Joint fault early warning model test method based on mathematical modeling and mechanism analysis

Abstract

The application discloses a joint fault early warning model test method and device based on mathematical modeling and mechanism analysis, electronic equipment and a storage medium. Among them, the method comprises the following steps: obtaining an early warning model graph of a fault early warning model; checking the functions and design of the fault early warning model based on the early warning model graph; encapsulating the fault early warning model into operator blocks of different function types based on the early warning model graph and a big data platform, so as to deploy the fault early warning model to the big data platform; setting the operator blocks based on the early warning model graph and the function types; and testing the fault early warning model. Through the technical scheme of the application, the fault early warning model can be tested in multiple aspects and targeted, so as to ensure that the fault early warning model can accurately perform early warning.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a test method, device, electronic device and storage medium for a joint fault early warning model based on mathematical modeling and mechanism analysis. Background Technology

[0002] Among related technologies, big data analytics and neural network models are widely used in smart power plant applications. However, thermal power plant systems are large, involve a wide range of disciplines, have strong coupling between systems, complex relationships, and variable operating environments. This results in fault warning models used for early warning of faults in thermal power plants having relatively complex structures and poor accuracy and timeliness in fault diagnosis. Therefore, testing fault warning models is also quite difficult. Summary of the Invention

[0003] This application provides a testing method, apparatus, electronic device, and storage medium for a joint fault early warning model based on mathematical modeling and mechanism analysis. It can test the fault early warning model to ensure that the model can accurately provide early warnings.

[0004] In a first aspect, embodiments of this application provide a testing method for a joint fault early warning model based on mathematical modeling and mechanistic analysis, comprising: obtaining an early warning model diagram of a fault early warning model; performing functional and design verification on the fault early warning model based on the early warning model diagram; encapsulating the fault early warning model into operator blocks of different functional types based on the early warning model diagram and a big data platform, so as to deploy the fault early warning model to the big data platform; setting the operator blocks based on the early warning model diagram and the functional types; and testing the fault early warning model; wherein, the early warning model diagram includes the early warning logic and model structure of the fault early warning model; the fault early warning model includes a mathematical fault early warning model and / or a mechanistic fault early warning model; the functional types include at least one of the following: time type, comparison type, function type, mathematical operation type, control type, device drive type, sequential control type, and start / stop type; and the testing includes at least one of the following: online testing, control strategy logic diagram testing, and joint testing.

[0005] In this technical solution, the fault early warning model can be deployed to a big data platform to simulate the actual operating environment of the fault early warning model and test it, thereby ensuring that the fault early warning model can accurately issue early warnings.

[0006] In one implementation, setting the operator block based on the early warning model diagram and the function type includes: responding to the function type of the operator block being the time type, setting the delay time and / or pulse time of the operator block according to the early warning model diagram; or, responding to the function type of the operator block being the comparison type, setting the input measurement point and comparison category according to the early warning model diagram, and setting the comparison threshold of the operator block according to the input measurement point; or, responding to the function type of the operator block being the function type, setting the function relationship and function value of the operator block according to the early warning model diagram; or, responding to the function type of the operator block being the mathematical operation type. The operation relationship of the operator block is set according to the warning model diagram; or, in response to the function type of the operator block being the control type, the control strategy of the operator block and the corresponding device to be controlled are set according to the warning model diagram; or, in response to the function type of the operator block being the device drive type, the drive device corresponding to the operator block is set according to the warning model diagram; or, in response to the function type of the operator block being the sequential control type, the sequential control logic of the operator block is set according to the warning model diagram; or, in response to the function type of the operator block being the start / stop type, the start / stop control command of the operator block and the corresponding start / stop control device are set according to the warning model diagram.

[0007] In one implementation, the test is the online test, and the testing of the fault warning model includes: acquiring a fault warning information database; acquiring test data based on the fault warning information database; performing static testing on the fault warning model to obtain a static test accuracy rate, and determining whether the static test accuracy rate reaches an accuracy rate threshold; performing a static warning display test on the fault warning model on the big data platform based on the test data, acquiring a static warning display accuracy rate, and determining whether the static warning display accuracy rate reaches a first accuracy rate threshold; and testing whether the fault warning model meets refresh rate requirements based on the test data.

[0008] In one optional implementation, the fault warning model is the mathematical fault warning model, and the method further includes: testing whether the computation time of the mathematical fault warning model meets the requirements based on the test data; and testing whether the accuracy of the mathematical fault warning model in data retrieval reaches a second accuracy threshold based on the test data.

[0009] In one implementation, the fault warning model is the mechanism fault warning model, the test is the control strategy logic diagram test, and the testing of the fault warning model includes: acquiring warning test data and determining whether the mechanism fault warning model can issue warning information normally based on the warning test data; and verifying the judgment control logic of the mechanism fault warning model.

[0010] In one implementation, the fault warning model is the mechanistic fault warning model, the test is the control strategy logic diagram test, and the testing of the fault warning model includes: acquiring warning test data and corresponding actual warning data; acquiring warning output data of the mathematical fault warning model based on the warning test data; comparing the warning output data with the actual warning data to obtain a comparison result; determining that the control strategy logic diagram test of the mathematical fault warning model has passed if the comparison result shows that the warning output data and the actual warning data are the same; or, optimizing the mathematical fault warning model based on the comparison result if the comparison result shows that the warning output data and the actual warning data are different.

[0011] In one implementation, the test is the joint test, and the testing of the fault warning model includes: verifying the function of the operator block based on the function type; constructing fault warning logic based on the fault warning model; in response to the function type of the operator block including the function type and / or the time type, deleting the operator block of the function type and / or the operator block of the time type based on the fault warning logic to obtain a reference fault warning model; inputting warning test data into the fault warning model to obtain first warning information; inputting the warning test data into the reference fault warning model to obtain second warning information; and determining whether the function of the operator block of the function type and / or the operator block of the time type is normal based on the first warning information and the second warning information.

[0012] In one optional implementation, the step of verifying the function of the operator block based on the function type includes: in response to the operator block's function type being the time type, verifying whether the delay time function and / or pulse time function of the operator block are normal based on the delay time and / or the pulse time; or, in response to the operator block's function type being the comparison type, verifying whether the comparison result of the operator block is correct based on the comparison threshold; or, in response to the operator block's function type being the function type, verifying whether the function calculation result of the operator block is correct based on the function relationship and the function value; or, in response to the operator block's function type being the mathematical operation... The operator block can be configured to: verify whether its operation result is correct based on the operational relationship; or, in response to the operator block's function type being the control type, verify whether the operator block can control the device to be controlled based on the control strategy; or, in response to the operator block's function type being the device drive type, verify whether the operator block can drive the drive device; or, in response to the operator block's function type being the sequential control type, verify whether the operator block can achieve automated control based on the control logic; or, in response to the operator block's function type being the start / stop type, verify whether the operator block can control the start / stop control device to open and / or close based on the start / stop control command.

[0013] In one implementation, there are multiple fault warning models, and the method further includes: generating a fault warning model library based on the multiple fault warning models; acquiring test input data and test output data; inputting the test input data into the multiple fault warning models to obtain the actual output data of the multiple fault warning models; determining whether the transfer of the fault warning model library is correct based on the actual output data and the test output data; sending a first input signal and a target fault warning model request signal to the multiple fault warning models based on the big data platform to obtain a first output result output by the fault warning model library; sending a second input signal and the target fault warning model request signal to the multiple fault warning models based on the big data platform to obtain a second output result output by the fault warning model library; wherein the second input signal is a deviation signal of the first input signal; and determining whether the application of the fault warning model library is normal based on the first output result and / or the second output result.

