• 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:
    There is provided a system (1) and method for creating and deploying one or more inference models (19) for use in telecontrol of the status of a first park of a first equipment (100). The system (1) includes pattern configuration data (111) for later use in a pattern creation application (13) for constructing one or more desired inference patterns (19) for the first equipment. The model configuration data (111) is adapted to the desired first device (100) and inference model (s) (19), and is provided in a user-readable and easily modifiable format (U) of the system (1). The model configuration data (111) is separated from the underlying processing algorithms that are used by the model-creating application (13) when constructing the desired inference model (s) (19) during a system operation learning mode (1). The system (1) can also be used in execution mode to derive a state of one or more of the first equipment (100) from operating data received from the given equipment (s) ( s).
    公开号:FR3025337A1
    申请号:FR1557885
    申请日:2015-08-24
    公开日:2016-03-04
    发明作者:Donna Louise Green；Brian David Larder；Peter Robin Knight；Olivier Thuong
    申请人:GE Aviation Systems Ltd；
    IPC主号:
    专利说明:

    [0001] The present invention relates to systems for creating and deploying inference models for use in telecontrol of the state of parks of different equipment. Without seeking to limit the applicability of the invention, the invention applies to the telecontrol of the state of aircraft engines or their organs, other systems or components of an aircraft or aircraft. other important forms of equipment. In many areas of industry and transportation, equipment condition monitoring is undertaken to allow an operator to derive equipment health status at the same time and / or predict his future state of health. Such systems are known in the aeronautics industry, manufacturers of aircraft engines controlling data received from the engines of one or more aircraft (s) in flight. The check may be initiated first to detect possible abnormal behavior of equipment (eg, an extremely high temperature or pressure reading in a stage of a high pressure turbine of a gas turbine engine and then to establish whether or not this abnormal behavior indicates an existing anomaly and, if so, to assess the existing anomaly. State control systems according to the prior art employ statistical models which, upon receipt of operating data from a running equipment, are able to derive received data whether or not they indicate an anomaly. in the equipment.
    [0002] However, the statistical models underlying these state telecontrol systems according to the prior art require extensive rewrite of the underlying source code for creating the statistical models to account for the different characteristics of different equipment. "Different" equipment means equipment that is of the same category but differs in one or more aspect (s) from its construction. This rewriting of the source code requires specialist skills and should normally be undertaken only by experienced computer scientists. This rewrite may be necessary even if the different equipment is in a common category (for example, a category of "gas turbine engines"), but differ from one another by one or more aspect (s) of their construction. . For example, statistical models used for state control of a fleet of a first aircraft gas turbine engine model will differ from models used for state control of a fleet of a second model aircraft engine gas turbine, even if there are common points between some of the parts constituting the two models of engines. This is true even if both engine models come from the same engine manufacturer. Differences in construction between the two engine models may very well result in different operating limits and different reaction behaviors for common parameters in corresponding members of the two engine models. Therefore, state-of-the-art telecontrol systems have required a great deal of time and effort from experienced computer scientists for absolutely every engine model to ensure that the developed statistical models can detect and evaluate accurately a behavior anomaly of a given engine model. This problem is equally applicable to state telecontrol systems used to detect and evaluate a behavior anomaly in other categories of equipment. Therefore, there is a need to find a more efficient way of creating an inference model for state-of-the-art equipment remote control, which is adaptable to reduce the time and effort required to create remote control systems. for equipment that are in the same category but differ in their creation.
    [0003] It is therefore proposed a system for creating and deploying one or more model (s) of inference for the telecontrol of the state of a first park of a first major equipment, the system comprising: a) a configuration file and / or a compiled configuration database, the compiled configuration file and / or configuration database being editable and user-readable and including template configuration data for later use by a user application. model creation to construct one or more desired inference model (s) for the first equipment, the model configuration data being adapted to the first equipment and the desired inference model (s) ( s); b) a template creation application, the template creation application being adapted to: receive pattern configuration data from the compiled configuration file and / or configuration database; and operate in learning mode and in execution mode, according to the received pattern configuration data; 3025337 4 for the learning mode, the template creation application being adapted to: receive operation history data corresponding to the plurality of operating states of the first equipment; soliciting one or more processing algorithms, the solicited processing algorithm (s) being determined by the model configuration data; and construct - using the received model configuration data, received operating history data and the requested processing algorithm (s) - the desired inference model (s) (s) so that the inference model (s) can (s) be used to derive a state of one or more of the first equipment (s) from the first fleet based on data from 15 received from the first equipment (s) of the first fleet during operation of the first equipment (s); during the execution mode, the template creation application being adapted to: receive operating data from one or more of the first equipment (s) of the first fleet during the operation of the first (s) ) equipment (s); and using the or at least one of the inference models constructed to process the received operating data, thereby deriving received operating data to a state of the first equipment (s). A second aspect of the invention proposes a method of constructing one or more inference model (s) intended to be used for the telecontrol of state of a first park of a first major equipment, the method comprising: 3025337 5 a) performing a compiled configuration file and / or a configuration database, the compiled configuration file and / or the configuration database being editable and readable (s) ) by a user and including template configuration data to be subsequently used by a template creation application to construct one or more desired inference templates for the first device, the template configuration data. being adapted to the first device and the desired model (s) of inference (s); B) implementing the template creation application; c) the template creation application receiving the template configuration data from the compiled configuration file and / or the configuration database; d) operating the model-learning application in the learning mode according to the received model configuration data, the model-creating application in learning mode: receiving operating history data corresponding to a plurality of operating states of the first equipment; requesting one or more processing algorithms, the requested processing algorithm (s) being determined by the model configuration data; and constructing - using the received pattern configuration data, received operating history data and the requested processing algorithm (s) - the desired inference model (s) (s), the inference models can thereby be used to derive a state from one or more of the first equipment (s) of the first fleet based on operational data received from the first one (s). (s) equipment (s) of the first fleet during operation of the first equipment (s). According to a third aspect of the invention, the method comprises successive steps of: operating the pattern creation application in run mode according to the pattern configuration data received by the pattern creation application, model-creation application, in execution mode: receiving operating data from one or more of the first equipment (s) of the first fleet during operation of the first equipment (s) ; and using the constructed inference model or at least one of the inference models constructed to process the received operating data to thereby derive, based on the received operating data, a status of the first one (s) ) equipment (s). According to a fourth aspect, the method is further adapted to construct one or more inference models for use in telecontrol of status of a second fleet of a second important equipment, the first and second second equipment belonging to the same category but differing in construction; knowing that: step (a) is performed for the second equipment to realize a corresponding second equipment configuration file and / or a second compiled configuration corresponding database containing model configuration data adapted to the second equipment and one or more desired inference model (s) for the second equipment; step (c) is performed for the second equipment, the template creation application receiving template configuration data from the corresponding second equipment configuration file and / or the corresponding second database. configuration compiled; and knowing that step (d) is performed for the second equipment to thereby construct the desired inference model (s) for the second equipment, the model (s) of desired inference (s) that can be used to derive a state of one or more of the second equipment (s) of the second fleet from operating data received from the second equipment (s) of the second park during operation of the second equipment (s). According to another aspect, the method according to the fourth aspect is adapted to comprise successive steps of: operating the pattern creation application in the execution mode in accordance with the adapted configuration data of 15 second equipment models received by the application of model creation, the model-creation application, in execution mode: receiving operating data from one or more of the second equipment (s) of the second fleet during operation of the second (s) equipment (s); and using the constructed inference model or at least one of the inference models constructed from the second equipment to process the operating data of the second equipment (s) received to thereby derive operating data. received a condition of the second equipment (s).
