A twin-plant-based hydropower plant equipment monitoring and early warning method and system

Abstract

The present invention discloses a method and system for monitoring and early warning of hydropower plant equipment based on a twin factory, which belongs to the field of digital twins, wherein the method includes: performing digital twin modeling on the equipment of the hydropower plant to obtain a model library of the hydropower plant equipment; acquiring characteristic operation equipment by collecting the real-time status of the equipment of the hydropower plant; calling the first equipment twin model of the characteristic operation equipment; collecting the current operation data of the characteristic operation equipment to obtain the second equipment twin model; performing model differentiation testing and outputting differentiation indicators; when the differentiation indicator is not within the preset differentiation indicator threshold, obtaining the first early warning information and issuing an early warning on the current maintenance quality. This application solves the technical problem that the maintenance quality of hydropower plant equipment cannot be accurately evaluated and warned in real time in the prior art, and achieves the technical effect of realizing real-time monitoring and multi-level early warning of the maintenance quality of hydropower plant equipment through differential analysis of the digital twin model.

Description

Technical Field

[0001] The present invention relates to the field of digital twins, and in particular to a method and system for monitoring and early warning of hydropower plant equipment based on twin factories. Background Art

[0002] With the increasing operation life and maintenance of hydropower plants, equipment deviations will occur before and after maintenance. If not discovered in time, it will lead to misjudgment of the equipment during actual operation, resulting in inaccurate equipment control. The existing hydropower plant equipment maintenance quality monitoring and early warning mainly relies on technicians to judge the maintenance quality based on experience, and cannot quantitatively evaluate the maintenance quality and accurately warn, affecting the smooth operation of the hydropower plant. Summary of the invention

[0003] This application aims to solve the technical problem that the existing technology cannot accurately evaluate and warn the maintenance quality of hydropower plant equipment in real time by providing a hydropower plant equipment monitoring and early warning method and system based on twin plants.

[0004] In view of the above problems, the present application provides a hydropower plant equipment monitoring and early warning method and system based on twin plants.

[0005] The first aspect disclosed in the present application provides a hydropower plant equipment monitoring and early warning method based on a twin factory, the method comprising: obtaining a hydropower plant equipment model library by performing digital twin modeling on the equipment of the hydropower plant; obtaining characteristic operating equipment by collecting the real-time status of the equipment in the hydropower plant, wherein the characteristic operating equipment is the operating equipment after maintenance in the hydropower plant; calling the first equipment twin model of the characteristic operating equipment according to the hydropower plant equipment model library, wherein the first equipment twin model is the digital twin simulation result of the characteristic operating equipment before maintenance; obtaining the second equipment twin model of the characteristic operating equipment by collecting the current operating data of the characteristic operating equipment, wherein the second equipment twin model is the digital twin simulation result of the characteristic operating equipment after maintenance ; Perform model testing based on the first device twin model and the second device twin model, including differentiation testing of random independent partition models and differentiation testing of random continuous partition models, and output differentiation indicators, wherein the random independent partition model includes a model of an independent operating device randomly selected from the first device twin model, and a model of the corresponding independent device in the second device twin model, and the random continuous partition model is a model of at least two continuous devices with a continuous collaborative operating relationship randomly selected from the first device twin model, and a model of the corresponding continuous device in the second device twin model; when the differentiation indicator is not within the preset differentiation indicator threshold, obtain a first warning information, and issue a warning on the current maintenance quality according to the first warning information.