[0014] Secondly, embodiments of this application provide a testing device for a fault early warning model, comprising: a first acquisition module for acquiring an early warning model diagram of the fault early warning model; a verification module for verifying the function and design of the fault early warning model based on the early warning model diagram; a deployment module for encapsulating the fault early warning model into operator blocks of different functional types based on the early warning model diagram and a big data platform, so as to deploy the fault early warning model to the big data platform; a first processing module for setting the operator blocks based on the early warning model diagram and the functional types; and a testing module for testing the fault early warning model; wherein, the early warning model diagram includes the early warning logic and model structure of the fault early warning model; the fault early warning model includes a mathematical fault early warning model and / or a mechanistic fault early warning model; the functional types include at least one of the following: time type, comparison type, function type, mathematical operation type, control type, device drive type, sequential control type, and start / stop type; and the tests include at least one of the following: online testing, control strategy logic diagram testing, and joint testing.

[0015] In one implementation, the first processing module is specifically configured to: in response to the function type of the operator block being the time type, set the delay time and / or pulse time of the operator block according to the early warning model diagram; or, in response to the function type of the operator block being the comparison type, set the input measurement point and comparison category according to the early warning model diagram, and set the comparison threshold of the operator block according to the input measurement point; or, in response to the function type of the operator block being the function type, set the function relationship and function value of the operator block according to the early warning model diagram; or, in response to the function type of the operator block being the mathematical operation type, set the delay time and / or pulse time of the operator block according to the early warning model diagram. The operation relationship of the operator block is set; or, in response to the function type of the operator block being the control type, the control strategy of the operator block and the corresponding device to be controlled are set according to the early warning model diagram; or, in response to the function type of the operator block being the device driver type, the driver device corresponding to the operator block is set according to the early warning model diagram; or, in response to the function type of the operator block being the sequential control type, the sequential control logic of the operator block is set according to the early warning model diagram; or, in response to the function type of the operator block being the start / stop type, the start / stop control command of the operator block and the corresponding start / stop control device are set according to the early warning model diagram.

[0016] In one implementation, the test is the online test, and the test module is specifically used for: acquiring a fault warning information database; acquiring test data based on the fault warning information database; performing static testing on the fault warning model to obtain a static test accuracy rate, and determining whether the static test accuracy rate reaches an accuracy rate threshold; performing a static warning display test on the fault warning model on the big data platform based on the test data, acquiring a static warning display accuracy rate, and determining whether the static warning display accuracy rate reaches a first accuracy rate threshold; and testing whether the fault warning model meets the refresh rate requirements based on the test data.

[0017] In one optional implementation, the fault warning model is the mathematical fault warning model, and the method further includes: testing whether the computation time of the mathematical fault warning model meets the requirements based on the test data; and testing whether the accuracy of the mathematical fault warning model in data retrieval reaches a second accuracy threshold based on the test data.

[0018] In one implementation, the fault warning model is the mechanism fault warning model, the test is the control strategy logic diagram test, and the test module is specifically used to: acquire warning test data, and determine whether the mechanism fault warning model can issue warning information normally based on the warning test data; and verify the judgment control logic of the mechanism fault warning model.

[0019] In one implementation, the fault warning model is the mechanistic fault warning model, the test is the control strategy logic diagram test, and the test module is specifically used for: acquiring warning test data and corresponding actual warning data; acquiring warning output data of the mathematical fault warning model based on the warning test data; comparing the warning output data with the actual warning data to obtain a comparison result; determining that the control strategy logic diagram test of the mathematical fault warning model has passed if the comparison result shows that the warning output data and the actual warning data are the same; or, optimizing the mathematical fault warning model based on the comparison result if the comparison result shows that the warning output data and the actual warning data are different.

[0020] In one implementation, the test is the joint test, and the test module is specifically used for: verifying the function of the operator block based on the function type; constructing fault warning logic based on the fault warning model; in response to the function type of the operator block including the function type and / or the time type, deleting the operator block of the function type and / or the operator block of the time type based on the fault warning logic, and obtaining a reference fault warning model; inputting the warning test data into the fault warning model to obtain first warning information; inputting the warning test data into the reference fault warning model to obtain second warning information; and determining whether the function of the operator block of the function type and / or the operator block of the time type is normal based on the first warning information and the second warning information.

[0021] In one optional implementation, the testing module is specifically configured to: in response to the operator block's function type being the time type, verify whether the delay time function and / or pulse time function of the operator block are normal based on the delay time and / or the pulse time; or, in response to the operator block's function type being the comparison type, verify whether the comparison result of the operator block is correct based on the comparison threshold; or, in response to the operator block's function type being the function type, verify whether the function calculation result of the operator block is correct based on the function relationship and the function value; or, in response to the operator block's function type being the mathematical operation type, verify whether the operation result is correct based on the mathematical operation type. The algorithm verifies whether the operation result of the operator block is correct; or, in response to the function type of the operator block being the control type, it verifies whether the operator block can control the device to be controlled based on the control strategy; or, in response to the function type of the operator block being the device drive type, it verifies whether the operator block can drive the drive device; or, in response to the function type of the operator block being the sequential control type, it verifies whether the operator block can achieve automated control based on the control logic; or, in response to the function type of the operator block being the start / stop type, it verifies whether the operator block can control the start / stop control device to start and / or stop based on the start / stop control command.

[0022] In one implementation, there are multiple fault warning models, and the device further includes: a generation module for generating a fault warning model library based on the multiple fault warning models; a second acquisition module for acquiring test input data and test output data; a second processing module for inputting the test input data into the multiple fault warning models to acquire the actual output data of the multiple fault warning models; a first judgment module for judging whether the transfer of the fault warning model library is correct based on the actual output data and the test output data; a third processing module for sending a first input signal and a target fault warning model request signal to the multiple fault warning models based on the big data platform to acquire a first output result output by the fault warning model library; a fourth processing module for sending a second input signal and the target fault warning model request signal to the multiple fault warning models based on the big data platform to acquire a second output result output by the fault warning model library; wherein the second input signal is a deviation signal of the first input signal; and a second judgment module for judging whether the application of the fault warning model library is normal based on the first output result and / or the second output result.

[0023] Thirdly, embodiments of this application provide an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the joint fault early warning model testing method based on mathematical modeling and mechanism analysis as described in the first aspect.

[0024] Fourthly, embodiments of this application provide a computer-readable storage medium for storing instructions that, when executed, cause the method described in the first aspect to be implemented.

[0025] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the steps of the joint fault early warning model testing method based on mathematical modeling and mechanism analysis as described in the first aspect.

[0026] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent from the following description. Attached Figure Description

[0027] The accompanying drawings are provided for a better understanding of this solution and do not constitute a limitation of this application. Wherein:

[0028] Figure 1This is a schematic diagram of a joint fault early warning model testing method based on mathematical modeling and mechanism analysis provided in an embodiment of this application;

[0029] Figure 2 This is a schematic diagram of another joint fault early warning model testing method based on mathematical modeling and mechanism analysis provided in the embodiments of this application;

[0030] Figure 3 This is a schematic diagram of another fault early warning model testing method based on mathematical modeling and mechanism analysis provided in the embodiments of this application;

[0031] Figure 4 This is a schematic diagram of another fault early warning model testing method based on mathematical modeling and mechanism analysis provided in the embodiments of this application;

[0032] Figure 5 This is a schematic diagram of another fault early warning model testing method based on mathematical modeling and mechanism analysis provided in the embodiments of this application;

[0033] Figure 6 This is a schematic diagram of a testing method for a fault early warning model library provided in an embodiment of this application;

[0034] Figure 7 This is a schematic diagram of a test device for a joint fault early warning model based on mathematical modeling and mechanism analysis provided in an embodiment of this application;

[0035] Figure 8 This is a schematic diagram of another fault early warning model test device based on mathematical modeling and mechanism analysis provided in the embodiments of this application;

[0036] Figure 9 This is a schematic block diagram of an example electronic device that can be used to implement embodiments of this application. Detailed Implementation

[0037] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of this application, including various details to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0038] In the description of this application, unless otherwise stated, " / " means "or", for example, A / B can mean A or B; "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. The various numerical designations such as "first", "second", etc., involved in this application are only for the convenience of description and are not used to limit the scope of the embodiments of this application, nor do they indicate the order of events.

