit a distributed measurement system and requiring a high precision of the measured phase synchronization and this at a low material cost. For example, in the area of electrical transmission networks, PMU ("Phase Measurement Unit") that perform synchronized phase measurements is required to have a fast response time, and this at the expense of cost and accuracy. On the other hand, the gain in precision brought about by the method according to the invention and its low cost especially allows economic monitoring of the dielectric state of transformer passages. In the industrial area, engine control or other remote processes, can profit from the invention to synchronize the different equipment (paper industry and conveyors especially). It is a question of comparing the phases of different measures such as those resulting from an angular sensor or any other sensor giving information on the cyclical status (displacement, speed, acceleration, tachometer) of an organ participating in the process. In the area of vibration measurement, and more particularly of modal analysis over large extensions as well as on a drilling platform, the invention allows for an exact synchronization of the phase measurements carried out by different devices located in different locations. In the area of location, such as sonar and radar, the invention allows an accurate estimate of the orientation of one or several wave fronts from a distribution of fixed or moving receivers.
    [00011] In short, the phase time synchronization method according to the invention uses several PCU phase measurement units and at least one PPU phase processing unit. unit "). Each PCU is connected to one (or several) measuring device as a sensor that can be part of the PCU or be fixed outside on another equipment, and can include a processing unit, a GPS receiver that provides a synchronization signal , a reference clock and a communication interface. The PPU can comprise a processing unit and a communication interface. The following steps can be performed on each PCU: the signal from the sensor to which the PCU is associated is previously digitized if necessary and forwarded to its processing unit; the processing unit receives a synchronization signal from the GPS receiver as well as a signal from the reference clock; the processing unit has a counter that receives the signal from the reference clock and increments its count in response to a time stamp as each clock strikes the signal of the reference clock; the counter is preferably set to zero in response to a time stamp on the sync signal provided by the GPS receiver as a sync signal transition, and the processing unit memorizes the "OPPS" value corresponding to an account value of the counter, that is, the time marker on the synchronization signal; the processing unit assigns an account value to some samples of the digitized signal; the processing unit selects a time block that has a finite number of samples (e.g. between 16 and 65,536 samples) in the digitized signal; the processing unit retains at least one and preferably two account values linked to samples in the selected block as well as the stored OPPS value; the processing unit retains a time reference e.g. {hour: minute: second} and optionally {day: month: year} provided by the GPS receiver for at least one of the selected block samples; the processing unit performs a transformation of the digitized (temporal) signal in a representation area where the components of interest of the digitized signal are distinguishable; the processing unit estimates and retains a phase value in the same way as a time characteristic value of one or more components of interest observed in the selected block, such as its frequency, its scale or its periodicity; and the processing unit transmits data representative of the retained values to the PPU or continues processing.
    The following steps can be performed on each PCU or PPU: the OPPS value is used to assign respective time values to the account values that have been retained, linked to the samples in the selected block; from the time values of the retained account values, the processing unit (of the PCU or PPU, as the case may be) assigns a time characteristic value (e.g. frequency, scale, periodicity) to each component of interest; from one of the time values of the retained account values, the processing unit assigns a time label to the selected block; if a time reference of the phase value does not correspond to a position of the assigned time label, from the time values of the retained account values, the processing unit adjusts the phase value of each component to match a determined time reference by the temporal label of the selected block; and if the preceding steps are performed by a PCU, the processing unit of the PCU transmits to the PPU data representative of the selected block's time label and phase value in the same way as the time characteristic value of one or more components observed in the selected block.
    [00012] The following steps can be performed in the PPU: among the blocks from several PCUs, the processing unit regroups those that have a close temporal label according to a predetermined similarity criterion; the processing unit converts the phase values of each component of interest for each grouped block according to a common time reference determined by a common time label assigned to the grouped blocks; and the processing unit provides the common time label, the converted phase value, in the same way as the time characteristic value of one or more components observed in the blocks grouped with the common time label, thus ending the synchronization of time. phase of the measurement signals.
    [00013] PCUs and PPUs can be a permanent, portable or mobile element of a system. The PPU can be integrated into the same processing unit as a PCU. Several PCUs can share a GPS receiver, a reference clock, and / or a communication interface.
    [00014] In the case of a sensor that provides an analog signal, it passes through an analog digital converter (CAN) that chooses as a sample and digitizes the signal. Before reaching the processing unit of a PCU, the signal can pass through protection and conditioning circuits. The conditioning circuit can include an amplifier, a filter, and / or an integrator, or derivator. A spectral folding filter ("anti-aliasing filter") can be included in the conditioning circuit or the converter.
    [00015] The digitized signal can undergo galvanic isolation before being received by the processing unit of a PCU. One or more digitization subunits can be connected to the processing unit of a PCU via a common bus.
    [00016] The processing unit of a PCU can assign an account value to all samples of the digitized signal and not to some samples.
    [00017] In the case of a continuous digitization of the signal coming from a sensor, the signal is preferably cut into successive blocks of time and may or may not be overlapped in time.
    [00018] The digitized signal may suffer a limitation before a block is thus extracted.
    [00019] The processing unit of a PCU can retain two account values of the counter, consisting, for example, respectively of the conta_0 corresponding to the first sample of the selected block of time and the conta_N-1 corresponding to the last sample of that same block. The values of conta_0 and conta_N-1 correspond to remote samples in the block or in the proximity of the selected time block.
    [00020] If there is a temporary loss of the synchronization signal, this can be the last OPPS value that is retained by the PCU for its calculation. The PCU can be configured to detect a loss of synchronization, label the corresponding blocks and generate the counter accordingly. The loss of synchronization can be detected, for example, by an account difference of several standard deviations reactively to a moving average of the last OPPS. The PCU can transmit the synchronization status of each block to the PPU, being aware of any loss of synchronization. The PCU can also give the PPU the status of the GPS receiver transmitted by the GPS receiver to the PCU.
    [00021] The OPPS value can be used in conjunction with the time reference value provided by the GPS receiver to assign a time sample to the counter account values that have been retained. The time sample can mark the beginning, middle or end of the block selected in the PCU, or another predetermined position of the block if desired.
    [00022] The time reference value provided by the GPS receiver can be expressed in another time unit if desired, e.g. in seconds from a given date.
    [00023] The processing unit of a PCU can also estimate and retain an amplitude value of one or more components observed in the selected block and transmit it to the PPU for processing with the other data.
    [00024] The grouping of the blocks in the PPU can be done with a given time interval, done as a response to a command or done every time a new data cohort arrives from the PCU.
    [00025] The common time reference can have a predetermined value, one of the values of the time labels of the grouped data cohort, or a label corresponding to an average time of the time labels of the grouped data cohort.
    [00026] A spectral window can be applied over the selected block submitted to the transform in order to limit an error introduced by a spectral overlap of the components. The component (s) can be the result of a Fourier transform, small wave analysis, stationary cycle analysis or any other process leading to assigning a phase value to a signal component. In the three specific cases, it will be the subject of frequency, scale and periodicity as a characteristic of a component.
    [00027] In the case of a passive radar location, a block can be selected according to a key (master) recognizable in the signal by the different PCUs. For example, the switch can correspond to a distinct RF transition from an AM, FM, TV or other station and have a good signal to noise ratio. Each reflection creates a reproduction of the key. In the PCU, the temporal position of a key or its reproduction is then established at two levels, summarily according to its envelope and finely according to its phase. A comparison of the keys captured by different PCU allows to associate the keys from the same emission, deducting the relative delays and Doppler limits. The blocks can be lifted in advance on more than one RF signal band in order to explore the statistical coincidence of the locations obtained to increase strength and accuracy.