[0006] Another aspect disclosed in the present application provides a hydropower plant equipment monitoring and early warning system based on a twin factory, the system comprising: an equipment twin modeling module, used to obtain a hydropower plant equipment model library by performing digital twin modeling on the equipment of the hydropower plant; a characteristic operation equipment module, used to obtain characteristic operation equipment by collecting real-time status of the equipment in the hydropower plant, wherein the characteristic operation equipment is the operation equipment after maintenance in the hydropower plant; a twin model retrieval module, used to call the first equipment twin model of the characteristic operation equipment according to the hydropower plant equipment model library, wherein the first equipment twin model is the digital twin simulation result of the characteristic operation equipment before maintenance; a maintenance twin model module, used to obtain the second equipment twin model of the characteristic operation equipment by collecting the current operation data of the characteristic operation equipment, wherein the second equipment twin model is the digital twin simulation result of the characteristic operation equipment after maintenance The digital twin simulation results of the equipment are output; a differentiation test module is used to perform model testing according to the first equipment twin model and the second equipment twin model, including differentiation testing of random independent partition models and differentiation testing of random continuous partition models, and output differentiation indicators, wherein the random independent partition model includes a model of an independent operating equipment randomly selected from the first equipment twin model, and a model of the corresponding independent equipment in the second equipment twin model, and the random continuous partition model is a model of at least two continuous equipment with a continuous collaborative operating relationship randomly selected from the first equipment twin model, and a model of the corresponding continuous equipment in the second equipment twin model; a maintenance quality warning module is used to obtain a first warning information when the differentiation indicator is not within a preset differentiation indicator threshold, and to warn the current maintenance quality according to the first warning information.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] By conducting digital twin modeling on the equipment of the hydropower plant, a hydropower plant equipment model library is obtained to obtain a digital twin model reference for the equipment in a normal state for differential analysis of subsequent models; by collecting the real-time status of the equipment in the hydropower plant, characteristic operating equipment is obtained, wherein the characteristic operating equipment is the operating equipment after maintenance in the hydropower plant, which is taken as the research object; the first equipment twin model of the characteristic operating equipment is called according to the hydropower plant equipment model library to obtain a digital twin model of the equipment in a normal operating state before maintenance; the second equipment twin model of the characteristic operating equipment is obtained by collecting the current operating data of the characteristic operating equipment to obtain a digital twin model of the equipment in the operating state after maintenance; based on Model testing is performed on the first device twin model and the second device twin model, including differentiation testing of random independent partition models and differentiation testing of random continuous partition models, and differentiation indicators are output to quantify the difference in operating status before and after equipment maintenance. When the differentiation indicator is not within the preset differentiation indicator threshold, the first warning information is obtained, and the current maintenance quality is warned according to the first warning information. The technical solution of real-time monitoring and early warning of maintenance quality solves the technical problem that the prior art cannot accurately evaluate and warn the maintenance quality of hydropower plant equipment in real time, and achieves the technical effect of real-time monitoring and multi-level early warning of the maintenance quality of hydropower plant equipment through differentiation analysis of digital twin models.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A flow chart of a method for monitoring and early warning of hydropower plant equipment based on twin plants is provided for the embodiment of the present application;

[0011] Figure 2 A flow chart of obtaining differentiated indicators in a method for monitoring and early warning of hydropower plant equipment based on twin plants is provided for an embodiment of the present application;

[0012] Figure 3 A structural schematic diagram of a hydropower plant equipment monitoring and early warning system based on twin factories is provided for an embodiment of the present application.

[0013] Explanation of the accompanying drawings: equipment twin modeling module 11, characteristic operation equipment module 12, twin model retrieval module 13, maintenance twin model module 14, differentiated testing module 15, maintenance quality early warning module 16. DETAILED DESCRIPTION

[0014] The overall idea of ​​the technical solution provided by this application is as follows:

[0015] The embodiment of the present application provides a method and system for monitoring and early warning of hydropower plant equipment based on a twin factory. First, the digital twin technology is used to model the hydropower plant equipment to obtain a first equipment twin model in a normal state, and the characteristic operating equipment after maintenance is remodeled to obtain a second equipment twin model, thereby realizing the acquisition of a digital twin model in two states before and after equipment maintenance. Then, by performing a differentiated comparison between the first equipment twin model and the second equipment twin model, a differentiated index is extracted. If the differentiated index is not within the preset differentiated index threshold, the first early warning information is obtained, and the maintenance quality of the current characteristic operating equipment is warned according to the first early warning information, so as to realize accurate monitoring and early warning of the maintenance quality of hydropower plant equipment.

[0016] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

[0017] Embodiment 1

[0018] like Figure 1 As shown, the embodiment of the present application provides a hydropower plant equipment monitoring and early warning method based on twin plants, the method comprising:

[0019] By conducting digital twin modeling on the equipment in the hydropower plant, a model library of the hydropower plant equipment is obtained;

[0020] In the embodiment of the present application, first, according to the operating conditions and monitoring requirements of the hydropower plant, the equipment for which the digital twin model is to be established is determined, including hydro-turbine generator sets, speed control systems, switchgear, hydraulic regulation systems, hydraulic mechanical equipment, etc. Then, temperature sensors, pressure sensors, vibration sensors, etc. are installed on the selected hydropower plant equipment to collect real-time data under normal operating conditions of the equipment, record input parameters, output parameters and various state parameters, and obtain original equipment data. Subsequently, the original equipment data is pre-processed by marking, cleaning, denoising, synchronization, etc., and a structured data set is generated for modeling. Subsequently, based on the pre-processed equipment data, a physical model, a data-driven model, or a hybrid modeling method combining the two is used to construct a digital twin model of equipment such as a hydropower unit and a hydraulic regulation system, and obtain a hydropower plant equipment model library.