[0039] Please see Figure 1 , Figure 1 This is a schematic diagram of a joint fault early warning model testing method based on mathematical modeling and mechanism analysis provided in an embodiment of this application. Figure 1 As shown, the method may include, but is not limited to, the following steps:

[0040] Step S101: Obtain the early warning model diagram of the fault early warning model.

[0041] It should be noted that the fault early warning model in this application embodiment can be a fault early warning model used for early warning of thermal power generation systems. This fault early warning model can be built based on a fault early warning information database of various systems in a thermal power plant. For example, DCS (Distributed Control System) multi-functional operators, Python programming language, Java programming language, and PLC (Programmable Logic Controller) can be used to specifically model the models of each system to establish this fault early warning model.

[0042] In the embodiments of this application, the early warning model diagram includes the early warning logic and model structure of the fault early warning model; the fault early warning model includes a mathematical fault early warning model and / or a mechanistic fault early warning model.

[0043] As an example, we obtain the early warning model diagram of the mathematical fault early warning model.

[0044] As another example, obtain the early warning model diagram of the mechanism fault early warning model.

[0045] As another example, we obtain the early warning model diagrams of mathematical fault early warning models and mechanistic fault early warning models.

[0046] It should be noted that, in the embodiments of this application, the mathematical fault early warning model is a fault early warning model that realizes online observation of the accident conditions that are about to occur or are occurring through computer language and machine learning, and completes the framework early warning of the system or equipment; the mechanism fault early warning model is based on the logic configuration of operator blocks and mathematical and mechanism model diagrams, and realizes the analysis and judgment of accidents by setting preset values, integrating multiple judgment conditions, and adding preset conditions, thereby realizing the early warning of the system or equipment.

[0047] Step S102: Based on the early warning model diagram, verify the function and design of the fault early warning model.

[0048] For example, based on the early warning model diagram, the function and design of the fault early warning model are checked to confirm whether the fault early warning model can realize the pre-designed early warning function according to the preset early warning logic.

[0049] In one implementation of this application, in response to determining the function of the fault warning model and the fact that it does not meet the functional and design requirements, the fault warning model can be remodeled to ensure the accuracy and authenticity of the fault warning model.

[0050] Step S103: Based on the early warning model diagram and the big data platform, the fault early warning model is encapsulated into operator blocks of different functional types, so as to deploy the fault early warning model to the big data platform.

[0051] In the embodiments of this application, the above-mentioned function types include at least one of the following: time type, comparison type, function type, mathematical operation type, control type, device drive type, sequential control type, and start / stop type.

[0052] For example, based on the early warning model diagram, the various parts of the fault early warning model are encapsulated into operator blocks of different functional types, and the operator blocks are deployed to the big data platform to integrate the fault early warning model with the big data platform.

[0053] Step S104: Configure the operator block based on the early warning model diagram and function type.

[0054] For example, based on the warning model diagram and the function type of each operator block, the internal settings of each operator block are set.

[0055] In one optional implementation, the above-mentioned setting of the operator block based on the early warning model diagram and function type includes: responding to the operator block's function type being time-based, setting the delay time and / or pulse time of the operator block according to the early warning model diagram; or, responding to the operator block's function type being comparison-based, setting the input measurement point and comparison category according to the early warning model diagram, and setting the comparison threshold of the operator block according to the input measurement point; or, responding to the operator block's function type being function-based, setting the function relationship and function value of the operator block according to the early warning model diagram; or, responding to the operator block's function type being mathematical... The operation type is determined by setting the operation relationship of the operator block according to the early warning model diagram; or, in response to the function type of the operator block being control type, the control strategy of the operator block and the corresponding device to be controlled are set according to the early warning model diagram; or, in response to the function type of the operator block being device drive type, the drive device corresponding to the operator block is set according to the early warning model diagram; or, in response to the function type of the operator block being sequential control type, the sequential control logic of the operator block is set according to the early warning model diagram; or, in response to the function type of the operator block being start / stop type, the start / stop control command of the operator block and the corresponding start / stop control device are set according to the early warning model diagram.

[0056] As an example, in response to the operator block's function type being time-based, the specific value of the operator block's delay time is set according to the warning model diagram.

[0057] As another example, in response to the operator block's function type being time-based, the specific value of the operator block's pulse time is set according to the warning model diagram.

[0058] As another example, in response to the operator block's function type being comparison type, the operator block is set to either small selection comparison (i.e., comparing to obtain the smaller value among multiple values) or large selection comparison (i.e., comparing to obtain the larger value among multiple values) according to the early warning model diagram. The measurement points corresponding to the measurement data input to the operator block are set, and the specific value of the comparison threshold for comparison with the measurement data is set.

[0059] As another example, in response to the operator block's function type being a function type, the function relationships and parameter values ​​used to calculate the input data within the operator block are set according to the early warning model diagram.

[0060] As another example, in response to the operator block's function type being mathematical operation type, the operational relationships and corresponding operational rules used to perform data calculations on the input data are set within the operator block according to the early warning model diagram.

[0061] As another example, in response to the operator block's function type being control type, the control strategy of the operator block and the corresponding device to be controlled are set according to the early warning model diagram. For example, the operator block is set to control the power level of the device to be controlled when it receives a pre-set signal.

[0062] As another example, in response to the operator block's function type being sequential control, the configuration and control requirements for the operator block's sequential control steps are set according to the early warning model diagram.

[0063] As another example, in response to the operator block's function type being start / stop, the operator block is configured to control the corresponding device to start or stop based on different received instructions, according to the early warning model diagram.

[0064] Step S105: Test the fault early warning model.

[0065] In the embodiments of this application, the above tests include at least one of the following: online testing, control strategy logic diagram testing, and joint testing.

[0066] By implementing the embodiments of this application, the fault warning model can be deployed to a big data platform to test the fault warning model in a simulated actual operating environment, thereby ensuring that the fault warning model can accurately issue warnings.

[0067] In one implementation, a fault warning information database can be obtained, and test data can be acquired based on this database to conduct online testing of the fault warning model. For an example, please refer to [link to example]. Figure 2 , Figure 2 This is a schematic diagram of another fault early warning model testing method based on mathematical modeling and mechanism analysis provided in an embodiment of this application. Figure 2 As shown, the method may include, but is not limited to, the following steps:

[0068] Step S201: Obtain the early warning model diagram of the fault early warning model.

[0069] In the embodiments of this application, step S201 can be implemented in any of the ways described in the embodiments of this application. The embodiments of this application do not limit this, nor will they be described in detail.

[0070] Step S202: Based on the early warning model diagram, verify the function and design of the fault early warning model.

[0071] In the embodiments of this application, step S202 can be implemented in any of the ways described in the embodiments of this application. The embodiments of this application do not limit this, nor will they be described in detail.

[0072] Step S203: Based on the early warning model diagram and the big data platform, the fault early warning model is encapsulated into operator blocks of different functional types, so as to deploy the fault early warning model to the big data platform.

[0073] In the embodiments of this application, step S203 can be implemented in any of the ways described in the various embodiments of this application. The embodiments of this application do not limit this, nor will they be described in detail.

[0074] Step S204: Configure the operator blocks based on the early warning model diagram and function type.

[0075] In the embodiments of this application, step S204 can be implemented in any of the ways described in the embodiments of this application. The embodiments of this application do not limit this, nor will they be described in detail.

[0076] Step S205: Obtain the fault warning information database.

[0077] For example, a fault warning information database is formed by acquiring fault information that requires early warning.

[0078] Step S206: Obtain test data based on the fault early warning information database.

[0079] For example, early warning information data is obtained from the fault early warning information database as test data, and the time interval of the test data is guaranteed to be a preset interval. The names of each measuring point from which the test data comes are determined (for example, the KKS (Kraftwerk-Kennzeichensystem, power plant identification system) code of the corresponding measuring point can be determined from the measuring point table).