    [0004] The equipment is not limited to any field of technology and may include (without limitation) mechanical, electrical, electromechanical, software / hardware systems. By way of example only and without seeking to limit the scope of the invention, the equipment may be an aircraft engine, an aircraft flight control system or a wind turbine. These examples of equipment categories are ordinarily intended to operate in a semi-autonomous or fully autonomous manner without the presence of maintenance personnel. State telecontrol systems are useful in helping to identify or predict the occurrence of an abnormality in such self-powered equipment. A non-limiting example of first and second equipment sharing the same category consists of a first model of an aircraft gas turbine engine and a second model of an aircraft gas turbine engine. While in the same category of aircraft gas turbine engines, the first and second aircraft gas turbine engine models (i.e. the first and the second equipment) differ from each other in the same category of aircraft gas turbine engines. one of the other by their construction and therefore by their reaction behavior. Therefore, the inference model (s) necessary to derive a state of a first given equipment (e.g., the first engine model) from operating data received from the first given equipment will / will necessarily be different from the model (s) necessary to derive a state of a second given equipment (for example, the second engine model) from operating data received from the second given equipment. An inference model constructed by the system or method according to the invention for the first equipment must be usable to deduce the state of any of the first equipment of the first equipment from operating data received from the first equipment. equipment during operation. The learning mode is a mode of operation of the model creation application in which the inference model (s) is / are constructed (s). Execution mode is an operating mode of the model-creation application in which the inference model (s) constructed in a prior learning mode for the first equipment is employed in analyzing operating data received from one or more of the first equipment (s) of the first fleet during operation of the first equipment (s). The system and method of the invention are equally applicable to use for the second fleet of second devices - both in the learning mode and in the execution mode. Therefore, when the system and method will be described later with reference to the first equipment, unless otherwise stated, the system and method should also be considered suitable for the second equipment. The model configuration data for the second equipment will necessarily differ from the first equipment.
    [0005] The resulting inferred state of the execution mode is preferably: i) an indicator of the presence of an operating anomaly of the first equipment (s) of the first park; and / or ii) a classification of the anomaly indicated in one or more probable anomaly state (s) of the first equipment (s) of the first fleet. Depending on the complexity of the inference model (s) constructed, the model (s) may even be able to estimate the time remaining before a failure for one or more of the first equipment (s). ) (or an organ of it / these). The desired user-derived state resulting from the execution mode is likely to affect the nature of the inference model required by and constructed in the learning mode. For example, an inference model that is suitable for producing an indicator of presence of an anomaly (as in (i) above) is likely to have less complexity than and a different functionality. those of an appropriate inference model for assessing the indicated anomaly (as in (ii) above). The accuracy and functionality of a given inference model for deriving the state of a given device from the first equipment based on its run mode operation data is dependent on the quality and diversity of the data of the device. history of operation provided to the template creation application in the learning mode. The operating history data should preferably correspond to a plurality of normal operating states and a plurality of possible fault states for the first equipment. The operating history data is preferably derived from the entire first fleet of first equipment instead of coming from only one of the first equipment. As a result, this must increase the ability of the inference model (s) constructed (to be deduced (in run mode) if, yes or no, operating data received from the given one, first equipment and during the operation of a given one, first equipment, indicate a state of abnormality and / or to evaluate the state of abnormality. Returning to the example of the case where the first equipment is a first model of an aircraft gas turbine engine, the normal operating states of the operating history data may include the conditions of maximum thrust loads encountered during operation. take-off of an aircraft equipped with such an engine model, as well as cruising thrust conditions at a number of different altitudes of the aircraft, as well as thrust conditions during landing; the data may also be provided from a variety of different climatic conditions in which the given engine model has been operated. Preferably, the same diversity must also be present in the part of the operating history data representative of one or more condition (s) of anomaly (s). In addition to being based on in-service data, the historical data may include experimental data resulting from engine model testing on a test setup to simulate different normal regimes and anomaly conditions. Non-limiting examples of anomalous conditions include turbine blade failure, ingestion of one or more birds by the engine and failure of one or more of the injector (s) present (s). ) in the combustion chamber of the engine. Obviously, the nature of the operating history data will change depending on the particular type of equipment considered.
    [0006] As explained in later paragraphs, the system and method according to the invention can operate in the presence of the configuration file and in the absence of a compiled configuration database, or in the presence of the database of configuration compiled and in the absence of a configuration file. According to yet another possibility, a preferred example of the invention uses both the configuration file and the compiled configuration database. The use of the term "the configuration file and / or the configuration database compiled" covers these various different scenarios.
    [0007] Preferably, the template creation application is a computer executable file. The template creation application undertakes the construction of the inference model (s) in the learning mode. The template creation application constructs the inference template (s) based on the information contained in the model configuration data received from the configuration file and / or the configuration database compiled. , this information having been adapted to the first equipment and the desired model (s) of inference. As set forth in the appended claims, in the execution mode, the template creation application also employs at least one inference model constructed to derive a state from one or more of the first equipment (s). after the operating data received from the first equipment (s) during its operation.