    [00028] The method makes it possible to increase the precision of phase time synchronization by reducing the temporal dispersion of the synchronization signal, with the following steps: [00029] transmitting from a PCU to the PPU and keeping in the PPU, for each block, the values of account used in block processing, including the OPPS value; apply a digital filter on the successive OPPS values accumulated by the PPU and thus generate new OPPS values; renew the account values of the PCU counter from the new filtered OPPS values; use the new OPPS values to assign time values to account values that have been retained; use the old account values to find the phase origin and time characteristic values of the components observed in the blocks having the same time label; recalculate, from the new OPPS and account values calculated in (b) and (d) and old values found in (e), the phase and time characteristic values of the components observed in the blocks having the same time label; and recalculating the time label of the regrouped blocks according to the filtered OPPS values or converting the phase values according to the old common time label.
    [00030] Steps (b), (c), (d) and (e) can be replaced by a correction of the phase value and time characteristic from values resulting from an application of a digital filter eg FIR (" Finite Impulse Response ") or IIR (" Infinite Impulse Response "), on the successive OPPS values accumulated by the PPU.
    [00031] Step (f) can be replaced by reducing a time drift of the reference clock signal with the following steps: generating, through interpolation OPPS values, a time transfer function that converts the new account values resulting from the filtered OPPS values in account values that would correspond to those issued by a counter powered by a constant frequency clock setting a constant OPPS value named later OPPSP; apply the transfer function to the new account values in order to correct them; use the new OPPSP value to assign a time value to account values that have been retained; and recalculating, from the OPPSP value and the corrected account time values, phase values and time characteristics of the components observed in the blocks having the same time label.
    [00032] In the event of a temporary loss of the synchronization signal, the PPU can resume calculating a PCU considering the OPPS accounts valid before and after the loss of synchronization. A linear interpolation of the missing accounts can be performed before proceeding with the recalculation of the time labels, phase values and temporal characteristic of the components observed in the blocks from the PCU reached by the loss of synchronization signal.
    [00033] Interpolation can be applied over a series consisting of a successive sum of a series of OPPS values.
    [00034] The renewed account values may not contain zeroing for a certain duration in order to give an account of continuous progression.
    [00035] An observed component can be a stationary cycle characterized by an amplitude, a periodicity and a phase, the periodicity being processed as the inverse of the frequency. [00036] The PPU can be configured to identify measures likely to be partial due, for example, to a climatic phenomenon such as rain or a phenomenon having a similar effect on the measures, in order to for example reject them and not use them in calculations serving to establish diagnoses based on the time synchronization of phase measurements according to the invention, such as for monitoring transformer passages (ie possible defect conditions). In such a case, the PPU can carry out the following steps: make successive assessments of time differences of phase angle differences according to the phase values of the components of the regrouped blocks; calculate standard deviations on successive assessments; and invalidate a measurement according to the corresponding standard deviation if it exceeds a predetermined rejection threshold.
    [00037] The PPU can therefore be configured to stop the calculations that serve to establish a diagnosis until the measurements are again valid, meaning that the disturbing phenomenon that causes significant transients on the differential measurements is finished. Successive evaluations can be made on time difference of tangent to phase angle differences if desired.
    BRIEF DESCRIPTION OF THE DRAWINGS
    [00038] A detailed description of the preferred embodiments of the invention will be given below with reference to the following figures: [00039] Figure 1 is a schematic diagram showing an architectural example of a system according to the invention. [00040] Figure 2 is a schematic diagram showing a phase measurement unit (PCU) according to the invention.
    [00041] Figure 3 is a schematic diagram showing a configuration shared between different phase measurement units (PCU) according to the invention.
    [00042] Figure 4 is a schematic diagram showing a processing carried out by a phase measurement unit (PCU) according to the invention.
    [00043] Figure 5 is a schematic diagram showing a phase measurement processing unit (PCU) of signals captured over a transformer passage according to the invention. [00044] Figure 6 is a schematic diagram showing a system according to the invention for monitoring transformer passages. DETAILED DESCRIPTION OF THE PREFERRED ACHIEVEMENTS [00045] In the context of this disclosure, the term "time characteristic" means a frequency, scale, periodicity or similar parameter of a component of interest in a signal. [00046] With reference to figure 1, an example of architecture of a system according to the invention is shown where several phase 1 measurement units (hereinafter also called PCU by "computing unit phase") are connected to a unit of measurement phase 2 processing (moreover also called PPU by "phase processing unit") via a local communication network 3 which may itself be connected to a wider network 4. Other PCUs (not shown) can be added through the wider network 4. In a possible configuration of the system according to the invention, the role of each PCU 1 is to estimate the amplitude, phase and frequency values of one or more spectral components of a measured signal by associating with these data a time label, while the role of the PPU 2 is to process the data coming from the PCU 1 in order to regroup them with the same time labels in order to finish the phase time synchronization of the measurement signals and transmit the data thus processed for example for use by a device asking for identical data or storage in a database.
    [00047] With reference to figure 2, a PCU 1 can be equipped with a digitization unit that includes a sensor 5 or another measuring device, integrated or fixed outside on a device (not shown), to produce a measurement signal in relation to a characteristic of the equipment to be monitored. Before a digital conversion of the analog signal by a converter 8, the signal can pass through a protection circuit 6 and a conditioning circuit 7 (e.g. amplifier, filter, integrator, derivator ...). A folding filter ("anti-aliasing filter") can be included in conditioning circuit 7 or converter 8. One or more scanning units can be connected to a processing unit 11 via a common bus 10. Preferably, each The scanning unit has galvanic isolation 9 in front of the bus 10 connecting it to the rest of the system. In the case of a digital output sensor, converter 8 is not necessary. The processing unit 11 receives a synchronization signal as well as a signal from a reference clock 13. The synchronization signal preferably comes from a GPS receiver 12, but it can also come from a carrier wave generated locally and transmitted by radio, electrical conduction or any other appropriate means of transmission if determined. The synchronization signal can take the form of a pulse per second, or another form providing a time stamp allowing time synchronization in a pre-established time unit. The processing unit 11 has a counter 14 which receives the signal from the reference clock 13 and increments its count in response to a time stamp such as a clock beat on the clock signal 13. The clock 13 has chosen stability specifications depending on the intended application and the material environment (eg temperature and stability of the feed). The counter 14 is preferably set to zero at a transition of the synchronization signal of the GPS receiver 12. The measurement signal can be digitized without interruption or by time blocks. In case of continuous digitization, the measurement signal is cut into successive blocks of time, which may or may not be overlapped in time. In a possible configuration of a PCU 1, the processing unit 11 evaluates the amplitude, frequency (or other time characteristic) and phase of one or more spectral components of a digitized signal block and assigns a time label to the block. These operations can be carried out by a processor 27 or by a similar circuit with memory in the processing unit 11. The data resulting from the processing are transmitted to the PPU 2 (illustrated in figure 1) via a communication interface 15.
    [00048] With reference to figure 3, several PCU 1 can share the same GPS receiver 12, the same reference clock 13 and the same communication interface 15.