[0080] In some embodiments of this application, the test data can be divided into digital signals and analog signals, and the states of the digital signals and analog signals can be confirmed to facilitate testing.

[0081] Step S207: Perform static testing on the fault warning model to obtain the static test accuracy and determine whether the static test accuracy reaches the accuracy threshold.

[0082] For example, the syntax, structure, process, and interface of the fault warning model program are analyzed and checked to obtain the program's accuracy as the static test accuracy, and to determine whether the static test accuracy reaches the accuracy threshold (e.g., 100%).

[0083] Step S208: Based on the test data, conduct a static test of the mathematical fault early warning model on the big data platform to obtain the accuracy rate of the static test of early warning display, and determine whether the accuracy rate of the static test of early warning display reaches the first accuracy threshold.

[0084] For example, test data is input into the fault warning model, causing the fault warning model to issue a warning message. This allows the big data platform to perform a static test of the warning display based on the warning message, obtain the accuracy rate of the static test of the warning display, and determine whether the accuracy rate of the static test of the warning display reaches the first accuracy threshold (e.g., 100%).

[0085] Step S209: Based on the test data, test whether the fault early warning model meets the refresh rate requirements.

[0086] For example, based on test data, the refresh rate of the warning display of the fault warning model is tested to determine whether the warning display refresh rate reaches the preset standard refresh time (e.g., 2 seconds).

[0087] In one optional implementation, the aforementioned fault warning model is a mathematical fault warning model, and the method further includes: testing whether the computation time of the mathematical fault warning model meets the requirements based on test data; and testing whether the accuracy of the mathematical fault warning model in data retrieval reaches a second accuracy threshold based on test data.

[0088] For example, the mathematical fault early warning model is invoked once every first preset time interval (e.g., 2 minutes), and the output result of the mathematical fault early warning model is obtained. The actual number of times the mathematical fault early warning model is invoked is obtained within a second preset time interval (e.g., 10 minutes) to test whether the computation time of the mathematical fault early warning model meets the requirements. The number of early warnings issued by the mathematical fault early warning model within a preset time period (e.g., 1 month) is compared with the actual situation to obtain the accuracy rate of the early warning, and it is determined whether this accuracy rate reaches a second accuracy threshold (e.g., 90%).

[0089] As an example, let's take a first preset time of 2 minutes and a second preset time of 10 minutes. Within the 10 minutes, the fault warning model is called every 2 minutes. The output of each call to the mathematical fault warning model is obtained, along with the total number of calls to the mathematical fault warning model within the 10 minutes. If the total number of calls to the mathematical fault warning model within 10 minutes is greater than or equal to 5, it is determined whether the computation time of the mathematical fault warning model meets the requirements; or, if the total number of calls to the mathematical fault warning model within 10 minutes is less than 5, it is determined that the computation time of the mathematical fault warning model does not meet the requirements.

[0090] By implementing the embodiments of this application, a fault warning information database can be obtained, and test data can be obtained based on the fault warning information database to conduct static testing, warning display static testing, and refresh rate testing on the fault warning model, thereby conducting online testing of the fault warning model and ensuring that the fault warning model can accurately issue warnings.

[0091] In one implementation, the aforementioned fault warning model is a mechanism-based fault warning model, and the aforementioned test is a control strategy logic diagram test. For an example, please refer to [link to example]. Figure 3 , Figure 3 This is a schematic diagram of another fault early warning model testing method based on mathematical modeling and mechanism analysis provided in the embodiments of this application. Figure 3 As shown, the method may include, but is not limited to, the following steps:

[0092] Step S301: Obtain the early warning model diagram of the fault early warning model.

[0093] In the embodiments of this application, step S301 can be implemented in any of the ways described in the various embodiments of this application. The embodiments of this application do not limit this, nor will they be described in detail.

[0094] Step S302: Based on the early warning model diagram, verify the function and design of the fault early warning model.

[0095] In the embodiments of this application, step S302 can be implemented in any of the ways described in the embodiments of this application. The embodiments of this application do not limit this, nor will they be described in detail.

[0096] Step S303: Based on the early warning model diagram and the big data platform, the fault early warning model is encapsulated into operator blocks of different functional types, so as to deploy the fault early warning model to the big data platform.

[0097] In the embodiments of this application, step S303 can be implemented in any of the ways described in the embodiments of this application. The embodiments of this application do not limit this, nor will they be described in detail.

[0098] Step S304: Configure the operator block based on the early warning model diagram and function type.

[0099] In the embodiments of this application, step S304 can be implemented in any of the ways described in the embodiments of this application. The embodiments of this application do not limit this, nor will they be described in detail.

[0100] Step S305: Obtain early warning test data and determine whether the mechanism fault early warning model can issue early warning information normally based on the early warning test data.

[0101] Understandably, since the input measurement point data is usually insufficient to trigger the early warning function of the mechanism fault early warning model, the early warning can be triggered by changing the relevant settings of the operator block. If the early warning is triggered immediately after the relevant settings are changed, it is determined that the mechanism fault early warning model can issue early warning information normally; otherwise, it is determined that the mechanism fault early warning model cannot issue early warning information normally.

[0102] In some embodiments of this application, if the fault warning model includes a time-type operator block with a delay, the warning delay trigger time can be verified by setting the delay time. If the warning is triggered after the delay time set in the operator block, it is determined whether the mechanism fault warning model can issue warning information normally; otherwise, it is determined that the mechanism fault warning model cannot issue warning information normally.

[0103] Step S306: Verify the judgment control logic of the mechanism fault early warning model.

[0104] Understandably, when a mechanism fault early warning model includes multiple decision control logics, each logic needs to be verified individually. Furthermore, after verifying one logic, the mechanism fault early warning model must be restored to its normal, non-alert state before proceeding to verify the next logic. During verification, it is crucial to maintain consistency between the precision of the input and output values ​​of each operator block.

[0105] By implementing the embodiments of this application, it is possible to determine whether the mechanism fault early warning model can issue early warning information normally based on early warning test data, and to verify the judgment control logic of the mechanism fault early warning model, so as to test the control strategy logic diagram of the mechanism fault early warning model, thereby ensuring that the mechanism fault early warning model can issue early warning normally.

[0106] In one implementation, the aforementioned fault warning model is a mathematical fault warning model, and the aforementioned test is a control strategy logic diagram test. For an example, please refer to... Figure 4 , Figure 4 This is a schematic diagram of another fault early warning model testing method based on mathematical modeling and mechanism analysis provided in the embodiments of this application. Figure 4 As shown, the method may include, but is not limited to, the following steps:

[0107] Step S401: Obtain the early warning model diagram of the fault early warning model.

[0108] In the embodiments of this application, step S401 can be implemented in any of the ways described in the various embodiments of this application. The embodiments of this application do not limit this, nor will they be described in detail.

[0109] Step S402: Based on the early warning model diagram, verify the function and design of the fault early warning model.

[0110] In the embodiments of this application, step S402 can be implemented in any of the ways described in the embodiments of this application. The embodiments of this application do not limit this, nor will they be described in detail.

[0111] Step S403: Based on the early warning model diagram and the big data platform, the fault early warning model is encapsulated into operator blocks of different functional types to deploy the fault early warning model to the big data platform.

[0112] In the embodiments of this application, step S403 can be implemented in any of the ways described in the embodiments of this application. The embodiments of this application do not limit this, nor will they be described in detail.

[0113] Step S404: Configure the operator block based on the early warning model diagram and function type.

[0114] In the embodiments of this application, step S404 can be implemented in any of the ways described in the embodiments of this application. The embodiments of this application do not limit this, nor will they be described in detail.

[0115] Step S405: Obtain the early warning test data and the corresponding actual early warning data.

[0116] Step S406: Based on the early warning test data, obtain the early warning output data of the mathematical fault early warning model.

[0117] For example, the early warning test data is input into the mathematical fault early warning model, and the early warning output data output by the mathematical fault early warning model is obtained.

[0118] Step S407: Compare the warning output data with the actual warning data to obtain the comparison results.