    [0008] Essentially, the system and method according to the invention segment i) the high level information which defines various characteristics associated with the first desired equipment and the desired model (s) of inference (s) according to ii) the complex processing algorithms and complex software coding necessary to construct the inference model (s). The high level information of i) is contained in the configuration data of the configuration file and / or the compiled configuration database that can be modified and readable. The complex processing algorithms and the software coding in support of ii) are contained in, or communicate with, the template creation application. This segmentation benefits from the fact that, for different equipment belonging to the same category (for example, the first and second models of aircraft gas turbine engines mentioned above), the construction of 25 corresponding inference models for Different equipment is likely to employ common tooling of underlying processing algorithms and supporting software codes. The use of the computable configuration file and / or the editable and readable configuration database provides a location for defining specific data of the structure and / or performance characteristics for the first one. equipment (or, alternatively, the second equipment), as well as to indicate a required functionality of the desired inference model (s) for the first (or second) important equipment, this is ie "model configuration data". In the context of the above non-limiting example of first and second equipment constituting first and second respective models of aircraft gas turbine engines, the model configuration data adapted to the first Aircraft gas turbine engine model will be different from the model configuration data adapted for the second model of aircraft gas turbine engine. Preferably, the system (and the corresponding method) comprises the configuration file, which configuration file is directly modifiable and user-readable and containing the model configuration data. The configuration file is advantageously made in the form of a spreadsheet or an extensible markup language (XML) format file. If the file is in the form of a spreadsheet, the spreadsheet may comprise a single sheet or multiple spreadsheets related to each other. Whatever the format of the specific file of the configuration file, the configuration file is directly modifiable and readable by a user. One advantage of this configuration file is that it creates a relatively simple interface for an inexperienced user to provide the information necessary for the template application to construct the inference model (s). If the configuration file contains the model configuration data, the system (and the corresponding method) preferably also comprises the compiled configuration database, the compiled configuration database being designed to receive the template configuration data and store the pattern configuration data thus received in an object code format readable by the template creation application, the template creation application being adapted to receive the template data configuration of compiled models by accessing the compiled configuration database. Advantageously, the compiled configuration data base comprises or is coupled to a compiler for compiling data received from the directly modifiable and readable configuration file. By storing the model configuration data in a compiled object code format, the compiled configuration database increases the speed with which the template creation application can access and understand template configuration data. This offers advantages if the template creation application requires multiple accesses (for example, during the learning mode) to access the same configuration data of 20 templates, storing template configuration data in a code format compiled object making it easier to reduce the time required for the template creation application to access the template configuration data in comparison with the case where the template configuration data is stored in an uncompiled format. The template configuration data of the configuration file advantageously includes an address for the location of the compiled configuration database. Alternatively, the system (and the corresponding method) can avoid the use of the configuration database by presenting a compiler on a communication circuit between the configuration file and the pattern creation application. , the template creation application being adapted to receive model configuration data directly from the configuration file via the non-staging compiler in a compiled format, the compiler for converting the model configuration data into a format of compiled object code readable by the template authoring application. This variant avoids the use of a configuration database (or analog database) for storing the model configuration data in a compiled format, but may require repeated use of the compiler absolutely. whenever the template creation application wants to access the same template configuration data. According to another possible configuration, the system (and the corresponding method) can avoid the use of the configuration file and, instead, include the compiled configuration database, the compiled configuration database containing the configuration data. configuration of models in a compiled format, the system further comprising a user interface allowing a user to interact directly with the compiled configuration database to perform at least one of the following two operations: i) enter the data configuration of templates in the compiled configuration database; and ii) modify the model configuration data already stored in the compiled configuration database. This variant of the invention allows a user to interact directly with the configuration database to enter model configuration data into the configuration data base as well as to modify configuration data. models already contained in it. In a preferred example, the user interface comprises a compiled software application that communicates with the configuration database compiled to enable the user to enter and / or modify the model configuration data. Such a compiled software application would be a convenient interface to allow a user to enter and / or modify template configuration data in the configuration database.
    [0009] The segmentation described above and the use of an editable and readable configuration file and / or configuration database allow anyone other than a specialist software to provide the system with sufficient instructions to allow the template creation application to create one or more inference templates adapted to the first device (or also to the second device). The processing algorithms solicited and used by the template creation application will have varying complexity, depending on the complexity of the task they are performing. Some of the processing algorithms will have low level functionality; for example, to perform smoothing, trend clearing, and filtering operations on received operating history data for the first device during learning mode or on received operating data from one or more of the first ( s) 25 equipment (s) of the first park during the execution mode. Other processing algorithms will have a relatively higher level of functionality to construct the inference model (s). For example, these higher level processing algorithms can coordinate the operations of the low level functionality algorithms and use the results of these low level algorithms to thereby construct the inference model (s) during the second stage. learning mode. The processing algorithm or at least one of the processing algorithms is advantageously integrated into the template creation application. According to another possibility or in addition, the processing algorithm or at least one of the processing algorithms is present in one or more server (s) and / or database (s) with which / which communicates template creation application. In the latter case, the model configuration data preferably includes an address for the location of the server or at least one of the servers and / or the database / databases with which / which communicates the template creation application. In either case, the processing algorithms are separate and distinct from the model configuration data. Thus, while the configuration file and / or the compiled configuration database can / may contain instructions, inputs, or other data (i.e., "model configuration data") which will then be used by a given processing algorithm mentioned in the file and / or the database, the file and / or the database does not contain / itself contain all the processing algorithms necessary for the construction of the model (s) of inference. This facilitates the modification and understanding of the configuration file and / or configuration database compiled for a non-specialized user of the system, i.e. someone other than an experienced computer scientist. As noted above, model configuration data is adapted to the first device and the desired inference model (s). The model configuration data advantageously comprises one or more of the following data (s): an identifier of the first device and a classification in a category to which the first device belongs; an indicator of the necessary functionality of the desired inference model (s); an indicator of the required functionality of the desired inference model (s); Values assigned to one or more parameter (s) associated with the operation of the first equipment; an identity of the processing algorithm or at least one of the processing algorithms for subsequent use by the template creation application to construct the desired inference model (s); and one or more input parameter (s) associated with the processing algorithm or at least one of the processing algorithms. In the context of the non-limiting example of the first equipment consisting of an aircraft gas turbine engine of a given make and model, the identifier of the first equipment may be the make and model of the equipment. engine, the classification being that of gas turbine engines. However, the nature of the identifier and classification into a category will vary depending on the nature of the first piece of equipment. Non-limiting examples of indicators of the required functionality of the desired inference model (s) include the specification of a desired desired result of using an inference model in the execution mode, and the information that the desired inference model to be constructed in the learning mode is intended to indicate only the presence of a malfunction of a given piece of equipment and / or is intended to evaluate the class the anomaly revealed. Non-limiting examples of the values assigned to one or more parameters associated with the operation of the first equipment include values of maximum take-off thrust, cruise-level thrust and specific fuel consumption rates. for a first model of an aircraft gas turbine engine. However, the nature of these values and their associated parameters will vary according to the nature of the first equipment; for example, if the first equipment was a flight control system, these values would likely include voltage or current values. As non-limiting examples of the input parameter (s) associated with the processing algorithm or at least one of the processing algorithms, mention may be made of pressure, temperature or pressure variations. and temperature in