    [00049] Referring to figure 4, there is shown a processing that a PCU 1 (as illustrated in figure 1) can perform. For practical reasons, the signal chosen as sample 16, the spectral window 17 and the count value 21 are presented as continuous values while in reality they are successions of discrete values. The signal chosen as sample 16 can correspond to the digitized signal or a limiter of the digitized signal. In order to limit the error introduced by the spectral overlap of the components ("spectral leakage", cf. J. Harris, "On the use of Windows for harmonic analysis with the discrete Fourier transform", Proceeding of IEEE, vol. 66, n ° 1, pp. 51-83,1978), preferably a spectral window 17 is applied over the signal block subjected to a transform 18. The spectral window 17 will preferably have a shape close to a Gaussian and will display a high level of rejection of the lateral lobes. Transform 18 converts the temporal signal into spectral information where the energy of a tone is regrouped at a frequency with an amplitude and a phase. Preferably, this transform will correspond to a fast Fourier transform ("FFT"). The spectral information is submitted to a component estimator 19. This estimator 19 finds the amplitude, phase and frequency of the spectral line order of one or several tones. The PCU 1 can stop processing at that point and transmit to the PPU 2 (as shown in figure 1) the frequency and phase amplitude values as well as the time stamp 20 of the GPS receiver 12 and three values of the counter 14 connected to the clock reference 13 (as shown in figure 2). Preferably, the three account values 21 of counter 14 consist respectively of an account_0 22 corresponding to the first sample of the selected time block, an account_N-1 24 corresponding to the last sample of that same block and an OPPS 23 value of counter 14 in o moment of zeroing it at the moment of the last transition of the synchronization signal (eg one pulse per second) coming from the GPS receiver 12. The time label can mark the beginning, the middle or the end of the selected block, or another point block specific if determined. It should be noted that another change such as a counting direction transition can be performed instead of zeroing the counter 14. Or, the counter 14 may not undergo any change as the counter's OPPS account value 14 that marks a time stamp provided by the synchronization signal is retained. In such a case, the rate of the timestamps provided by the synchronization signal will preferably be higher than a counter recording cycle 14 in order to simplify the processing of the timestamps. Only the OPPS 23 account value and another account value can be transmitted if determined.
    [00050] The following processing can be performed on a PCU 1 (as illustrated in figures 1 and 2) or PPU 2 (as illustrated in figure 1). For the frequency estimation, the phase adjustment with a time reference given as soon as the generation of time labels, the estimator 19 explores the information 20 transmitted to it by the GPS receiver 12 as well as three account values 21 of the counter 14 connected to the reference clock 13. Information 20 on the GPS receiver 12 corresponds to the current hour / minute / second (and possibly day / month / year) label in progress. The OPPS value 23 allows to characterize the frequency of the reference clock 13 and thus to give a time value to each account 21. For example, for a reference clock of 100 MHz, the account 21 has an OPPS value 23 which can vary typically some units around 100 million samples. A Késimode account will then have the time label.
    [00051] The particular case exposed in figure 4 where the counter 14 is set to zero between the first and the last sample of the selected block is to be considered if it is the case. In this situation, the "second" value is increased for the calculation of the time samples after the counter 14 is set to zero, and the possible cases of exceeding the values of seconds, minutes and hours (and days, months, years if applicable) are processed. If there is a temporary loss of the synchronization signal, this can be the last OPPS value that is retained by the PCU for its calculation in equation (1) and the k value can significantly exceed the OPPS value in order to count several seconds. The k value will include the wrap around of counter 14 if applicable.
    [00052] If the reference clock 13 is chosen for its stability, on the contrary, the other watches in the different converters 8 can substantially derive. The account values conta_N-1 24 and conta_0 22 correspond to the time labels tCOnta_N-i and tCOnta_o according to the transformation given in (1). The time labels tCOnta_N-1 and tConta_o allow to characterize the average frequency of each converter 8. Thus the equation [00053] where N is the number of samples in the block, allows the conversion in Hertz of the frequency expressed in spectral line number 1 of a tone. We emphasize that if the estimator 19 performed an interpolation, i is not an integer.
    [00054] A phase value only makes sense if that value is referenced in a time position. For example, the classic Fourier transform algorithm references the phase relative to the first temporal sample of the block. Whether in PCU 1 or PPU 2, it is sometimes necessary to convert the phase value of a tone to another time reference. Whether in radians (3) [00055] the phase correction applied when moving from time reference t1 to reference t2, t1 and t2, the time labels being expressed in seconds.
    [00056] PCU 1 can then transmit to the PPU 2 the amplitude, phase and frequency values calculated according to (2) as well as a single time label calculated according to (1) by which the phase is referenced according to (3) if required. You can also transmit a GPS status and a "synchro" or "hors synchro" status of the OPPS value used in your calculation. Optionally, as explained below, a reduction in noise from the GPS receiver 12 and a compensation for deviations from the reference clock 13, which consists of adding to the analysis results of each block the three values conta_0 22, OPPS 23 and conta_N-1 24 of the counter 14 and transmitting the whole to the PPU 2 can be performed.
    [00057] Regarding the phase of stationary cycle phenomena, the present method of time synchronization can be applied as follows. The objective is to position angularly or temporally, which is the same, a stationary cycle according to a determined time reference. A first way is to use in the synchronization method a small wave transform with a small wave similar to the stationarity cycle present. Temporally, the phase zero then corresponds to the beginning of a cycle when the value 2n (N-1) / N corresponds to the phase of the last sample of the cycle. A second mode uses a harmonic analysis where the stationarity cycle is considered with a sum of harmonic components. The phase of each component is then taken into account and synchronized individually by the method according to the invention. A harmonic position can be deduced from each of these phases and from the set of these positions, according to the weighting of the choice (harmonic amplitude, power, amplitude x frequency ...), the group limit corresponding to the set is estimated. of harmonics.
    [00058] Referring again to figure 1, the phase processing unit (PPU) 2 constitutes a processing unit 25 that includes a processor 28 or a similar circuit with memory, which collects data from the different PCU 1 by means of a communication interface 26. The processing unit 25 regroups (or selects) in the first time all data that has a close time label. This grouping can be done at a given time interval, done in response to a command or made at each arrival of a new cohort of data from the PCU 1. The selection of nearby time labels ensures that the measurement blocks corresponding treated in the different PCU 1 are approximately overlapping in time. This overlap allows you to stay close to the maximum accessible accuracy. Maximum accessible precision is defined as the Cramer-Rao terminal (cf. C. Rife and R. Boorstyn, "Single-tone parameter estimation from discrete-time observation", IEEE Transactions on Information Theory, IT-20, n ° 5 , pp.591-598.
    1974) for the estimation of a continuous tone to which is added (1) the contribution of the temporal labeling errors in the estimation of the frequency and phase of the tone and (2) the contribution of the time overlap deviations between the blocks of the different measures having taking into account that the tone varies slowly in amplitude and frequency. Processing unit 25 of PPU 2 calculates new phase values for each measure according to a common time reference applying equation (3). The common time reference should be as close as possible to the time labels being processed if you want to minimize errors on the adjustment of the phase values. This time reference can be a predetermined value, one of the values of the labels of the processing cohort, or a label corresponding to the average time of the labels of the processing cohort. [00059] Referring also to figure 2, referring to the option to reduce the noise of the GPS receiver 12 and to compensate for deviations from the reference clock 13 of a PCU 1, the three values count_0 22, OPPS, 23 and conta_N-1 24 of counter 14 transmitted by a PCU 1 can be used to find the origin values (k, Θ) -position of an account and phase value. Let us determine that the noise of the GPS receiver 12 and the deviations of the reference clock 13 fix spectral distributions almost in reverse: the noise corresponding to the temporal differences of the synchronization given by the GPS receiver 12 is located for the short periods of the order of the second, when the deviations of reference clock 13 are important for longer periods, in the order of several tens of minutes. A FIR or IRR filter can therefore be applied over the successive OPPS values accumulated by the PPU 2 in order to reduce the noise of the GPS receiver 12. The filtered result of the risks of the GPS receiver 12 gives a good estimate of the behavior of the reference clock 13. It is a question of renewing the account values of meter 14 from new filtered OPPS values. To compensate for deviations from the reference clock 13, it is a question of finding the transformation of the time record that presents a continuous OPPS: the time curve resulting from an interpolation of the sum of the OPPS values is then seen as the inverse transfer function of the sought. The time transfer function is applied to the renewed account amounts. Equations (1), (2) and (3) are then taken up with new filtered and corrected values. Let us determine that other methods of phase synchronization, such as synchronous labeling, do not allow this fine correction. However, the counterpart of this correction is an additional limit on the final delivery of the result, this limit corresponding minimally to the half width of the FIR filter applied to the OPPS values plus a certain calculation time.