[0119] For example, the warning output data is compared with the actual warning data to obtain the comparison results of whether the warning output data and the actual warning data are consistent.

[0120] Step S408: In response to the comparison result that the warning output data is the same as the actual warning data, determine that the logic diagram test of the control strategy of the mathematical fault warning model has passed; or, in response to the comparison result that the warning output data is different from the actual warning data, optimize the mathematical fault warning model based on the comparison result.

[0121] As an example, in response to the comparison result showing that the warning output data is the same as the actual warning data, it is determined that the logic diagram test of the control strategy of the mathematical fault warning model has passed.

[0122] As another example, if the comparison result shows that the warning output data is different from the actual warning data, it means that the mathematical fault warning model has missed reports and / or false reports. In this case, the mathematical fault warning model is optimized based on the comparison result.

[0123] By implementing the embodiments of this application, the early warning output data of the mathematical fault early warning model can be obtained based on the early warning test data, and the early warning output data can be compared with the actual early warning data to obtain the comparison results. Based on the comparison results, the control strategy logic diagram of the fault early warning model can be tested to ensure that the fault early warning model can accurately issue early warnings.

[0124] In one implementation, the above tests are joint tests. For an example, please refer to [link to example]. Figure 5 , Figure 5 This is a schematic diagram of another fault early warning model testing method based on mathematical modeling and mechanism analysis provided in the embodiments of this application. Figure 5 As shown, the method may include, but is not limited to, the following steps:

[0125] Step S501: Obtain the early warning model diagram of the fault early warning model.

[0126] In the embodiments of this application, step S501 can be implemented in any of the ways described in the embodiments of this application. The embodiments of this application do not limit this, nor will they be described in detail.

[0127] Step S502: Based on the early warning model diagram, verify the function and design of the fault early warning model.

[0128] In the embodiments of this application, step S502 can be implemented in any of the embodiments of this application. The embodiments of this application do not limit this and will not be described in detail.

[0129] Step S503: Based on the early warning model diagram and the big data platform, the fault early warning model is encapsulated into operator blocks of different functional types to deploy the fault early warning model to the big data platform.

[0130] In the embodiments of this application, step S503 can be implemented in any of the embodiments of this application. The embodiments of this application do not limit this and will not be described in detail.

[0131] Step S504: Configure the operator block based on the early warning model diagram and function type.

[0132] In the embodiments of this application, step S504 can be implemented in any of the embodiments of this application. The embodiments of this application do not limit this and will not be described in detail.

[0133] Step S505: Perform functional verification on the operator block based on the function type.

[0134] For example, based on the function type corresponding to each operator block, each operator block is functionally verified to test whether each operator block can achieve the corresponding preset function.

[0135] In one optional implementation, the above-mentioned functional verification of the operator block based on the function type includes: responding to the operator block's function type being time-based, verifying whether the delay time function and / or pulse time function of the operator block are normal based on the delay time and / or pulse time; or, responding to the operator block's function type being comparison-based, verifying whether the comparison result of the operator block is correct based on a comparison threshold; or, responding to the operator block's function type being function-based, verifying whether the function calculation result of the operator block is correct based on the function relationship and function value; or, responding to the operator block's function type being number-based... Based on the operation type, verify whether the operation result of the operator block is correct; or, in response to the function type of the operator block being control type, verify whether the operator block can control the device to be controlled based on the control strategy; or, in response to the function type of the operator block being device drive type, verify whether the operator block can drive the device; or, in response to the function type of the operator block being sequential control type, verify whether the operator block can achieve automated control based on control logic; or, in response to the function type of the operator block being start / stop type, verify whether the operator block can control the start / stop control device to start and / or stop based on start / stop control commands.

[0136] As an example, if the function type of the operator block is time-based, test whether the function of the operator block triggering the delay time is normal, based on the specific value of the delay time set in the operator block.

[0137] As another example, in response to the operator block's function type being time-based, the function of triggering the pulse time in the operator block is tested to determine if it functions correctly, based on the specific value of the pulse time set within the operator block.

[0138] As another example, in response to the operator block's function type being comparison type, a test value is input to the operator block to obtain its output. Based on the test value, the output, and the corresponding comparison threshold, the operator block's comparison function is tested to determine if it functions correctly. For example, assuming the operator block's comparison category is small selection comparison, the comparison threshold is 2, and the test value is 1: If the operator block's output is 1, then the operator block's comparison function is normal; conversely, if the operator block's output is 2, then the operator block's comparison function is abnormal.

[0139] As another example, in response to the operator block's function type being a function type, the test value is input into the operator block to obtain the operator block's output result. If the operator block's output result is the same as the actual result obtained based on the operator block's function relationship, the function's parameter values, and the test value, the operator block is judged to be functioning normally; otherwise, the operator block is judged to be functioning abnormally.

[0140] As another example, in response to the operator block's function type being mathematical operation type, the test value is input into the operator block to obtain the operator block's output result. If the operator block's output result is the same as the actual result calculated based on the operator block's operation relationship, operation rule, and test value, the operator block is judged to be functioning normally; otherwise, the operator block is judged to be functioning abnormally.

[0141] As another example, in response to the operator block's function type being control, a pre-set control signal is sent to the operator block, and it is detected whether the operator block can respond to the received control signal and control the device to be controlled. If the operator block can respond to the received control signal and control the device to be controlled, the operator block is considered to be functioning normally; otherwise, the operator block is considered to be functioning normally.

[0142] As another example, in response to the operator block's function type being sequential control, a pre-set sequential control signal is sent to the operator block. It is then checked whether the operator block can respond to the received sequential control signal and control according to the configured sequential control step sequence. If the operator block can respond to the received sequential control signal and control according to the configured sequential control step sequence, the operator block is considered to be functioning normally; otherwise, the operator block is considered to be malfunctioning.

[0143] As another example, in response to the operator block's function type being start / stop, a pre-set start / stop control command is sent to the operator block. The system then checks whether the operator block can respond to the received start / stop control command and control the corresponding device to start or stop. If the operator block can respond to the received start / stop control command and control the corresponding device to start or stop, the operator block is considered to be functioning normally; otherwise, the operator block is considered to be malfunctioning.

[0144] Step S506: Construct fault early warning logic based on the fault early warning model.

[0145] For example, the early warning logic is based on a fault early warning model to construct a fault early warning model for early warning.

[0146] Step S507: In response to the function type of the operator block, including function type and / or time type, delete the operator block of function type and / or time type based on the fault warning logic, and obtain the comparison fault warning model.

[0147] As an example, in response to the function type of the operator block, including the function type, the operator block of the function type is deleted from the fault warning model based on the fault warning logic to obtain the reference fault warning model.

[0148] As another example, in response to the function type of the operator block including the time type, the time type operator block is removed from the fault warning model based on the fault warning logic to obtain the reference fault warning model.

[0149] As another example, in response to the function type of the operator block including function type and time type, the operator block of function type and the operator block of time type are deleted from the fault warning model based on the fault warning logic to obtain the comparison fault warning model.

[0150] Step S508: Input the early warning test data into the fault early warning model to obtain the first early warning information.

[0151] For example, the early warning test data is input into the fault early warning model, and the first early warning information output by the fault early warning model is obtained.

[0152] Step S509: Input the early warning test data into the reference fault early warning model to obtain the second early warning information.

[0153] For example, the early warning test data is input into the reference fault early warning model, and the second early warning information output by the reference fault early warning model is obtained.

[0154] Step S510: Based on the first warning information and the second warning information, determine whether the function type operator block and / or the time type operator block are functioning normally.

[0155] It should be noted that, in the embodiments of this application, the first warning information and the second warning information may include the warning result and / or the warning time.

[0156] As an example, in response to the consistency of the first and second warning messages, it is determined that the function type operator block and / or the time type operator block is malfunctioning.

[0157] As another example, in response to a discrepancy between the first and second warning messages, it is determined that the function type operator block and / or the time type operator block are functioning correctly.