space and / or time, encountered in a given element of a first aircraft gas turbine engine model. Again, however, the nature of the input parameters will vary depending on the category of the equipment. As noted above, if the first equipment was, for example, in the category of flight control systems, the input parameters would likely include various voltages or intensities in one or more portions of the flight control system. . In addition, the model configuration data advantageously further includes one or more of the following data (s): values assigned to one or more physical property (s) of the first equipment; Values defining an upper and / or lower limit for the input parameter or at least one of the input parameters; set values for the processing algorithm or at least one of the processing algorithms; A control command value defining whether or not the pattern creation application should operate in the learning mode or the execution mode; and an alert set value defining an action to be performed or a warning to be provided to a user of the system 10 based on a result of the inference tool (s) constructed during use in the run mode. A non-limitative example of values assigned to one or more physical property (s) of the first device comprises a mass or dimension (s) of one or more members of a first model of gas turbine engine of an aircraft. However, the physical properties used will vary depending on the nature of the first equipment. A non-limitative example of values defining an upper and / or lower limit for the input parameter and / or at least one of the input parameters comprises a maximum value of pressure or temperature. These limit values may include the maximum temperature and / or pressure (s) that can be supported by one or more members of a first model of a gas turbine engine. aircraft; For example, a maximum allowable temperature in the combustion chamber or a maximum permissible temperature for one or more of the blades of a turbine in the turbine section of the engine. In addition or alternatively, these values may represent minima or maxima of operating parameters which, if present in the received operating history data (for the learning mode) or in operation received during the operation of the first equipment (for execution mode), would be indisputably outliers and, therefore, rather indicate the presence of an anomaly in a sensor provided to provide the data. Preferably, in this case, the template creation application would not take into account data received during the learning or execution modes, the values of which would exceed the defined minima or maxima, which would prevent the application of creation of model is biased by sensor read errors during i) construction of inference models (during learning mode) or ii) use of inference models constructed (during run mode). Non-limiting examples of alert set point values defining an effector action or warning to be communicated to a user of the system based on a result of the inference tool (s) constructed during the use in the execution mode include producing an audible or visual warning signal for the system user to notify him in good time of an abnormality detected by the system and / or to provide him with identifying information. the cause of the anomaly. The anomaly detected may, for example, be an excessive variability of a given range of values for the mass flow rate in the turbine section of a first aircraft gas turbine engine model, the identified cause of the anomaly being a turbine rotor blade failure or a fault in the fuel flow to the engine combustion chamber. The precise nature of the identifiers, indicators, parameters, values and setpoints above will depend on the category of the first equipment. If the first equipment, for example, fell within the category of aircraft flight control systems, the values and setpoints could very well relate to electrical input / output signals provided by the flight control system. . As indicated above in the description, the system and method of the invention can be applied to any equipment in which there is a need or desire to remotely control the state of the equipment. Such equipment may include mechanical, electrical, electromechanical, and hardware / software systems (or a combination thereof). The invention will be better understood on studying the detailed description of an embodiment taken by way of nonlimiting example and illustrated by the appended drawings in which: FIG. 1 represents a first example of a system according to the invention; invention, associated with a first fleet of a first model aircraft engine gas turbine engine; FIG. 2 represents a second example of a system according to the invention, associated with the first park of the first aircraft gas turbine engine model; FIG. 3 illustrates an example of the content of a configuration file that can be used with the system examples of FIGS. 1 and 2; and FIG. 4 shows the data flows between different parts of the system of FIG. 1 during its operation for first and second parks of corresponding first and second equipment. It should be noted that the figures are intended to illustrate non-limiting examples of the invention. Figure 1 shows a first example of a system 1, the elements of which are shown surrounded by a discontinuous line boundary. Figure 1 shows the system 1 used with a first fleet of first equipment 100. The first fleet of first equipment 100 illustrated consists of a plurality of engines of a first model of aircraft gas turbine engine 5 used in a certain number of aircraft. However, as indicated in the above general description, the equipment is not limited to any technological sector. In other non-illustrated embodiments, the equipment may be a flight control system of an aircraft, a wind turbine, or any other form. The system and method of the invention are applicable to any equipment for which it may be necessary or desirable to remotely control the condition of the equipment. The system 1 of Figure 1 includes a configuration file 11, a compiled configuration database 12, and a model 13 authoring application. The configuration file 11 is contained in a memory module 141 of a computer 14. The computer 14 also includes a processor 142 coupled to the memory module 141, the processor being adapted to execute the template creation application 13. The template creation application 13 is an executable computer application capable of executing a template. or more tasks when executed by the processor 142. The computer 14 is coupled to a display screen 143 visible to a user U of the system 1 to allow the user to read the contents of the configuration file 11 The computer 14 is also coupled to an input device 144 such as a keyboard to allow the user U to modify the configuration file 11 and thus act on the function of the Modeling Application 13.
    [0010] In the example illustrated in FIG. 1, the computer communicates with a first server 15, a second server 16 and a third server 17. The communication can be provided by a wired and / or radio connection between the computer 14 and the computer 14. and each of the servers 15, 16, 17, as indicated in FIG. 1 by the continuous lines extending between the servers 15, 16, 17 and the computer 14. In the illustrated example, the first server 15 contains the Compiled configuration database 12. The second server 16 contains a library of processing algorithms accessible and used by the model-creating application 13. In another possible example not shown in the figures, the processing algorithms are contained in the computer 14, for example by being stored locally in the memory module 141 or by being coded for integration directly into the model-making application 13. The third Server 17 contains an operating history data library corresponding to normal operation and abnormalities for the first model aircraft engine 100 gas turbine engine. The origin of these operating history data is presented in the above general description. It is preferred that these operating history data come from multiple engines of the first aircraft gas turbine engine model 100 that collectively constitute the first fleet 100. The normal operating states covered by operation may include the operation of the first aircraft gas turbine engine model at the maximum rated thrust (as might be encountered during take-off of an aircraft corresponding to the first gas turbine engine model 30). aircraft), and cruise operation at a number of different altitudes and for a number of different meteorological conditions. The anomalous conditions covered by the operating history data may include a rupture of a turbine blade in the first model of an aircraft gas turbine engine, the ingestion of one or more birds. by the engine and the failure of one or more of the injectors in the combustion chamber of the engine. In other possible embodiments not illustrated in the figures, the contents of the three different servers 15, 16, 17 can be assembled in a smaller number of servers; for example, the compiled configuration database 12, the processing algorithms and the operating history data can be placed in a common server communicating with the computer 14.
    [0011] For the system 1 of Figure 1, the configuration file 11 is in a directly modifiable and user-readable format U. Non-limiting examples of the format of the configuration file 11 include a spreadsheet file comprising one or more sheets each other, or an extensible markup language (XML) format file. For example, the configuration file 11 may comprise one or more drop-down menu (s) allowing the user U to choose one or more option (s) corresponding to characteristics of the first equipment 100 and the model 25. In Figure 1 (as well as in FIG. 3), the configuration file 11 contains model configuration data 111. The model configuration data 111 is configured for a user-defined inference. Subsequent use by the pattern creation application 13 to allow the model-creating application to construct a 26 or more inference model (s) for the first model of the gas turbine engine 100 aircraft. The model configuration data 111 is adapted to the first aircraft gas turbine engine model 100 and any model of inference desired by the user U. An exemplary form of these model configuration data 111 is explained in more detail in later paragraphs, with reference to FIG. 3 The compiled configuration database 12 shown in FIG. 1 is adapted to receive from the configuration file 11 the model configuration data 111 and to store the data. configuration of models thus received in a compiled object code format readable by the template creation application 13. In the system example 1 illustrated in FIG. 1, the compiled configuration database 12 is contained in FIG. first server 15. The model creation application 13 is designed to access the compiled configuration database 12 in the first server 15 for the first server 15. r thus access the compiled model configuration data 111 stored in the configuration database 12. By storing the model configuration data 111 in a compiled object code format, the compiled configuration database 12 provides access quickly from the template creation application 13 to the template configuration data 111 in a format easily understandable by the template creation application. Although not shown in Figure 1, the compiled configuration database 12 may include or be coupled to a compiler for compiling model configuration data 111 received from the directly modifiable and readable configuration file 11.