    [00060] The synchronized phase values, such as corrected frequency values and amplitude values can be transmitted abroad via the communication interface 26. Outside, these data can be explored in contexts as varied as the monitoring forecast, equipment diagnostics or process control.
    [00061] The data processing functionalities of PPU 2 can be integrated into one, several or all PCU 1 if necessary to reduce material costs and expand the method's field of application. The material features and data processing features of the PCU 1 and PPU 2 units can be merged into one unit.
    [00062] As a non-limiting example, the system and method according to the invention allow online monitoring of transformer passages.
    [00063] With reference to figure 5, an example of installation of a system according to the invention with a sensor 29 of a passage 30 of a transformer is shown. The PCU 1 has, in the illustrated case, 6 acquisition channels that can be connected to as many sensors if desired, eg via the common bus 10. Overvoltage protection devices 31, 6 are included in the passage 29 sensor 30 and at the entrance of each channel, and can be interconnected by a shielded twisted pair 39. A channel is preferably composed of a conditioning circuit 7 with a shunt 32, a sigma-delta 8 digital analog converter (24-bit CAN) and an isolation circuit digital 9. This assembly has a high immunity to noise and a rise in voltage relative to ground, and the intrinsic synchronization of converters 8 to over-sampling when commanded by a common clock. In addition, the oversampling converter 8 allows the use of a simpler folding filter in the conditioning circuit 7, which helps to minimize disturbances in the phase angle and frequency amplitude measurements of the network.
    [00064] The system can be configured to calculate the phase and amplitude of all channels every minute. The results can be saved in a local storage device over the network (not shown) and transferred outside the site once a day to a central database (not shown) for analysis.
    [00065] A relative measurement method that uses two or more passages in parallel over the same electrical phase and which calculates the ratio of the amplitudes and the tangent of the phase angle between the fundamental components of the pass-through insulation currents is recommended. The internal dielectric insulation of high voltage bushings constitutes a stack of conductive sheets and interleaved dielectric sheets. A deterioration is characterized by a damage to one or more dielectric sheets, which lately generates a short circuit between sheets. The damage of the dielectric modifies the phase relationship between the current of the 60Hz fundamental component of the network (or another operating frequency if applicable) passing through the insulation of the passage and the voltage at the terminals of that insulation. The leaves in partial or total short-circuit directly influence the value of the equivalent capacity of the passage, affecting the amplitude of the 60Hz component current. The monitoring of the internal dielectric insulation of passages implies a monitoring of the temporal evolution of the phase and current values from the moment of the installation of the equipment, in the hypothesis that the passages were in good condition for this placement. Typically, for a passage constituting a hundred leaves, a diagnosis of defect in one of the passages is attributed when the tangent value of the voltage / current phase angle (tanõ) varies from 0.005 or more. Likewise, a deviation of more than 1% over the time evolution of the relative amplitudes between two bushings means the presence of at least one short circuit in the sheets. The tangent of the synchronized phase difference between two passages is sensitive to any change in the power / dispersion factor of one of the passages, and the amplitude ratio is sensitive to changes in the capitance of one of the passages. If relative measurements are made using three elements of the equipment in parallel, then the defective passage can be identified. The interphase voltage asymmetry will not influence the interpretation, since the applied voltage is essentially the same for all equipment connected in parallel. The method of summing the currents of three passages connected to the three phases, for example, to the primary of a transformer, is less sensitive. On the contrary, in the case of several sheets in short-circuit, this method allows to confirm and / or determine the diagnosis. For example, in the case of an installation comprising only two transformers, the method of summing the currents makes it possible to target the transformer with the defective passage if the defect is accentuated over one of the passages.
    [00066] With reference to figure 6, the processing unit 11 of each PCU 1 (as illustrated in figure 2) is configured to calculate a current phasor of sensor 29 (as illustrated in figure 5) and transfer the time-labeled measurements by GPS for the processing unit of a PPU 2, which in the illustrated case is a pass diagnostics unit, using a communication network 3 that may already exist in such facilities. Each PCU 1 is connected to the sensors 29 of the transformer passages 34 and installed in its box. A GPS antenna 35 is connected to the synchronization receiver (e.g. GPS receiver 12 as illustrated in figure 2) of each PCU 1.
    [00067] PPU 2 can be located in a substation control building. You can receive the PCU 1 phasors, calculate the Atanõ (time differences of the tangents of differences in phase angles) and amplitude relationships, save the data in memory, perform trend analysis, issue a local diagnosis and transmit alarms if necessary to a maintenance center 36 eg connected to network 3 via a platform 37 and an enterprise network 38. PPU 2 can also allow maintenance personnel to explore and analyze relevant data from a distance.
    [00068] Local network 3 in the course of the substation connects the control construction to the boxes of high-voltage transformers 34. The use of an existing network allows a significant reduction in the costs of implementation and maintenance.
    [00069] Referring again to figure 5, the shunt values 32 are preferably chosen to adapt the nominal current of the sensor 29, which is a function of the capacity (pF) and the voltage of the passage, in the input range of the converter 8. PCU 1 can constitute an FPGA ("Field-Programmable Gate Array") 33 used for the temporal labeling of converter samples 8 in the resolution of the reference clock 13. Counter 14 implemented in FPGA 33 is operated by the clock 13 eg at low temperature deviation timed at 125 MHz. Counter 14 is preferably set to zero by the signal of a pulse per second (1PPS) from the GPS receiver 12. The intervals between pulses are also used to calculate the clock frequency Reference 13. Since the signal noise provided by the GPS receiver 12 used for sampling, processing can be carried out to reduce the two noises as explained above.
    [00070] In addition to the time labeling, the FPGA 33 can also be used with the first phase of signal processing, buffer and interface with processor 27. Processor 27 is responsible for calculating the phase angle and amplitude of the time-labeled signals coming from from converters 8. PCU 1 transmits the resulting phasor data through its Ethernet 15 optical fiber port to the PPU 2 as shown in figure 6.
    [00071] The effect of harmonics on the voltage can cause a significant error in digital processing that is based on a zero-pass detection. In the present case, the delta sigma converter 8 converts the signal e.g. at 50k samples per second. In order to reduce the processing power requirements, the digital signal can be filtered by a low step FIR and limited to ten. With a sampling rate of 5kHz, a fast Fourier transform (FFT) processes 83.3 samples per cycle. The maximum number of cycles processed by an FFT is fixed by the stability of the network frequency, and the minimum by the type of spectral window and the intended rejection of components under synchronization.