[0158] By implementing the embodiments of this application, the function of the corresponding operator block of the fault warning model can be verified based on the function type of the operator block, and fault warning logic can be constructed based on the fault warning model to determine whether the function type operator block and / or time type operator block are functioning normally, so as to ensure that the fault warning model can accurately issue warnings.

[0159] In one implementation, multiple fault warning models can be used. A fault warning library can be generated based on these multiple models for transfer testing and application testing. For an example, please refer to [link to example]. Figure 6 , Figure 6 This is a schematic diagram of a testing method for a fault early warning model library provided in an embodiment of this application. For example... Figure 6 As shown, the method may include, but is not limited to, the following steps:

[0160] Step S601: Generate a fault early warning model library based on multiple fault early warning models.

[0161] For example, multiple fault early warning models corresponding to multiple systems in a thermal power plant can be combined to generate a fault early warning model library.

[0162] Step S602: Obtain test data for the fault early warning model library.

[0163] In the embodiments of this application, the test data for the fault warning model library includes the test data corresponding to each fault warning model in the fault warning model library.

[0164] Step S603: Input the test data of the fault early warning model library into the fault early warning model library and obtain the output data of the fault early warning model library.

[0165] For example, the test data corresponding to each fault warning model in the fault warning model library test data is input into the corresponding fault warning model to obtain the first output data of the fault warning model library.

[0166] Step S604: Detect whether the first output data includes the model outputs of multiple fault warning models, and determine whether the transfer of the fault warning model library is correct based on the detection results.

[0167] As an example, in response to the output data including the model output of each of the multiple fault warning models, it is determined that the fault warning model library transfer is correct.

[0168] As another example, in response to the output data not including the model output of each of the multiple fault warning models, it is determined that the fault warning model library transfer is correct.

[0169] Step S605: Send the first input signal and the target fault early warning model request signal to the fault early warning model library, and obtain the first output result output by the fault early warning model library.

[0170] For example, a first input signal is sent to the fault early warning model library through a big data platform, and a target fault early warning model request signal is sent to request the target early warning model to issue an early warning based on the first input signal, and the first output result output by the fault early warning model library is obtained.

[0171] In the embodiments of this application, the target fault warning model is any one of a plurality of fault warning models.

[0172] Step S606: Send a second input signal to the fault early warning model library and obtain the second output result output by the fault early warning model library.

[0173] For example, a second input signal is sent to the fault warning model library through a big data platform, and a target fault warning model request signal is sent to request the target warning model to issue a warning based on the second input signal, and the first output result output by the fault warning model library is obtained.

[0174] In the embodiments of this application, the second input signal is a deviation signal of the first input signal.

[0175] For example, the first input signal can be a normal signal that has a significant impact on the model, and the second input signal can be a signal of the same type as the first input signal but deviating from the normal value.

[0176] Step S607: Based on the first output result and / or the second output result, determine whether the application of the fault early warning model library is normal.

[0177] As an example, if the first output result is the output result of the target fault warning model, and the first output result and the second output result are inconsistent, it is determined that the application of the fault warning model library is normal.

[0178] As another example, if the response is that the first output result is the output result of the target fault warning model, and the first output result is consistent with the second output result, it is determined that the application of the fault warning model library is abnormal.

[0179] As another example, in response to the first output result not being the output result of the target fault warning model, it is determined that the application of the fault warning model library is abnormal.

[0180] By implementing the embodiments of this application, a fault warning library can be generated based on multiple fault warning models, and the fault warning model library can be transferred and applied to ensure that the fault warning model library can provide warnings normally.

[0181] In some embodiments of this application, for problems discovered during testing, the optimized fault warning model can be tested again after the problems are optimized and resolved. This allows for the optimization of the fault warning model's warning performance based on the test results.

[0182] In some embodiments of this application, before implementing the joint fault early warning model testing method based on mathematical modeling and mechanism analysis provided in any embodiment of this application, the software and hardware configuration of the big data platform can be checked, and the software configuration of the big data platform can be checked and tested.

[0183] In some embodiments of this application, data communication tests can also be performed on the fault warning model library. For example, output measurements can be performed on some switch signals and analog signals of different ranges (e.g., 0%, 25%, 50%, 75%, and 100%) of the fault warning model library. The measurement results are used to determine whether the signal output of the fault warning model library is normal. The data reception of the fault warning system is tested based on the switch signals and analog signals of different ranges. Analog signals of different ranges are sent to the fault warning model library to detect whether the fault warning model library can receive the signals normally, thereby determining whether the signal reception of the fault model library is normal.

[0184] In some embodiments of this application, the signal transmission between the big data platform and the fault early warning model library can also be tested to determine if it is normal. For example, one of the big data platform and the fault model library can be used as the server and the other as the client. The client sends a signal request, and the server can be tested to see if it can receive the signal by forcing the signal, thereby determining whether the signal transmission between the big data platform and the fault model library is normal.

[0185] In some embodiments of this application, the display test of the fault warning model and the big data platform can also be performed. For each fault warning model, normal operation-related data and fault operation-related data are input into the big data platform. After verification by the fault model library, it is checked whether the big data platform displays the model warning results normally; and it is also checked whether the push of the model warning results is normal (e.g., whether the relevant devices can receive the push of the model warning results).

[0186] Please see Figure 7 , Figure 7 This is a schematic diagram of a test device for a joint fault early warning model based on mathematical modeling and mechanism analysis, provided in an embodiment of this application. Figure 7 As shown, the device 700 includes: a first acquisition module 701, used to acquire the early warning model diagram of the fault early warning model; a verification module 702, used to perform functional and design verification of the fault early warning model based on the early warning model diagram; a deployment module 703, used to encapsulate the fault early warning model into operator blocks of different functional types based on the early warning model diagram and the big data platform, so as to deploy the fault early warning model to the big data platform; a first processing module 704, used to set the operator blocks based on the early warning model diagram and the functional types; and a testing module 705, used to test the fault early warning model; wherein, the early warning model diagram contains the early warning logic and model structure of the fault early warning model; the fault early warning model includes a mathematical fault early warning model and / or a mechanistic fault early warning model; the functional types include at least one of the following: time type, comparison type, function type, mathematical operation type, control type, device drive type, sequential control type, and start / stop type; and the tests include at least one of the following: online testing, control strategy logic diagram testing, and joint testing.

[0187] In one implementation, the first processing module 704 is specifically configured to: respond to the operator block's function type being time-based, set the delay time and / or pulse time of the operator block according to the early warning model diagram; or, respond to the operator block's function type being comparison-based, set the input measurement point and comparison category according to the early warning model diagram, and set the comparison threshold of the operator block according to the input measurement point; or, respond to the operator block's function type being function-based, set the function relationship and function value of the operator block according to the early warning model diagram; or, respond to the operator block's function type being mathematical operation-based, set the delay time and / or pulse time of the operator block according to the early warning model diagram. The diagram sets the operational relationships of the operator blocks; or, in response to the operator block's function type being control type, the control strategy of the operator block and the corresponding controlled device are set according to the early warning model diagram; or, in response to the operator block's function type being device drive type, the drive device corresponding to the operator block is set according to the early warning model diagram; or, in response to the operator block's function type being sequential control type, the sequential control logic of the operator block is set according to the early warning model diagram; or, in response to the operator block's function type being start / stop type, the start / stop control command of the operator block and the corresponding start / stop control device are set according to the early warning model diagram.

[0188] In one implementation, the testing is an online test. The testing module 705 is specifically used for: acquiring a fault warning information database; acquiring test data based on the fault warning information database; performing static testing on the fault warning model to obtain the static test accuracy rate and determining whether the static test accuracy rate reaches the accuracy rate threshold; based on the test data, performing static testing of the fault warning model on a big data platform to obtain the static test accuracy rate of the warning display and determining whether the static test accuracy rate of the warning display reaches the first accuracy rate threshold; and testing whether the fault warning model meets the refresh rate requirements based on the test data.