    [0012] In the second exemplary system 2 shown in FIG. 2, a compiler C is present between the memory module 141 and the model creation application 13, no compiled configuration database 12 or any equivalent means 5. being provided for storing the model configuration data 111 in a compiled object code format. The model configuration data 111 is communicated to the model creation application 13 directly by the compiler C, without intermediate storage in a compiled object code format. In this second exemplary system 2, whenever the template creation application 13 has to access the template configuration data 111, the compiler C steps in to compile the received model configuration data 111 in an object code format. Therefore, the compiler C is involved in compiling the template configuration data 111 as required by the template creation application 13. In this way, the system 2 of FIG. 2 avoids the recourse to a configuration database 12 (or equivalent) for storing the model configuration data 111 in a compiled format. In another possible example of a system not shown in the drawings, the use of the configuration file 11 is dispensed with. In this alternative system, the compiled configuration database 12 contains the model configuration data 11. a compiled format, the system comprising a user interface usable to allow the user U to interact directly with the configuration database compiled to perform at least one of the following operations: i) enter the configuration data models 111 in the compiled configuration database 12; and ii) modify the configuration data of templates already stored in the compiled configuration database.
    [0013] This variant of the invention allows the user to interact directly with the configuration database 12 to directly enter the model configuration data 111 into the configuration database as well as modify the configuration data of the configuration database. models 111 already contained in it. In a preferred example, the user interface would include a compiled software application communicating with the compiled configuration database 12 to allow the user to enter and / or modify the model configuration data. Such a compiled software application provides a convenient interface for allowing a user to enter and / or modify template configuration data in the configuration database. The user interface may also include a display screen and keyboard input device (similar or the same as the display screen 143 and the keyboard 144) to facilitate the interaction between the user U and the keyboard. user interface. For all the systems described and / or illustrated, the model creation application 13 can be used in two distinct modes: a learning mode and an execution mode as explained in the above general description. The operation of the pattern creation application 13 in both learning and execution modes is determined by the model configuration data 111, the model configuration data being adapted to the first gas turbine engine model 100 and the (x) desired inference model (s) by the user U. In the learning mode, the model creation application 13 is used to construct one or more inference model (s) 19 ( see Figure 4) for the first aircraft gas turbine engine model 100. In the learning mode, the model creation application 13 of the illustrated example is designed to: - access the third server 17 to acquire the operating history data representative of the first model of the gas turbine engine 100 aircraft; accessing the second server 16 to request one or more of the processing algorithms contained in the second server, based on the model configuration data 111; and - constructing - using i) compiled model configuration data 111, ii) acquired operating history data, and iii) the requested processing algorithm (s) - one or more inference model (s) 19 for the first aircraft gas turbine engine model 100. As indicated in the foregoing general description, the inference models 19 are constructed so as to be able, during a subsequent execution mode of operation of the model-creation application 13, to derive from operating data received from the given engine, a state of any one of the engines of the first fleet 100 of the first aircraft gas turbine engine model. The inference models 19 are mathematical calculation models relating to the first fleet 100 of the first aircraft gas turbine engine model. By virtue of the fact that the inference models 19 (see FIG. 4) are constructed from operating history data representative of the first aircraft gas turbine engine model 100, these mathematical models are able to draw conclusions based on real-time operating data received from one or more of the engines of the first fleet 100. These inferences include an indication of whether or not an abnormality is present during operation. of a given engine of the first fleet 5 100, and / or the classification of the nature of the anomaly indicated. The operation of the template creation application 13 in the learning and execution modes depends on the content of the template configuration data 111 - this applies regardless of the format in which the template configuration data is stored; that is, irrespective of whether a configuration file 11 and / or a compiled configuration database 12 are used. The model configuration data 111 is adapted to the first model of a gas turbine engine. aircraft of the first fleet 100 and any model of inference 19 desired by the user U. The desired inference model 19 may be of a type which, when used in the execution mode, the user U an indicator as to the presence of an anomaly in the operation of any one of the engines of the first park 100. According to another possibility or in addition, the desired model of inference can be of a type who, s of use in execution mode, provides the user U with a classification of the anomaly indicated in one or more condition (s) of probable anomaly (s) for the given engine. Depending on the complexity of the desired inference model 19 constructed by the model-making application 13, the inference model provides an estimate of the time remaining before a failure for the given engine of the first park 100. As noted above, Figure 3 illustrates an exemplary arrangement of the pattern configuration data 111 of the configuration file 11. The model configuration data 111 comprises: an identifier 1111 of the first equipment item 100 and a classification in a category corresponding to the first equipment. In the illustrated example, the identifier comprises the make and model of the first aircraft gas turbine engine model 100 to be the subject of the inference model (s). Illustrated example, the classification category is that of aircraft gas turbine engines; an indicator 1112 of the required functionality of the desired inference model (s). In the illustrated example, this includes the specification of one or more result (s) required as a result of using an inference model 19 in the execution mode, and that, yes or no, the desired inference model is intended to allow the indication of the presence of a malfunction of an engine of the first fleet 100 and / or is intended to classify the indicated anomaly; values 1113 assigned to one or more parameter (s) associated with the operation of the first aircraft gas turbine engine model 100. In the illustrated example, these values 1113 include maximum take-off thrust values, thrust at cruising altitude, and specific fuel consumption rates for the first aircraft gas turbine engine model; an identity 1114 of the processing algorithm or at least one of the processing algorithms for later use by the model processing application 13 to construct the inference model (s) 19 ; one or more input parameter (s) 1115 associated with the processing algorithm or at least one of the processing algorithms. In the illustrated example, the input parameter (s) 1115 includes / understands the pressure, the temperature and the variations of pressure and temperature in space and / or time encountered in a given organ. the first model of an aircraft gas turbine engine 100; values 1116 assigned to one or more physical properties of the first aircraft gas turbine engine model. In the illustrated example, these values 1116 comprise a mass or a dimension of one or more members (s) of the first aircraft gas turbine engine model 100; values 1117 defining an upper and / or lower limit for the input parameter and / or at least one of the input parameter (s). In the example illustrated, these values 1117 comprise maximum pressure and / or temperature values, in particular the maximum temperature and / or pressure (s) permissible (s) that can / can withstand one or more members (15) of the first model of 100 gas turbine engine aircraft. In addition or alternatively, these values may represent pressure or temperature minima or maxima (or any other input parameter) that, if present in the operating history data received for the mode. 20 or if they are present in the operating data received during operation of the first aircraft gas turbine engine model in the run mode, would be indisputably outliers and, therefore, would rather indicate the existence of a anomaly in a sensor responsible for providing the data; setpoints 1118 for the processing algorithm or at least one of the processing algorithm (s); a command set value 1119 defining whether or not the pattern creation application 13 should operate in the learning mode or in the execution mode; and 3025337 33 - an alarm set point 1120 defining a measurement to be taken or an alert to communicate to a user of the system 1, 2 according to a result of the inference model (s) constructed (s) during use in the run mode. The measurements defined by the alarm set point 1120 may include providing an audible warning signal or a visual warning signal to the user U to warn him in good time of an anomaly. detected by the system 1, 2 and / or to provide the user with an identity of the cause of the anomaly. Such a visual warning signal may be communicated to the user U by the display screen 143, the display screen also comprising an acoustic speaker (not shown in the figures) to allow communication of the signal from the display. sound warning to the user.