    [00072] The product of the spectral window, the FFT and the parameter estimation of the spectral component are performed by CPU 27. Since reference clock 13 is more accurate than that of converter 8, the sampling frequency is estimated from of the last 1PPS account value. The phase is referenced on the time label generated from account values and GPS data. Atanó values are estimated by PPU 2 (as shown in figure 6) considering the time labels and the respective frequencies of reference watches 13.
    [00073] The use of a lateral lobe spectral window with a high rejection rate allows a rejection that exceeds 90 dB of harmonic and under synchronous components. The precision of spectral estimation is a function of the signal-to-noise ratio: [00074] in dB under the spectral lobe where G represents the processing gain factor of the spectral window, N represents the number of temporal samples processed by the FFT, and ao / aw represents the relationship between peak signal amplitude and white noise RMS amplitude value.
    [00075] Choosing a Blackman-Harris window, for N = 4096 samples, the signal-to-noise ratio under the spectral lobe is SNR (dB) = ao / ov / (dB) + 32.6dB, expressed in dB. The phase standard deviation for a passage A, expressed in degrees, is linked by [00076] In the case of uncorrelated noise between measurements, the Atanó standard deviation is the quadratic sum of the standard deviations of two phase estimates. Since σβ / ~ θσβ, the Atano standard deviation is [00077] expressed in dB. Taking into account the analog white noise and the equivalent noise from the converter 8, the typical Atano accuracy exceeds -100 dB or 0.001%. The corresponding dispersion limit is 27 ns, in the same range as the GPS noise. This precision can be achieved with a standard galvanic system. The measurement accuracy is not disturbed by the reference clock 13 or the deviation of the network frequencies. The contribution of the Atanõ dispersion of the PPU 2 is the quadratic sum: [00078] where aops = Aígps ■ 2 π ■ 60Hz and Aígps is the GPS noise. In the case of non-correlated noise between PCU 1, GPS noise is the quadratic sum of noise from two different GPS 12 receivers. Typical measured values take a contribution of GPS noise At2GPs of 50 ns or 0.0018% in equation (7).
    [00079] PPU 2 can be configured to identify measures likely to be partial due to, for example, a climatic phenomenon such as rain or a phenomenon having a similar effect on the measures, in order to for example reject them and not use them in the calculations that serve to establish a diagnosis of monitoring passages. As indicated above, important transients can be observed on a differential measurement Atanõ between a 60 Hz wave resulting from a capacitive coupler over two passages. Some of these transients may be associated with the presence of rain. However, automated monitoring without interruption requires a search for a trend free of inopportune transients such as those caused by rain. The proposed method calls for a local temporal stability of the measures estimated from a standard deviation over several successive Atanõ estimates. A measure can be considered invalid when the standard deviation exceeds a pre-established threshold level. The threshold level can be set manually or determined automatically by a simple statistical calculation when sufficient measures are accumulated. In a possible configuration of the system according to the invention, the PPU 2 calculates the successive Atão values in the fixed measurement interval eg of 2 seconds, performing eg 36 rapid Fourier transforms (TRF) of 1024 samples with an overlap of 75 % over the duration of the interval, in the moments that follow the operation of the system. The following two TRFs are distant from 256 samples. The two extreme TRFs are preferably removed to keep only the other 34. Considering the overlap of the TRFs, the successive values of Atanõ's estimate are not totally independent: the weight of statistical sampling is not 34 but rather closer to 15. A longer duration or shorter FRT can be statistically advantageous. In the presence of light precipitation a flicker of measurements of a zero average appears, suggesting that there is no significant water film and that the precipitation evaporates faster than the water supply. When precipitation increases, a continuous component will appear (i.e., zero average) in the scintillation. This component can correspond to a passage that will be wetter than the other. The statistical distribution of the standard deviation over the dispersion of instantaneous Atano measurements, calculated with a log application of the calculation in decibels before statistical binning, presents an overlap of two types of distribution. The first distribution is Gaussian and corresponds to the measurement noise in the absence of disturbance. The second distribution being the first to the right and corresponds to the disturbances attributed to the precipitations. The establishment of a detection threshold is a compromise between a sensitivity and a probability of false detention. The threshold can be set at a distance of two to three times from the standard deviation of an undisturbed value. The maximum probability is obtained by signal processing manipulations performed in a representation space where the noise is Gaussian, as is the case here. The threshold value will possibly be several times higher for low voltage side measurements that are disturbed by inverters. The threshold can initially be set with a high value as an initial value in a moving average that gradually adjusts the threshold value to the average standard deviation plus three times the standard deviation of the standard deviation estimated from the filtered values. A min / max terminal with a defect alert can limit the excursion threshold in order to guarantee strength. The analysis results of the measurements can be the result of a comparison between the threshold and the max {ETY (Atanõ), Én} where ETY represents an estimate of Atanõ and the moving average Én is simply a weighting of the standard deviation (ETY) with a forgetting factor of 25%. You can write Én = 0.75 · Én + 0.25 ^ ETYn where ETYn = ETY (tan (õn, canalx) - (tan (õn, canaly)). This means at the same time allows you to respond instantly to a sudden rise in the ETY and extend the rejection of the estimated values of some measurements after a peak in ETY, thus guaranteeing a cleaning around a peak in ETY. For multi-transformer monitoring, the excess thresholds can be combined over the different passages. The combination must accept the defect of a passage that will lead to an excess threshold for the calculations that imply this, for example, it is possible that intermittent partial discharges in the leaves of a passage may increase the dispersion of the instant ETY. probable simultaneously in two passages, you can adjust the decision to "more than one pass" = precipitations. There is no relationship between the fluctuations of the signal-to-noise ratio values and the means of Atano or the standard deviations of Atano. The standing disturbances of the 60Hz components are of low amplitude and constitute a random part and a deterministic part between two passages. The standing fluctuations of the 60Hz component are partly correlated from one passage to another for the same phase: in the comparison of the tanõ, this noise is largely eliminated by the tangent to tangent subtraction.
    [00080] The method and system according to the invention also allow monitoring the appearance of cracking in a structure such as that of a drilling platform that uses two measuring points or more in parallel installed on the structure and that calculates the amplitudes and phases of vibration modes that affect the structure. The appearance of a fissure changes the division of the bellies and knots in the structural modes of the platform and also shifts their frequencies. The monitoring of this division allows the detection of the symptoms of a crack diagnosis. The method according to the invention allows inexpensive installation without wiring between the different measuring points. Typically, the measurement points comprise triaxial accelerometers for strain gauges and displacement sensors. Preferably, time synchronization will be done by GPS. At the end of the processing carried out by the method, the amplitude, phase and frequency values of the structural modes can be compared with a digitized model to which a suspicious anomaly is inserted in order to observe the correspondence between respective modes and determines the correction of the suspected anomaly. [00081] In a context of passive sonar or radar localization, the measurements processed by the method according to the invention can come from two passive receivers or more in parallel, at least one emitter and one target, the processing then calculating the phase angle between at least one signal component emitted by the emitter and reflected by the target and calculating the fine frequency of the components emitted and reflected. The receivers and transmitters can be radio metric, ultrasonic or audible. The time synchronization will be given by GPS, a luminous electric radio wave or any other means of communication capable of providing the required synchronization signal. In addition to the information brought by the transient fronts of the wave impulses, the phase information adds precision to the respective location of the emitters, receivers and targets. Fine frequency information adds precision to the estimation of the respective speeds of emitters, receivers and targets. The method allows a more accurate measurement at a lower cost and gives access to the use of lower frequencies where, without the proposed method, the use of the wavefront would have an accuracy limited by the wavelength.