[0189] In one optional implementation, the fault warning model is a mathematical fault warning model, and the test module 705 is specifically used to: test whether the computation time of the mathematical fault warning model meets the requirements based on test data; and test whether the accuracy of the mathematical fault warning model in data retrieval reaches the second accuracy threshold based on test data.

[0190] In one implementation, the fault warning model is a mechanism fault warning model, the test is a control strategy logic diagram test, and the test module 705 is specifically used to: acquire warning test data, and determine whether the mechanism fault warning model can issue warning information normally based on the warning test data; and verify the judgment control logic of the mechanism fault warning model.

[0191] In one implementation, the fault warning model is a mechanism-based fault warning model, and the test is a control strategy logic diagram test. The test module 705 is specifically used for: acquiring warning test data and corresponding actual warning data; acquiring warning output data of the mathematical fault warning model based on the warning test data; comparing the warning output data with the actual warning data to obtain a comparison result; determining that the control strategy logic diagram test of the mathematical fault warning model has passed if the comparison result shows that the warning output data and the actual warning data are the same; or, optimizing the mathematical fault warning model based on the comparison result if the comparison result shows that the warning output data and the actual warning data are different.

[0192] In one implementation, the test is a joint test, and the test module 705 is specifically used for: verifying the function of the operator block based on the function type; constructing fault warning logic based on the fault warning model; in response to the function type of the operator block including function type and / or time type, deleting the function type operator block and / or time type operator block based on the fault warning logic, and obtaining a reference fault warning model; inputting the warning test data into the fault warning model to obtain the first warning information; inputting the warning test data into the reference fault warning model to obtain the second warning information; and determining whether the function type operator block and / or time type operator block are functioning normally based on the first warning information and the second warning information.

[0193] In one optional implementation, the test module 705 is specifically configured to: in response to the operator block's function type being time-based, verify whether the delay time function and / or pulse time function of the operator block are normal, based on the delay time and / or pulse time; or, in response to the operator block's function type being comparison-based, verify whether the comparison result of the operator block is correct based on a comparison threshold; or, in response to the operator block's function type being function-based, verify whether the function calculation result of the operator block is correct based on the function relationship and function value; or, in response to the operator block's function type being mathematical operation-based... The following methods can be used to verify the correctness of the operation result of the operator block based on the operation relationship; or, in response to the function type of the operator block being control type, to verify whether the operator block can control the device to be controlled based on the control strategy; or, in response to the function type of the operator block being device drive type, to verify whether the operator block can drive the device; or, in response to the function type of the operator block being sequential control type, to verify whether the operator block can achieve automated control based on the control logic; or, in response to the function type of the operator block being start / stop type, to verify whether the operator block can control the start / stop control device to start and / or stop based on start / stop control commands.

[0194] In one implementation, there are multiple fault warning models, and the device further includes: a generation module, a second acquisition module, a second processing module, a first judgment module, a third processing module, a fourth processing module, and a second judgment module. For an example, please refer to [link to example]. Figure 8 , Figure 8 This is a schematic diagram of another fault early warning model test device based on mathematical modeling and mechanism analysis provided in an embodiment of this application. Figure 8 As shown, the device 800 includes: a generation module 806 for generating a fault warning model library based on multiple fault warning models; a second acquisition module for acquiring test input data and test output data; a second processing module 807 for inputting the test input data into multiple fault warning models and acquiring the actual output data of the multiple fault warning models; a first judgment module 808 for judging whether the transfer of the fault warning model library is correct based on the actual output data and test output data; a third processing module 809 for sending a first input signal and a target fault warning model request signal to multiple fault warning models based on a big data platform to acquire the first output result of the fault warning model library; a fourth processing module 810 for sending a second input signal and a target fault warning model request signal to multiple fault warning models based on a big data platform to acquire the second output result of the fault warning model library; wherein the second input signal is a deviation signal of the first input signal; and a second judgment module 811 for judging whether the application of the fault warning model library is normal based on the first output result and / or the second output result.

[0195] The apparatus of this application embodiment can be used to deploy a fault warning model to a big data platform to test the fault warning model in a simulated actual operating environment, thereby ensuring that the fault warning model can accurately issue warnings.

[0196] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0197] Based on the embodiments of this application, this application also provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to execute the joint fault warning model testing method based on mathematical modeling and mechanism analysis of any of the foregoing embodiments.

[0198] Based on the embodiments of this application, this application also provides a computer-readable storage medium, wherein computer instructions are used to cause a computer to execute the joint fault early warning model testing method based on mathematical modeling and mechanism analysis according to any of the foregoing embodiments provided in this application.

[0199] Please see Figure 9 ,like Figure 9The diagram shown is a schematic block diagram of an example electronic device that can be used to implement embodiments of the present application. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present application described and / or claimed herein.

[0200] like Figure 9 As shown, device 900 includes a computing unit 901, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 902 or a computer program loaded from storage unit 908 into random access memory (RAM) 903. RAM 903 may also store various programs and data required for the operation of device 900. The computing unit 901, ROM 902, and RAM 903 are interconnected via bus 904. Input / output (I / O) interface 905 is also connected to bus 904.

[0201] Multiple components in device 900 are connected to I / O interface 905, including: input unit 906, such as keyboard, mouse, etc.; output unit 907, such as various types of monitors, speakers, etc.; storage unit 908, such as disk, optical disk, etc.; and communication unit 909, such as network card, modem, wireless transceiver, etc. Communication unit 909 allows device 900 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0202] The computing unit 901 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 901 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 901 performs the various methods and processes described above, such as a joint fault warning model testing method based on mathematical modeling and mechanistic analysis. For example, in some embodiments, the joint fault warning model testing method based on mathematical modeling and mechanistic analysis can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as storage unit 908. In some embodiments, part or all of the computer program can be loaded and / or installed on device 900 via ROM 902 and / or communication unit 909. When the computer program is loaded into RAM 903 and executed by computing unit 901, one or more steps of the joint fault warning model testing method based on mathematical modeling and mechanism analysis described above can be performed. Alternatively, in other embodiments, computing unit 901 can be configured to perform the joint fault warning model testing method based on mathematical modeling and mechanism analysis by any other suitable means (e.g., by means of firmware).

[0203] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard parts (ASSPs), systems-on-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0204] The program code used to implement the methods of this application may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0205] In the context of this application, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM) or flash memory, optical fiber, compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0206] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a cathode ray tube (CRT) or LCD (Liquid Crystal Display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0207] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), the Internet, and blockchain networks.

[0208] Computer systems can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is established by computer programs running on the respective computers and having a client-server relationship with each other. A server can be a cloud server, also known as a cloud computing server or cloud host, a hosting product within the cloud computing service system that addresses the shortcomings of traditional physical hosts and VPS (Virtual Private Server) services, such as high management difficulty and weak business scalability. Servers can also be servers for distributed systems or servers integrated with blockchain technology.

[0209] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this application can be achieved, and this is not limited herein.

[0210] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A joint fault early warning model testing method based on mathematical modeling and mechanism analysis, characterized in that, The method comprises: obtaining an early warning model graph of a fault early warning model; wherein the fault early warning model is a fault early warning model for early warning of a thermal power generation system; based on the early warning model graph, checking the functions and design of the fault early warning model; based on the early warning model graph and a big data platform, encapsulating the fault early warning model into an operator block of different function types to deploy the fault early warning model to the big data platform; based on the early warning model graph and the function types, setting the operator block; testing the fault early warning model; wherein the early warning model graph contains early warning logic and model structure of the fault early warning model; the fault early warning model comprises a mathematical fault early warning model and / or a mechanism fault early warning model; the function types comprise at least one of the following: time type, comparison type, function type, mathematical operation type, control type, device driving type, sequence control type and start-stop type; the testing comprises at least one of the following: online testing, control strategy logic graph testing and joint testing; wherein the mathematical fault early warning model is a fault early warning model that realizes online observation of an accident condition about to occur or being in progress through computer language and machine learning, and completes framework early warning of a system or device; the mechanism fault early warning model is a model that realizes analysis and judgment of an accident through setting of preset values, fusion of multiple judgment conditions and increase of preset conditions, thereby realizing early warning of a system or device, based on an operator block and a mathematical and mechanism model graph logic configuration; the testing is the joint testing, and the testing of the fault early warning model comprises: based on the function types, function verification of the operator block; based on the fault early warning model, construction of fault early warning logic; in response to the function types of the operator block comprising the function type and / or the time type, based on the fault early warning logic, deletion of the operator block of the function type and / or the operator block of the time type, and acquisition of a contrast fault early warning model; input of early warning test data into the fault early warning model to acquire first early warning information; input of the early warning test data into the contrast fault early warning model to acquire second early warning information; based on the first early warning information and the second early warning information, determination of whether the functions of the operator block of the function type and / or the operator block of the time type are normal.