    [0014] The adaptability of the systems 1, 2 to the creation of inference models adapted to different equipment parks is explained with reference to FIG. 4. FIG. 4 is a simplified representation of the elements of the system 1 (some elements being absent ), but annotated to indicate the data flow in the system 1 and the nature of the results of the system during the operation learning mode of the pattern creation application 13. Figure 4 illustrates the presence of the first fleet 100 of first equipment which (as previously indicated) consists of a plurality of engines of a first aircraft gas turbine engine model used on a number of aircraft. However, Figure 4 also illustrates the presence of a second park 200 of second equipment. The second fleet of second equipment 200 illustrated consists of a plurality of engines of a second aircraft gas turbine engine model 30 used in a number of aircraft. Therefore, it can be seen that the first and second equipment 100, 200 belong to the same category of aircraft gas turbine engines. However, the first and second equipment 100, 200 differ from one another in one or more aspects of their construction (and hence in their performance) resulting from the fact that they are two different engine models. The configuration file 11 and the compiled configuration database 12 correspond to the first model 100 of an aircraft gas turbine engine. However, in Fig. 4 there is also shown a separate configuration file 21 and configuration database 22 which corresponds to the second aircraft gas turbine engine model 200. Although not shown in Figure 4, in one example, the configuration databases 12, 22 are both located in the first server 15. Alternatively, the databases 12, 22 may be located in respective different servers capable of communicating with the computer 14 and the model 13 creation application. The configuration data of models 111, 211 of the configuration files 11, 21 are adapted. to the respective first and second motor models 100, 200 and any desired inference model 19, 29 for the first and second respective models of gas turbine engine 100, 200 of aircraft. In the example illustrated in Figure 4, the second server 16 contains the processing algorithms (previously presented for system 1 of Figure 1). On the other hand, the third server 17 contains respective operating history data libraries corresponding to normal operating conditions and anomalies for the first engine model 100 to 3025337 aircraft gas turbine and the second engine model. 200 aircraft gas turbine. When used in the learning mode, the model-creating application 13 builds a first series 5 of inference models 19 corresponding to the first model of the aircraft gas turbine engine 100 and a second series of model aircraft. inference 29 corresponding to the second engine model 200 gas turbine aircraft. The model creation application 13 constructs the respective series of inference models 19, 29 for the first and second engine models 100, 200 respectively using i) the compiled model configuration data 111, 211 stored in the respective compiled configuration databases 12, 22, ii) the operating history data contained in the third server 17, corresponding to the first and second respective models 100, 200 of engines; and iii) the solicited processing algorithm (s) contained in the second server 16. In the example illustrated in FIG. 4, the configuration files 11, 21 and the configuration databases compiled 12, 22 for the first and second aircraft gas turbine engine models 100, 200 coexist with each other, the pattern creation application 13 successively or simultaneously accessing the respective configuration databases 12 , 22 when it serves to build the respective sets of inference patterns 19, 20. In another possible example, the configuration file 22 can be obtained by modifying the contents of the configuration file 11, only one of the configuration files. 11, 22 can exist at a given moment. During the execution mode, the pattern creation application 13 uses the respective series of inference patterns 19, 29, to derive the state of one or more of any of the first or second park engines. 100, 200. In the run mode, the pattern creation application 13 receives operating data 101, 201 (preferably in near real time) from one or more of the first and second respective fleet engines 100 , 200 (see Figure 4). In the illustrated example of FIG. 4, these operating data 101, 201 are transmitted by radio from the first and second motor parks 100, 200 to be received by a transceiver (not shown) coupled to the The model creation application 13 then uses the inference model series 19 which corresponds to the respective park 100, 200 from which the operating data 101, 201 are received in order to process the operating data. received in accordance with the respective pattern configuration data 111, 211 to thereby derive a state of one or more of the first or second motors 100, 200. As explained in the foregoing general description, the state deduced may be an indication of the presence of a malfunction of one, 20 given engines of the first or second park 100, 200. According to another possibility or in or be, the state deduced can be a classification of the anomaly detected and / or a time remaining predicted before the failure of one or more organ (s) of the given engine.
    权利要求:
    Claims (15)
    [0001]
    REVENDICATIONS1. A system (1) for creating and deploying one or more inference patterns (19) for use in telecontrolling the status of a first fleet of first equipment (100), the system (1) is characterized in that it comprises: a) a configuration file (11) and / or a compiled configuration database (12), the configuration file (11) and / or the configuration database compiled (12). ) being editable and user-readable (U) and containing model configuration data (111) for use by a model-creating application (13) to construct one or more desired inference models ( s) for the first equipment (100), the model configuration data (111) being adapted to the desired first equipment (100) and inference model (s) (19); b) a template creation application (13), the template application (13) being adapted to: receive template configuration data (111) from the configuration file (11) and / or the database compiled configuration (12); and operate in learning mode and in execution mode, according to the received pattern configuration data (111); for the learning mode, the pattern creation application (13) being adapted to: receive operation history data corresponding to a plurality of operating states of the first equipment (100); Requesting one or more processing algorithms, the solicited processing algorithm (s) being determined by the model configuration data (111); and construct - using the received model configuration data (111), received operating history data and the requested processing algorithm (s) - the model (s) of desired inference (19) so that the desired inference model (s) (19) can (nt) be used to infer a state of one or more of the first equipment (s) (s) (100) of the first park 10 based on operating data received from the first equipment (s) (100) of the first park during operation of the first equipment (s) ( 100); for the run mode, the model creation application (13) being adapted to: receive operating data from one or more of the first equipment (100) of the first fleet during operation of the first equipment (100); and using the constructed inference model (19) or at least one of the constructed inference model (s) (19) to process the received operating data to thereby derive a state of the first one (s). (s) equipment (100) based on received operating data.
    [0002]
    2. System (1) according to claim 1, the system (1) comprising the configuration file (11), the configuration file (11) being directly modifiable and readable by a user and containing the model configuration data ( 111).
    [0003]
    The system of claim 2, the system (1) further comprising the compiled configuration database (12), the compiled configuration database (12) being adapted to receive from the configuration file (11). the model configuration data (111) and store the model configuration data (111) thus received in a compiled object code format readable by the model creation application (13), the model creation application (13) being adapted to receive the compiled model configuration data (111) by accessing the compiled configuration database (12).