    [00082] The method according to the invention also allows the simultaneous monitoring of several measurement points located with electrical potentials, not allowing cabling of these measurement points and therefore requiring PCU 1 in the form of autonomous sensors. For example, for the monitoring of cutting chambers of high power circuit breakers, the vibration and acoustic measurement, such as the electric radio measurement, provide information rich in diagnostic symptoms of the electromechanical state of the equipment. However, in these two cases, the autonomous sensor must be incorporated in the cutting chamber to maximize the signal ratio of one cutting chamber over that of the other cameras. The sensor is therefore located at the high voltage line potential. In addition to the issue of autonomous power that can be done in different ways, synchronization is often problematic. The phase synchronization according to the invention allows to solve this problem. The measurement of interest occurs at the moment of a circuit breaker switching. The autonomous sensor is preferably configured to sleep between two switches. For reasons of energy saving, you may be required to take action before the first synchronization signal arrives (the GPS receiver 12 is out of voltage just before). The method can then result in retaining the account of two successive synchronization transitions in the temporal proximity of the analyzed block. We need that the circuit breaker chambers are connected in series between them through the same circuit breaker. The electrical radio and acoustic vibration behavior of these chambers are interconnected and need to synchronously compare the signatures, including the phase of the components that make up these signatures. The implantation of a synchronous digitalization in time as used in the PMU is not possible for the autonomous sensors because such a means would not allow a deep sleep of the sensor considering the synchronization in real time that implies an electrical consumption of several orders of magnitude greater by proposed means without counting the material costs also higher. An autonomous sensor based on a PCU 1 typically comprises inside or near a vibrating acoustic sensor such as an accelerometer, a measuring antenna, a communication antenna, a temperature measurement, a current measurement, the latter can also serve as a source of food. The processing carried out by the PPU 2 that receives the data from the PCU 1 can re-perform a phase synchronization.
    [00083] Instead of a GPS signal or other connection specifically dedicated to fine time synchronization, it is possible to explore some ambient signals available for all PCU 1, such as an electrical radio broadcast (AM, FM, TV station), an audible signal or optical. The method then involves measuring a common ambient signal as a reference to correct the risks of small time amplitudes of a first level of coarse synchronization. We are talking here about a double differential, be it a difference between a reference channel of a PCU 1 and its other channels and a difference between the values coming from two PCU 1. In one possible realization, each PCU 1 dedicates one of its channels analogous to the measurement and processing of the synchronization signal. A first coarse synchronization means, such as a simple inter-PCU 1 communication according to the IEEE 1588 standard, performs an approximate staging of the PCU 1's clocks 13 and operates the counters 14. The method requires the creation of a limit calibration table of receiving PCU 1 to a position of the fine sync source. The values in the calibration table can be estimated by a simple wave propagation calculation considering the respective PCU 1 and source positions, with the calculated limit values being relative (to PCU 1 the closest to the source for example). Two equivalent phase correction approaches can be used. In a first case, the phase measured on the synchronization reference channel is taken into account to correct the phase of the components of the other channels in PCU 1, in which case the phase correction to be brought in a phase comparison between two PCUs 1 will be a function of a phase delay value entered in the calibration table for the compared PCU 1. In a second case, the phases of components measured by two PCU 1 are subtracted, in which case the phase correction to be brought will be a function of the phase difference of the synchronization signal measured by each PCU 1 and an entered phase delay value. in the calibration table for each PCU 1. All of these phase corrections must be reset in the temporal area according to the frequency of the synchronization signal and transported back to the phase area considering the frequency of the component measured over the compared channels. In this way, the GPS price is saved but a measurement channel is lost. It is also necessary that the sampling rate is at least twice as high as that of the synchronization signal and that the inaccuracy of the first level coarse synchronization is less than the period of the synchronization signal.
    [00084] Although embodiments of the invention have been illustrated in the drawings together and described above, it seems evident to persons skilled in the art that modifications can be brought to these embodiments without departing from the invention.
    权利要求:
    Claims (22)
    [1]
    1. Method of synchronizing the phase phase of signals from respective measuring devices, characterized by the fact that it comprises the steps of: for each measuring device: receiving a synchronization signal available on each measuring device; producing a reference clock signal having a higher rate than the synchronization signal; operating a reference clock signal response counter to produce account values; complete the synchronization signal with account values provided by the accountant; selecting at least one block of time having a finite number of samples in the signal from the measuring device; establish temporal locations of at least two samples from each time block with the completed synchronization signal; estimate a phase value and a time characteristic of at least one component of the signal from the measuring device in each time block; assign to each time block a time label derived from the completed synchronization signal; and producing data representative of at least one component, the phase value, the time characteristic, the time locations and the time label for each time block; and for the set of all measuring devices: regroup the data relating to the blocks of time having close time labels according to a predetermined similarity criterion with the same time labels serving as common time references; and calculating new phase values of at least one component in the time blocks according to the respective common time references and the corresponding time locations for the phase time synchronization of the signals coming from the measuring devices.
    [2]
    2. Method, according to claim 1, characterized in that the account values provided by the meter are based on a cyclic element of the reference clock signal, the meter being operated in order to change the account value or a direction account in response to a time stamp on the synchronization signal, the sample locations being relative to the corresponding account values in the case of the time stamp.
    [3]
    3. Method, according to claim 1, characterized by the fact that it also comprises the step to receive an available time signal in each measurement device, in which the time label assigned to each time block is based on a unit of measure of time indicated by the time signal in the case of a time marker in the synchronization signal.
    [4]
    4. Method, according to claim 3, characterized in that the synchronization signal and the time signal result from the same signal.
    [5]
    5. Method, according to claim 1, characterized by the fact that the temporal characteristic of at least one component of the signal is estimated as a function of the temporal locations of the samples in the time block.
    [6]
    6. Method, according to claim 1, characterized by the fact that the estimated phase value is adjusted according to the time location corresponding to the time label of the time block.
    [7]
    Method according to claim 1, characterized in that at least one component is obtained by performing a transform of the signal from the measuring device in a representation domain where at least one component is distinguishable.
    [8]
    8. Method, according to claim 1, characterized by the fact that it also comprises, for the set of measuring devices, the steps of: accumulating the successive account values of the samples of the time blocks that coincide with time markers in the signal synchronization; and correcting the phase values and the time characteristics of at least one component observed in the blocks having the same time label from new account values that result from digital processing on the accumulated account values.
    [9]
    9. Method, according to claim 1, characterized by the fact that it also comprises, for the set of measuring devices, the steps of: making successive estimates of time differences of phase angle differences according to the phase values of the components the regrouped blocks; calculate statistical differences on successive estimates; and invalidate a measurement if the corresponding statistical difference exceeds a predetermined rejection threshold.