2. The method of claim 1, wherein, the setting of the operator block based on the early warning model graph and the function types comprises: in response to the function types of the operator block being the time type, setting of a delay time and / or a pulse time of the operator block according to the early warning model graph; or in response to the function types of the operator block being the comparison type, setting of an input measuring point and a comparison category according to the early warning model graph, and setting of a comparison threshold of the operator block according to the input measuring point; or in response to the function types of the operator block being the function type, setting of a function relationship and a function value of the operator block according to the early warning model graph; or In response to the function type of the operator block being the mathematical operation type, setting an operation relationship of the operator block according to the early warning model graph; or, In response to the function type of the operator block being the control type, setting a control strategy and a corresponding to-be-controlled device of the operator block according to the early warning model graph; or, In response to the function type of the operator block being the device driving type, setting a driving device corresponding to the operator block according to the early warning model graph; or, In response to the function type of the operator block being the sequence control type, setting a sequence control logic of the operator block according to the early warning model graph; or, In response to the function type of the operator block being the start-stop type, setting a start-stop control instruction and a corresponding start-stop control device of the operator block according to the early warning model graph.

3. The method of claim 1, wherein, The test is the online test, and the testing of the fault early warning model comprises: obtaining a fault early warning information library; obtaining test data based on the fault early warning information library; performing static testing on the fault early warning model to obtain a static testing accuracy rate, and determining whether the static testing accuracy rate reaches an accuracy rate threshold; based on the test data, performing early warning display static testing on the mathematical fault early warning model on the big data platform to obtain an early warning display static testing accuracy rate, and determining whether the early warning display static testing accuracy rate reaches a first accuracy rate threshold; based on the test data, testing whether the fault early warning model meets a refresh rate requirement.

4. The method of claim 3, wherein, The fault early warning model is the mathematical fault early warning model, and the method further comprises: based on the test data, testing whether a calculation time of the mathematical fault early warning model meets a requirement; based on the test data, testing whether an accuracy rate of data retrieval of the mathematical fault early warning model reaches a second accuracy rate threshold.

5. The method of claim 1, wherein, The fault early warning model is the mechanism fault early warning model, and the test is the control strategy logic graph test, and the testing of the fault early warning model comprises: obtaining early warning test data and determining whether the mechanism fault early warning model can normally issue early warning information based on the early warning test data; verifying a judgment control logic of the mechanism fault early warning model.

6. The method of claim 1, wherein, The fault early warning model is the mechanism fault early warning model, and the test is the control strategy logic graph test, and the testing of the fault early warning model comprises: obtaining early warning test data and corresponding actual early warning data; based on the early warning test data, obtaining early warning output data of the mathematical fault early warning model; comparing the early warning output data with the actual early warning data to obtain a comparison result; in response to the comparison result being that the early warning output data is the same as the actual early warning data, determining that the mathematical fault early warning model control strategy logic graph test passes; or in response to the comparison result being that the early warning output data is not the same as the actual early warning data, optimizing the mathematical fault early warning model based on the comparison result.

7. The method of claim 1, wherein, The function verification of the operator block based on the function type comprises: In response to the function type of the operator block being the time type, verifying whether the time delay function and / or the pulse time function of the operator block is normal based on a time delay time and / or a pulse time; or, In response to the function type of the operator block being the comparison type, verifying whether the comparison result of the operator block is correct based on a comparison threshold; or, In response to the function type of the operator block being the function type, verifying whether the function calculation result of the operator block is correct based on a function relationship and a function value; or, In response to the function type of the operator block being the mathematical operation type, verifying whether the operation result of the operator block is correct based on an operation relationship; or, In response to the function type of the operator block being the control type, verifying whether the operator block can control a to-be-controlled device based on the control strategy; or, In response to the function type of the operator block being the device driving type, verifying whether the operator block can drive a driving device; Or, In response to the function type of the operator block being the sequential control type, verifying whether the operator block can realize automatic control based on control logic; or, In response to the function type of the operator block being the start-stop type, verifying whether the operator block can control the opening and / or closing of a start-stop control device based on a start-stop control instruction.

8. The method of claim 1, wherein, The method further comprises: Generating a fault early warning model library based on a plurality of fault early warning models; Obtaining fault early warning model library test data; Inputting the fault early warning model library test data into the fault early warning model library to obtain output data of the fault early warning model library; Detecting whether the output data includes model output of the plurality of fault early warning models, and determining whether the fault early warning model library transition is correct according to the detection result; Sending a first input signal and a target fault early warning model request signal to the fault early warning model library to obtain a first output result output by the fault early warning model library; Sending a second input signal to the fault early warning model library to obtain a second output result output by the fault early warning model library; wherein the second input signal is a deviation signal of the first input signal; Based on the first output result and / or the second output result, determining whether the application of the fault early warning model library is normal.

9. A device for testing a combined failure early-warning model based on mathematical modeling and mechanism analysis, characterized in that, The apparatus comprises: A first obtaining module for obtaining a warning model graph of a fault early warning model; wherein the fault early warning model is a model for early warning of a thermal power generation system; A checking module for checking the function and design of the fault early warning model based on the warning model graph; A deployment module for encapsulating the fault early warning model as an operator block of different function types based on the warning model graph and a big data platform, so as to deploy the fault early warning model to the big data platform; A processing module for setting the operator block based on the warning model graph and the function types; and A processing module for setting the operator block based on the warning model graph and the function types. The test module is configured to test the fault early warning model; wherein the early warning model graph comprises early warning logic and a model structure of the fault early warning model; the fault early warning model comprises a mathematical fault early warning model and / or a mechanism fault early warning model; the function type comprises at least one of a time type, a comparison type, a function type, a mathematical operation type, a control type, a device driving type, a sequence control type, and a start-stop type; the test comprises at least one of online testing, control strategy logic graph testing, and joint testing; wherein the mathematical fault early warning model is a model for realizing online observation of an accident condition about to occur or being in progress and completing framework early warning of a system or a device in the form of a computer language and machine learning; the mechanism fault early warning model is a model for realizing early warning of a system or a device by setting a preset value, fusing multiple judgment conditions, and increasing preset conditions based on an operator block and a mathematical and mechanism model graph logic configuration. The test is the joint test, and the test module is specifically configured to: perform function verification on the operator block based on the function type; construct fault early warning logic based on the fault early warning model; in response to the function type of the operator block comprising the function type and / or the time type, delete the operator block of the function type and / or the operator block of the time type based on the fault early warning logic, and obtain a contrast fault early warning model; input early warning test data into the fault early warning model to obtain first early warning information; input the early warning test data into the contrast fault early warning model to obtain second early warning information; determine whether the function of the operator block of the function type and / or the operator block of the time type is normal based on the first early warning information and the second early warning information.

10. An electronic device, comprising: comprise: at least one processor; and a memory connected in communication with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the joint fault early warning model testing method based on mathematical modeling and mechanism analysis according to any one of claims 1 to 8.

11. A computer readable storage medium storing instructions, wherein the instructions, when executed by a processor, cause the processor to perform operations comprising: When the instructions are executed, the method according to any one of claims 1 to 8 is implemented.

Images