    [0004]
    The system (1) according to claim 2, the system further comprising a compiler (C) disposed on a communication circuit between the configuration file (11) and the pattern creation application (13), the a template creation application (13) adapted to receive template configuration data (111) directly from the configuration file (11) via the compiler (C) without intermediate storage in a compiled format, the compiler (C) serving converting the template configuration data (111) into a compiled object code format readable by the template creation application (13).
    [0005]
    The system (1) of claim 1, the system (1) comprising the compiled configuration database (12), the compiled configuration database (12) containing the model configuration data (111) under a compiled format, the system (1) further comprising a user interface usable by a user (U) for interacting directly with the compiled configuration database (12) to perform at least one of the following operations: model configuration data (111) in the compiled configuration database (12); and ii) modifying the pattern configuration data (111) already stored in the compiled configuration database (12). 3025337 40
    [0006]
    The system (1) according to any one of the preceding claims, wherein the processing algorithm or at least one of the processing algorithms (i) is / are integrated into the model creation application (13). ) and / or (ii) is / are included in one or more server (s) (15, 16, 17) and / or a database or at least one of the databases with which communicates the template creation application (13).
    [0007]
    The system (1) of claim 6, wherein the model configuration data (111) comprises one or more of the following data: an address for the location of the server or at least one of the server (s) (15, 16, 17) and / or a database or at least one of the databases with which the model creation application (13) communicates; an identifier (1111) of the first equipment (100) and a classification in a category to which the first equipment (100) falls; an indicator (1112) of the required functionality of the desired inference model (s) (19); assigned values (1113) to one or more parameters associated with the operation of the first equipment (100); an identity (1114) of a processing algorithm or at least one of the processing algorithms for subsequent use by the template creation application (13) to construct the template (s) of desired inference (19); one or more input parameters (1115) associated with the processing algorithm or at least one of the processing algorithms; Values (1116) assigned to one or more physical properties of the first equipment (100); values (1117) defining an upper and / or lower limit for the input parameter or at least one of the input parameters (1115); set values (1118) for the processing algorithm or at least one of the processing algorithms; a command set value (1119) defining whether or not the pattern creation application (13) is to operate in the learning mode or the execution mode; and an alert setpoint (1120) defining a measurement to be taken or a warning to be communicated to a user (U) of the system (1) according to a result obtained by the inference tool (s) built (s) during use in the run mode.
    [0008]
    A method for constructing one or more inference models for use in telecontrol of the status of a first fleet of first equipment (100), the method comprising: a configuration file (11) and / or a compiled configuration database (12), the configuration file (11) and / or the compiled configuration database (12) being modifiable (s) and user-readable (U) and containing template configuration data (111) to be subsequently used by a template-creating application (13) to construct one or more desired inference templates (s) (19) for the first device (100), the model configuration data (111) being adapted to the desired first device (100) and inference model (s) (19) ; b) implementing the template creation application (13); C) the pattern creation application (13) receiving the pattern configuration data (111) from the compiled configuration file (11) and / or the compiled configuration database (12); 5 d) operating the template creation application (13) in learning mode according to the received template configuration data (111), the template creation application (13), in the learning mode: receiving data operating history 10 corresponding to a plurality of operating states of the first equipment (100); requesting one or more processing algorithms, the solicited processing algorithm (s) being determined by the pattern configuration data (111); and constructing, using the received model configuration data (111), received operating history data and the requested processing algorithm (s) - the model (s) of desired inference (19), the inference models (19) can thus be used to derive a state from one or more of the first equipment (100) of the after operating data received from the first equipment (100) of the first park during operation of the first equipment (100).
    [0009]
    The method of claim 8, the method comprising performing the configuration file (11), the configuration file (11) being directly modifiable and user-readable and containing the model configuration data (111).
    [0010]
    The method of claim 9, the method further comprising performing the compiled database configuration (12), the compiled configuration database (12) being adapted to receive the model configuration data. (111) from the configuration file (11) and store the model configuration data (111) thus received in a compiled object code format readable by the model creation application (13), the creation application model (13) receiving the compiled model configuration data (111) by accessing the compiled configuration database (12).
    [0011]
    The method of claim 8, the method comprising performing the compiled configuration database (12), the compiled configuration database (12) containing the model configuration data (111) in a compiled format. the system (1) further comprising a user interface for a user to interact with the compiled configuration data base (12) to perform at least one of the following operations: i) entering template configuration data ( 111) in the compiled configuration database (12); and ii) modifying the template configuration data (111) already stored in the compiled configuration database (12).
    [0012]
    The method of any one of claims 8 to 11, wherein the processing algorithm or at least one of the processing algorithms (i) is / are integrated into the template creation application (13). ) and / or (ii) is / are included in one or more servers (15, 16, 17) and / or one or more database (s) with which and / or which / communicates the application of model creation (13). 3025337 44
    [0013]
    The method of claim 12, wherein the model configuration data (111) contains one or more of the following data: an address for the location of the server or one of the servers (15, 16, 17) and / or a database or databases with which and / or with which / communicates the template creation application (13); an identifier (1111) of the first equipment (100) and a classification in a category to which the first equipment 10 (100) falls; an indicator (1112) of the required functionality of the inference model (s) (19); values (1113) assigned to one or more parameter (s) associated with the operation of the first piece of equipment (100); An identity (1114) of the processing algorithm or at least one of the processing algorithms for subsequent use by the pattern creation application (13) when constructing the model (s) of desired inference (19); one or more input parameters (1115) associated with the processing algorithm or at least one of the processing algorithms; values (1116) assigned to one or more physical property (s) of the first piece of equipment (100); values (1117) defining an upper and / or lower limit for the input parameter or at least one of the input parameters (1115); set values (1118) for the processing algorithm or at least one of the processing algorithms; A command set value (1119) defining whether or not the pattern creation application (13) is to operate in the learning mode or in an execution mode; and an alert setpoint (1120) defining a measure to be taken or a warning to be communicated to a user (U) of the system (1) according to a result obtained by the tool (s) of inference built during use in run mode.
    [0014]
    A method according to any one of claims 8 to 13, the method comprising the successive steps of: operating the pattern creation application (13) in run mode according to the model configuration data (111) received by the pattern creation application (13), the pattern creation application (13), in execution mode: receiving operating data from one or more of the first piece of equipment (100); ) of the first park during operation of the first equipment (100); and using the inference model or at least one of the constructed inference models (19) to process the received operating data to thereby derive a state of the first equipment (s). (100) from the received operating data.