[0021] Through digital twin modeling, a hydropower plant equipment model library covering all types of equipment in the hydropower plant is obtained. The equipment twin models contained in it reflect the normal operating status and performance parameters of the corresponding equipment, providing a basis for subsequent equipment monitoring and early warning analysis.

[0022] By collecting the real-time status of the equipment in the hydropower plant, characteristic operating equipment is obtained, wherein the characteristic operating equipment is the operating equipment after maintenance in the hydropower plant;

[0023] In the embodiment of the present application, first, the real-time status data such as temperature and vibration collected by the sensor is transmitted to the power plant management system database through dedicated line access to obtain the real-time status of the equipment in the hydropower plant. Then, the power plant management system database of the hydropower plant is connected to obtain the information of the equipment that has been recently repaired, including the equipment number, maintenance time, etc. Then, for the equipment that has been recently repaired, it is confirmed that its maintenance work has been completed and the equipment has been restored to normal production operation. These devices are characteristic operation equipment.

[0024] Calling a first equipment twin model of the characteristic operation equipment according to the hydropower plant equipment model library, wherein the first equipment twin model is a digital twin simulation result of the characteristic operation equipment before maintenance;

[0025] In the embodiment of the present application, a hydropower plant equipment model library is deployed on a cloud server, and a unique equipment code is established as a retrieval index. After determining the characteristic operation equipment, the corresponding digital twin model is found in the hydropower plant equipment model library through the code of the characteristic operation equipment, and the first equipment twin model is obtained. The digital twin model is a twin model of the characteristic operation equipment in a normal operation state before maintenance, providing data support for subsequent model comparison and analysis.

[0026] By collecting current operation data of the characteristic operation device, a second device twin model of the characteristic operation device is obtained, wherein the second device twin model is a digital twin simulation result of the characteristic operation device after maintenance;

[0027] In the embodiment of the present application, after obtaining the first device twin model of the characteristic operation equipment (i.e., the digital twin simulation result before maintenance), the second device twin model is further obtained, i.e., the digital twin simulation result of the characteristic operation equipment after maintenance. Specifically, first, the state monitoring data of the characteristic operation equipment after maintenance is obtained in the power plant management system database; then, the state monitoring data of the characteristic operation equipment after maintenance is preprocessed by marking, cleaning, denoising, synchronizing, etc., to generate a structured data set for modeling; then, based on the preprocessed state monitoring data, the second device twin model of the characteristic operation equipment is constructed in the same way as the first device twin model, laying the foundation for evaluating the equipment maintenance quality.

[0028] Perform model testing according to the first device twin model and the second device twin model, including differentiation testing of random independent partition models and differentiation testing of random continuous partition models, and output differentiation indicators, wherein the random independent partition model includes a model of an independent operating device randomly selected from the first device twin model, and a model of the corresponding independent device in the second device twin model, and the random continuous partition model is a model of at least two continuous devices with a continuous collaborative operating relationship randomly selected from the first device twin model, and a model of the corresponding continuous device in the second device twin model;

[0029] In an embodiment of the present application, the model test includes a differentiation test of a random independent partition model and a differentiation test of a random continuous partition model to output a differentiation index. Among them, the random independent partition model includes a model of an independent operating device randomly selected from the first device twin model, and a model of a corresponding independent device selected from the second device twin model, and there is no continuous collaborative working relationship between these independent devices. Then the selected device digital twin model is subjected to a differentiation test to compare the differences between the two models before and after maintenance. The test content may include differences in equipment parameters, status data, operating data, output data, etc., so as to obtain the differentiation index of the random independent partition model. Through the differentiation test of the independent operating equipment model, the impact of the maintenance quality of a single device can be evaluated.

[0030] The random continuous partition model includes randomly selecting models of at least two continuous devices that have a continuous collaborative working relationship from the first device twin model, and selecting models of corresponding continuous devices from the second device twin model, and there is a continuous collaborative working relationship between these continuous devices. Then, the selected device digital twin model group is subjected to a differentiation test to compare the differences between the two model groups before and after maintenance. The test content can also include differences in parameters, status, operating data, and output data, so as to obtain differentiation indicators of the random continuous partition model. Through the differentiation test of the continuous device group, the maintenance quality of the collaborative device group and the impact on the collaborative performance of the device group can be evaluated.

[0031] By performing differentiation tests on the random independent partition model and the random continuous partition model, the difference between the first device twin model and the second device twin model can be comprehensively evaluated, and differentiation indicators can be output for evaluation. Among them, independent devices can reflect the maintenance quality of a single device, and continuous devices can reflect the maintenance quality of a collaborative device group.