    [10]
    10. Phase time synchronization system of signals from respective measuring devices, characterized by the fact that it comprises: for each measuring device, a phase measurement unit comprising: a receiver to receive a synchronization signal available in each phase measurement unit; a clock for producing a reference clock signal having a higher rate than the synchronization signal; and a processing unit; and for the set of measuring devices, a phase processing unit comprising a processing unit; the processing unit of each phase measurement unit being configured to receive the signal from the corresponding measuring device, receive the synchronization signal, receive the reference clock signal, provide a counter operating in response to the reference clock signal to produce account values, complete the synchronization signal with the account values provided by the accountant, select at least one block of time having a finite number of samples in the signal from the measuring device, establish temporal locations of at least two samples of each time block with the completed synchronization signal, estimate a phase value and a time characteristic of at least one signal component from the measuring device in each time block, and produce data representative of at least one component, the value phase, time characteristic, and time locations; the processing unit of one of each phase measurement unit and the phase processing unit that is configured to assign to each time block a time label derived from the completed synchronization signal, the time label being part of the data for each time block; and the processing unit of the phase processing unit that is configured to regroup the data related to the time blocks having close time labels according to a predetermined similarity criterion with the same time labels serving as common time references, and to calculate new phase values of at least one component in the time blocks according to the respective common time references and the corresponding time locations for the time synchronization of signals from the measurement units.
    [11]
    11. System, according to claim 10, characterized in that the account values provided by the meter are based on a cyclic element of the reference clock signal, the meter being operated in order to change the account value or a direction account in response to a time deviation in the synchronization signal, the sample locations being relative to the corresponding account values in the case of the time deviation.
    [12]
    12. System, according to claim 10, characterized in that the processing unit of each phase measurement unit is configured to receive an available time signal in each measuring device, and the time label assigned to each time block be based on a unit of time measurement indicated by the time signal in the case of a time deviation in the synchronization signal.
    [13]
    13. System according to claim 12, characterized in that the synchronization signal and the time signal result from the same signal.
    [14]
    14. System, according to claim 10, characterized by the fact that the temporal characteristic of at least one component of the signal is estimated as a function of the temporal locations of the samples in the time block.
    [15]
    15. System, according to claim 10, characterized by the fact that the estimated phase value is adjusted according to the temporal location corresponding to the time label of the time block.
    [16]
    16. System according to claim 10, characterized in that the processing unit of each phase measurement unit is configured to obtain at least one component that performs a transformation of the signal from the measurement device in a representation area where at least one component is distinguishable.
    [17]
    17. System according to claim 10, characterized by the fact that a group of phase measurement units share at least the same receiver and the same clock via a common bus between at least one of the same receiver and the same clock and the processing units of the group of units of phase measurement.
    [18]
    18. The system according to claim 10, characterized in that each phase measurement unit and the phase processing unit comprise respective communication interfaces that connect to a network.
    [19]
    19. System, according to claim 10, characterized by the fact that the receiver is a GPS receiver.
    [20]
    20. System according to claim 10, characterized by the fact that it also comprises, for each phase measurement unit, at least one of: a filter to filter the signals from the measuring devices; an amplifier to amplify the signals from the measuring devices; an integrator or a tap to integrate or divert signals from measuring devices; a limiter to limit the signals coming from the measuring devices; for each measurement device that provides an analog signal, a digital analog converter to digitize the analog signal from the measurement device into a digitized signal; and galvanic isolation to isolate the phase measurement unit from the corresponding measuring device.
    [21]
    21. System, according to claim 10, characterized by the fact that each common time reference is defined by a predetermined value associated with the grouped blocks, or a label corresponding to a temporal average of the time labels of the grouped blocks.
    [22]
    22. System according to claim 10, characterized by the fact that the processing unit of the phase processing unit is configured to: make successive estimates of time differences of phase angle differences according to the phase values of the components of the blocks regrouped; calculate static differences on successive estimates; and invalidate a measurement if the corresponding statistical difference exceeds a predetermined rejection threshold.
    类似技术:
    公开号 | 公开日 | 专利标题
    BR112012024003B1|2020-03-10|METHOD AND SYNCHRONIZATION SYSTEM OF SIGNAL PHASE TIME FROM RESPECTIVE MEASURING DEVICES
    US10228409B2|2019-03-12|Fault location using traveling waves
    US10375108B2|2019-08-06|Time signal manipulation and spoofing detection based on a latency of a communication system
    US10656198B2|2020-05-19|Electric power system monitoring using high-frequency signals
    CN102495281B|2013-09-04|Method for measuring phasor frequency of power system
    US10581237B2|2020-03-03|High-frequency electric power system signal processing system
    CN103116083B|2014-12-10|Very low frequency | / low frequency | frequency range ground lightening detection positioning system
    CN106771861B|2019-01-25|Complex electric network Fault Locating Method based on wide area traveling wave energy and time difference
    CN204389910U|2015-06-10|A kind of calibration system of power information acquisition system master clock
    Zhang et al.2014|Synchrophasor time skew: Formulation, detection and correction
    CN109425844B|2022-01-18|Calibration method and system for data sampling
    CN102004441A|2011-04-06|Adaptive crystal oscillator frequency timekeeping method
    Baranski et al.2012|Evaluation of multiple ground flash charge structure from electric field measurements using the local lightning detection network in the region of Warsaw
    CN103809156A|2014-05-21|Regional high-resolution lightening radiation source locating system
    Mohamed et al.2014|A novel approach in the measurement of high frequency disturbances from very large electrical systems
    CN105759599A|2016-07-13|Calibration system and method for main station clock of power utilization information collection system
    SE1651596A1|2016-12-30|Time synchronization of intelligent electronic devices
    CN108872704A|2018-11-23|Wireless phasor measuring set and method based on Active noise cancellation algorithm
    Riendeau et al.2010|On-line monitoring of transformer bushings using a new decentralized measurement system
    US9581618B1|2017-02-28|Method and apparatus for triggered capture of voltage phase angle data on an alternating current power grid
    RU185831U1|2018-12-19|STORM DEVICE FOR REGISTRATION OF THUNDER DISCHARGES AND PLACES OF DAMAGE TO INSULATION ON VL
    Wu2005|New implementations of wide area monitoring system in power grid of China
    Dimitrijevic et al.2007|Remote wireless calibration of equipment in the distributed measurement systems