    [0015]
    A method according to any one of claims 8 to 14, the method being further adapted to construct one or more inference models (19) to be used in telecontrol of the status of a second park. a second piece of equipment (200), the first (s) and second (s) equipment (100, 200) belonging to the same category but differing in their construction, knowing that: step (a) is performed for the second equipment (200) 30 in order to perform, for the second equipment (200), a corresponding configuration file (21) and / or a compiled configuration database (22) containing model configuration data (211) adapted to the second equipment (200) and one or more desired inference model (s) (29) for the second equipment (200); step (c) is performed for the second device (200), the template creation application (13) receiving the template configuration data (211) from the configuration file (21) and / or the database corresponding compiled configuration data (22) corresponding to the second equipment; and knowing that step (d) is performed for the second equipment (200) to thereby construct the desired inference model (s) (29) for the second equipment (200), the model (s) (s) (29) can thereby be used to infer a state of one or more of the second equipment (200) of the second park from data of received from the second equipment (200) of the second park during operation of the second equipment (200).
    类似技术:
    公开号 | 公开日 | 专利标题
    FR3025337B1|2019-08-16|SYSTEM FOR CREATING AND DEPLOYING A MODEL OF INFERENCE
    EP2997343B1|2021-07-28|Method and system for the vibration analysis of an engine
    CA2943397C|2019-10-15|Method for assessing whether or not a measured value of a physical parameter of an aircraft engine is normal
    AU2017238862A1|2018-11-01|Computer systems and methods for creating asset-related tasks based on predictive models
    EP2756280B1|2018-07-11|System for monitoring a measurement chain of a turbojet engine
    FR2869698A1|2005-11-04|METHOD FOR MONITORING AND DIAGNOSING MACHINES
    CA2888716A1|2014-05-01|System for monitoring a set of components of a device
    EP2376989A1|2011-10-19|Identification of defects in an aircraft engine
    CA2914819C|2021-05-25|Methods of creating a database and of formulating a map of operating states of aircraft and a method of monitoring the operation of an associated aircraft
    EP3350660B1|2021-01-27|Decision aid system and method for the maintenance of a machine with learning of a decision model supervised by expert opinion
    CA2746537A1|2010-07-08|Standardization of data used for monitoring an aircraft engine
    CA2944120A1|2015-10-08|Method and device for monitoring a parameter of a rocket engine
    EP3163445B1|2021-02-24|Mechanism for analysing correlation during performance degradation of an application chain
    US10635095B2|2020-04-28|Computer system and method for creating a supervised failure model
    FR2990547A1|2013-11-15|CENTRALIZED PARAMETRABLE MAINTENANCE SYSTEM FOR AN AIRCRAFT
    CA2966306A1|2016-05-12|Tool for validating a system for monitoring an aircraft engine
    FR3025886B1|2019-07-19|SYSTEM AND METHOD FOR CONTROLLING PARTS
    FR2957170A1|2011-09-09|Equipment monitoring system designing tool for engine of aircraft, involves processing unit evaluating quantification of monitoring system based on quality value of result corresponding to output quality value associated with output module
    FR3012636A1|2015-05-01|METHOD FOR NON-REGRESSION OF A DESIGN TOOL OF AN AIRCRAFT ENGINE MONITORING SYSTEM
    FR3026785B1|2019-08-02|MONITORING A SET OF THE PROPULSIVE SYSTEM OF AN AIRCRAFT
    FR3035232A1|2016-10-21|SYSTEM FOR MONITORING THE HEALTH CONDITION OF AN ENGINE AND ASSOCIATED CONFIGURATION METHOD
    WO2020115440A1|2020-06-11|Computing environment system for monitoring aircraft engines
    WO2016087770A1|2016-06-09|Method for monitoring the operation of a turbomachine
    US11232371B2|2022-01-25|Computer system and method for detecting anomalies in multivariate data
    CA2813619A1|2013-10-24|Process and configuration device for an alert management system in an aircraft
    同族专利:
    公开号 | 公开日
    GB2529637B|2017-07-05|
    GB201415079D0|2014-10-08|
    GB2529637A|2016-03-02|
    US11113610B2|2021-09-07|
    FR3025337B1|2019-08-16|
    US20160063384A1|2016-03-03|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US20040243636A1|2003-03-18|2004-12-02|Smartsignal Corporation|Equipment health monitoring architecture for fleets of assets|
    US7853431B2|2006-09-29|2010-12-14|Fisher-Rosemount Systems, Inc.|On-line monitoring and diagnostics of a process using multivariate statistical analysis|
    US20100223211A1|2000-10-11|2010-09-02|Johnson Gregory A|Decision service method and system|
    US6609051B2|2001-09-10|2003-08-19|Daimlerchrysler Ag|Method and system for condition monitoring of vehicles|
    US7107585B2|2002-07-29|2006-09-12|Arm Limited|Compilation of application code in a data processing apparatus|
    US6751536B1|2002-12-04|2004-06-15|The Boeing Company|Diagnostic system and method for enabling multistage decision optimization for aircraft preflight dispatch|
    US7640145B2|2005-04-25|2009-12-29|Smartsignal Corporation|Automated model configuration and deployment system for equipment health monitoring|
    US20070088570A1|2005-10-18|2007-04-19|Honeywell International, Inc.|System and method for predicting device deterioration|
    CA2676441C|2006-02-03|2015-11-24|Recherche 2000 Inc.|Intelligent monitoring system and method for building predictive models and detecting anomalies|
    EP1818746A1|2006-02-10|2007-08-15|ALSTOM Technology Ltd|Method of condition monitoring|
    US20070203869A1|2006-02-28|2007-08-30|Microsoft Corporation|Adaptive semantic platform architecture|
    WO2008044227A2|2006-07-17|2008-04-17|Open Pricer|Customer centric revenue management|
    FR2905763B1|2006-09-11|2008-12-12|Eurocopter France|METHOD AND SYSTEM FOR DIAGNOSING AN AIRCRAFT FROM MEASUREMENTS CARRIED OUT ON THE AIRCRAFT.|
    GB2496386A|2011-11-08|2013-05-15|Ge Aviat Systems Ltd|Method for integrating models of a vehicle health management system|US11181898B2|2017-11-10|2021-11-23|General Electric Company|Methods and apparatus to generate a predictive asset health quantifier of a turbine engine|
    US11086637B1|2018-04-03|2021-08-10|Akamai Technologies, Inc.|Configuration transformation and delivery|
    US11200069B1|2020-08-21|2021-12-14|Honeywell International Inc.|Systems and methods for generating a software application|
    法律状态:
    2016-08-25| PLFP| Fee payment|Year of fee payment: 2 |
    2017-08-25| PLFP| Fee payment|Year of fee payment: 3 |
    2017-11-17| PLSC| Publication of the preliminary search report|Effective date: 20171117 |
    2018-07-20| PLFP| Fee payment|Year of fee payment: 4 |
    2019-07-22| PLFP| Fee payment|Year of fee payment: 5 |
    2020-07-21| PLFP| Fee payment|Year of fee payment: 6 |
    2021-07-22| PLFP| Fee payment|Year of fee payment: 7 |
    优先权:
    申请号 | 申请日 | 专利标题
    GB1415079.1|2014-08-26|
    GB1415079.1A|GB2529637B|2014-08-26|2014-08-26|System for building and deploying inference model|
    [返回顶部]