[0032] Further, such as Figure 2 As shown, the embodiment of the present application includes:

[0033] Establishing a differentiated identification network layer, wherein the differentiated identification network layer includes a digital twin network, and the digital twin network includes a first data input channel and a second data input channel;

[0034] Connecting the first data input channel to the first device twin model to receive a device performance test data set from the first device twin model;

[0035] Connecting the second data input channel to the second device twin model to receive a device performance test data set from the second device twin model;

[0036] A similarity analysis is performed based on the first data input channel and the second data input channel to obtain a differentiation index.

[0037] Furthermore, the embodiments of the present application include:

[0038] According to the first data input channel and the second data input channel, the energy consumption change characteristics of the equipment before and after the maintenance of the characteristic operation equipment, the equipment operation sensitivity characteristics, the equipment recovery rate characteristics and the equipment start-stop power characteristics are obtained;

[0039] A similarity analysis is performed based on the equipment energy consumption change characteristics, equipment operation sensitivity characteristics, equipment recovery rate characteristics, and equipment start-stop power characteristics to obtain differentiation indicators.

[0040] In a preferred embodiment, first, the basic framework of the digital twin network is constructed, and structures such as the industrial Internet are selected to realize the interconnection of device models. Two data interfaces are created in the digital twin network, namely the first input channel and the second input channel. Among them, the connection object of the first input channel is the first device twin model; the connection object of the second input channel is the second device twin model. Then, the constructed digital twin network is embedded in the differential identification network layer to realize the construction of the differential identification network layer. Subsequently, an adaptation connection is established between the first data input channel and the output interface of the first device twin model through wireless communication. Once the first device twin model runs the simulation, the performance data set output by it, which contains time series data such as device temperature and vibration, can be directly imported into the differential identification network through the first data input channel. At the same time, an adaptation connection is established between the second data input channel and the interface of the second device twin model through wireless communication. Once the second device twin model is simulated, the result of its output containing the latest device performance data set can enter the differential identification network through the second data input channel. At this point, the digital twin model of the characteristic operating equipment in two states before and after maintenance has been realized, and the interfaces are connected with the corresponding data input channels respectively, laying the foundation for the status comparison of the equipment before and after maintenance.

[0041] Then, an energy consumption feature extraction module is integrated in the differential identification network, connected to the first input channel and the second input channel, and the energy consumption feature extraction module is used to extract energy consumption data from the simulation results of the first device twin model and the second device twin model, respectively, to obtain the energy consumption change characteristics of the equipment before and after the maintenance of the characteristic operation equipment. At the same time, a fault response analysis module is integrated in the differential identification network, the fault model is input into the first device twin model and the second device twin model, the output fluctuation is detected, the fluctuation amplitude and stabilization time of the two device twin models are identified, the dynamic sensitivity score of the equipment operation is generated, and the equipment operation sensitivity characteristics before and after the maintenance of the characteristic operation equipment are obtained. At the same time, a fault recovery evaluation unit is integrated in the differential identification network, the fault signal is input into the first device twin model and the second device twin model, the respective fault duration and the time period of system stable recovery are identified, and the equipment recovery rate characteristics before and after the maintenance of the characteristic operation equipment are obtained. At the same time, a steady-state evaluation module is integrated in the differential identification network, the startup and shutdown simulation of the equipment is responded to in the first device twin model and the second device twin model, the excessive current jitter in each process is detected, the corresponding index is output, and the equipment start-stop power characteristics before and after the maintenance of the characteristic operation equipment are obtained. Subsequently, the equipment energy consumption change characteristics, equipment operation sensitivity characteristics, equipment recovery rate characteristics, and equipment start-stop power characteristics before and after the maintenance of the characteristic operation equipment are extracted at the same time, and each characteristic index is normalized and mapped to the range of 0-1. Then, the Euclidean distance calculation formula is used to calculate the distance between the normalized equipment energy consumption change characteristics, equipment operation sensitivity characteristics, equipment recovery rate characteristics, and equipment start-stop power characteristics before and after maintenance, and the Euclidean distance value is determined. After that, the Euclidean distance value of each characteristic index is obtained to obtain the average Euclidean distance value as a differentiation index. Among them, the smaller the average Euclidean distance value, the smaller the differentiation of the equipment energy consumption change characteristics, equipment operation sensitivity characteristics, equipment recovery rate characteristics, and equipment start-stop power characteristics before and after maintenance.