    Peretto et al.2008|A distributed system for the synchronized acquisition of fast voltage transients
    同族专利:
    公开号 | 公开日
    EP2550509A1|2013-01-30|
    AU2011232283A1|2012-10-11|
    AU2011232283B2|2016-05-12|
    CA2699596A1|2011-09-24|
    EP2550509A4|2015-04-15|
    CA2792376C|2017-10-31|
    EP2550509B1|2016-07-27|
    PL2550509T3|2017-01-31|
    BR112012024003A2|2016-08-30|
    US20130018620A1|2013-01-17|
    WO2011116479A1|2011-09-29|
    CN102859334B|2016-02-03|
    US9002672B2|2015-04-07|
    CA2792376A1|2011-09-29|
    CN102859334A|2013-01-02|
    ES2600008T3|2017-02-06|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    DE3938520A1|1989-11-21|1991-05-29|Teves Gmbh Alfred|METHOD AND SYSTEM FOR MEASURING DATA EVALUATION AND EVALUATION|
    KR960009539B1|1993-12-27|1996-07-20|Hyundai Electronics Ind|Apparatus for synchronizing time using satellite|
    SE507227C2|1996-09-16|1998-04-27|Ericsson Telefon Ab L M|Method and device for synchronizing time stamping|
    CN1163894C|1998-12-17|2004-08-25|松下电器产业株式会社|Frequence control/phase synchronizing circuit|
    US6297702B1|2000-01-10|2001-10-02|Honeywell International Inc.|Phase lock loop system and method|
    US6407699B1|2000-04-14|2002-06-18|Chun Yang|Method and device for rapidly extracting time and frequency parameters from high dynamic direct sequence spread spectrum radio signals under interference|
    US6497699B1|2000-08-09|2002-12-24|The Research Foundation Of State University Of New York|Hybrid neuroprosthesis for the treatment of brain disorders|
    US7133397B2|2001-09-07|2006-11-07|L-3 Communications Corporation|Time division duplex system utilizing global positioning system timing signals for access point synchronization|
    US7656985B1|2002-02-19|2010-02-02|Nortel Networks Limited|Timestamp-based all digital phase locked loop for clock synchronization over packet networks|
    US7031859B2|2002-03-11|2006-04-18|Piesinger Gregory H|Apparatus and method for identifying cable phase in a three-phase power distribution network|
    US6662124B2|2002-04-17|2003-12-09|Schweitzer Engineering Laboratories, Inc.|Protective relay with synchronized phasor measurement capability for use in electric power systems|
    US7471241B1|2005-07-25|2008-12-30|Chun Yang|Global navigation satellite system receivers based on satellite signal channel impulse response|
    FR2889331B1|2005-07-28|2008-02-01|Sercel Sa|APPARATUS AND METHOD FOR DERIVING COMPENSATION OF LOCAL CLOCK USED AS SAMPLING FREQUENCY|
    US7630863B2|2006-09-19|2009-12-08|Schweitzer Engineering Laboratories, Inc.|Apparatus, method, and system for wide-area protection and control using power system data having a time component associated therewith|
    EP1912108A1|2006-10-12|2008-04-16|Rohde & Schwarz GmbH & Co. KG|Device for providing a plurality of clock signals|
    BRPI0700841A|2007-03-20|2008-11-04|Santos Eduardo Pedrosa|real-time monitoring system of capacitive bushings of power transformers, power reactors, current transformers, circuit breakers and the like|
    WO2009055931A1|2007-11-02|2009-05-07|Novatel Inc.|System and method for distributing time and frequency over a network|
    CN101231337B|2008-02-15|2010-07-28|哈尔滨工程大学|High-precision time synchronizing apparatus|
    US8041981B2|2008-09-25|2011-10-18|Apple Inc.|Synchronizing timing domains based on state variables|
    CN101420747B|2008-11-12|2011-04-06|华为技术有限公司|Synchronization method, base station, network server and communication system|
    CN101557258B|2009-02-27|2014-07-02|工业和信息化部通信计量中心|Method and system for using synchronous digital hierarchy to realize high-accuracy time synchronization|EP2837553B1|2012-04-09|2022-01-19|CyTroniQ Ltd.|System for static and dynamic positioning or controlling motion of marine structure|
    EP2926604B1|2012-11-30|2019-08-07|Valmet Automation Oy|Multi-channel sensor measurement method and system|
    CN103051316B|2012-12-12|2015-03-18|中北大学|Synchronous triggering system in distributed type measurement|
    DE102013202320A1|2013-02-13|2014-08-28|Robert Bosch Gmbh|Method for detecting measured value, involves detecting measurement value to sampling point, through which time base of sensor clock generator is predetermined, so that unit is adjusted to time base of clock generator|
    US10372842B2|2013-03-14|2019-08-06|Xerox Corporation|Method and device for calibrating and updating a power model|
    TWI463145B|2013-04-30|2014-12-01|Univ Ishou|Phase synchronization measurement system|
    CN103308821B|2013-05-13|2015-07-01|沈阳农业大学|Intermittent arcing ground-fault line selection method based on improved phase locked loop|
    CN103491624B|2013-09-09|2016-05-25|电子科技大学|The method for synchronizing time of event time label in sensor network|
    US9277515B2|2013-11-04|2016-03-01|Samsung Electronics Co., Ltd|Precise time tagging of events over an imprecise link|
    JP6403957B2|2014-02-27|2018-10-10|横河電機株式会社|MEASUREMENT SYSTEM, MEASUREMENT MANAGEMENT DEVICE, MEASUREMENT DEVICE, AND MEASUREMENT METHOD|
    US20160154041A1|2014-04-18|2016-06-02|North China Electric Power University|A Synchrophasor Measurement Method Applying to P Class Phasor Measurement Unit |
    CN104459318B|2014-11-13|2017-06-16|广东电网有限责任公司电力科学研究院|The measuring method and system of electric harmonic|
    JP6285566B2|2014-11-14|2018-02-28|シャープ株式会社|Signal processing apparatus, signal processing method, and computer program|
    CN104483563B|2014-11-27|2017-03-08|广东电网有限责任公司电力科学研究院|The synchronous sampling method of electric power signal and system|
    CN104483545B|2014-11-27|2017-05-31|广东电网有限责任公司电力科学研究院|The harmonic measuring method and system of power system|
    EP3070483B1|2015-03-17|2017-09-13|ABB Schweiz AG|A method for monitoring transformer bushings, and a system therefor|
    CN105301437B|2015-09-18|2018-03-20|江苏省电力公司常州供电公司|Transformer station's transformer secondary polarity detection method based on GPS sync identifications|
    EP3153872A1|2015-10-06|2017-04-12|ABB Schweiz AG|Method and system for determining phasor components of a periodic waveform|
    CN107543671B|2016-07-15|2019-02-19|北京卫星环境工程研究所|More and exciting movement controller and digit phase isochronous controller matching process|
    US10859471B2|2016-10-03|2020-12-08|Epro Gmbh|Synchronization of machine vibration data after collection|
    CN106500942A|2016-11-25|2017-03-15|北京强度环境研究所|A kind of upright state simple structure modal test system|
    CN107290724B|2017-06-02|2020-05-22|北京理工大学|High-dynamic signal parameter estimation method based on delay correlation function|
    US10971166B2|2017-11-02|2021-04-06|Bose Corporation|Low latency audio distribution|
    US10320407B1|2018-02-06|2019-06-11|Analog Devices Global Unlimited Company|Low power synchronization of multiple analog to digital converters|
    CN108847921B|2018-07-06|2020-09-29|哈尔滨工业大学（深圳）|Distributed vibration synchronous continuous monitoring system|
    EP3650872B1|2018-11-08|2021-06-09|ABB Power Grids Switzerland AG|Relative bushing parameter method to avoid temperature influence in transformer absolute bushing parameter monitoring|
    CN110703246B|2019-10-24|2020-11-27|中国科学院电子学研究所|Signal synchronization device and system|
    CN110895321B|2019-12-06|2021-12-10|南京南瑞继保电气有限公司|Secondary equipment time mark alignment method based on recording file reference channel|
    CN111608645B|2020-05-28|2021-10-08|中国矿业大学|Directional ground penetrating radar device for drilling|
    法律状态:
    2019-01-08| B06F| Objections, documents and/or translations needed after an examination request according [chapter 6.6 patent gazette]|
    2019-09-10| B06U| Preliminary requirement: requests with searches performed by other patent offices: procedure suspended [chapter 6.21 patent gazette]|
    2020-01-21| B09A| Decision: intention to grant [chapter 9.1 patent gazette]|
    2020-03-10| B16A| Patent or certificate of addition of invention granted [chapter 16.1 patent gazette]|Free format text: PRAZO DE VALIDADE: 20 (VINTE) ANOS CONTADOS A PARTIR DE 16/03/2011, OBSERVADAS AS CONDICOES LEGAIS. |
    优先权:
    申请号 | 申请日 | 专利标题
    CA2699596A|CA2699596A1|2010-03-24|2010-03-24|System and method of phase synchronization of signals produced by respective units of measure|
    CA2699596|2010-03-24|
    PCT/CA2011/050143|WO2011116479A1|2010-03-24|2011-03-16|Method and system for the time synchronization of the phase of signals from respective measurement devices|
    [返回顶部]