[0042] When the differentiation index is not within the preset differentiation index threshold, first warning information is obtained, and a warning is issued on the current maintenance quality according to the first warning information.

[0043] In an embodiment of the present application, first, query the historical maintenance records, determine the normal range of changes in the characteristic operation equipment before and after maintenance, determine the upper and lower limits of the differentiation score, and obtain the preset differentiation index threshold. Then, compare the obtained differentiation index with the preset differentiation index threshold. When the differentiation index is higher than the upper limit of the preset differentiation index threshold or lower than the lower limit of the preset differentiation index threshold, the differentiation index is not within the preset differentiation index threshold, indicating that the equipment maintenance has a greater impact on its operating parameters and there are greater risks. The first warning information is automatically obtained. The warning information is pushed to the monitoring personnel or the maintenance management platform in the format of SMS or email, and timely warns of the maintenance quality problems of the current characteristic operation equipment. Once the maintenance personnel receive the first warning information, they will immediately review or remediate the characteristic operation equipment to avoid accidents and improve the safety of the equipment.

[0044] Furthermore, the embodiments of the present application include:

[0045] Connect to the equipment maintenance management system to obtain the historical maintenance times of the characteristic operation equipment;

[0046] If the historical maintenance times of the characteristic operation equipment are returned as null, calling the first equipment twin model of the characteristic operation equipment according to the hydropower plant equipment model library;

[0047] Obtaining a third device twin model of the feature operation device, wherein the third device twin model is a digital twin simulation result when the feature operation device performs abnormally before maintenance;

[0048] The first device twin model and the third device twin model are used to obtain abnormality levels, and the warning information is managed in a graded manner according to the abnormality levels.

[0049] In a preferred embodiment, first, the corresponding application program interface is called to connect to the equipment maintenance management system, which contains complete equipment maintenance records, including equipment code, maintenance times, maintenance time, maintenance personnel, maintenance reasons and other information, so as to obtain a complete historical maintenance record of a specific operating device, and extract the historical maintenance times from it. If no maintenance record of the characteristic operating device is found in the equipment maintenance management system, its historical maintenance times are returned as null. At this time, according to the pre-built hydropower plant equipment model library, the first digital twin model of the characteristic operating device when it is working normally is called. At the same time, the abnormal monitoring data when the characteristic operating device is abnormal before maintenance is obtained in the power plant management system database, and the abnormal monitoring data is pre-processed such as marking, cleaning, and denoising. Based on the pre-processed abnormal monitoring data, the third equipment twin model of the characteristic operating device is constructed in the same way as the first equipment twin model, laying the foundation for evaluating the abnormal level of the equipment. Subsequently, the first equipment twin model and the third equipment twin model are respectively run to obtain the output data sets of the two models, calculate the difference between the indicators in the output data sets, determine the abnormal amplitude, and then determine the matching abnormal level according to the difference in the magnitude of the abnormal amplitude, such as the first warning, the second warning, etc. Then, according to the abnormality level, the corresponding warning content is called from the warning template library to generate information, and according to the severity of the abnormality level, the appropriate distribution channel is selected to push the warning information, so as to achieve timely detection and matching of abnormal status of equipment, promote graded warning, and refine the monitoring and early warning management of equipment.

[0050] Furthermore, the embodiments of the present application include:

[0051] If the historical maintenance times of the characteristic operation equipment is not empty, obtain n differentiation indicators, where n is the return value of the historical maintenance times;

[0052] generating, based on the n differentiation indicators, a performance degradation curve of the operating equipment based on the maintenance identification feature;

[0053] The performance degradation curve of the equipment running based on the maintenance identification feature is subjected to degradation rate detection, and when the degradation rate is greater than or equal to a preset degradation rate, second warning information is obtained, and a warning is issued on the continuous maintenance quality according to the second warning information.

[0054] Furthermore, n differentiation indicators are obtained, including:

[0055] According to the historical maintenance times of the characteristic operation equipment, a differentiation index between two adjacent maintenance times is calculated to obtain a first differentiation index, a second differentiation index, ... an nth differentiation index;

[0056] Among them, the first differentiation index is the differentiated recognition result of the first device twin model and the second device twin model under the first maintenance number, the second differentiation index is the differentiated recognition result of the first device twin model and the second device twin model under the second maintenance number,…, the nth differentiation index is the differentiated recognition result of the first device twin model and the second device twin model under the nth maintenance number;

[0057] The first differentiation index, the second differentiation index, ... the nth differentiation index are integrated to obtain n differentiation indexes.

[0058] In a preferred embodiment, after obtaining the historical maintenance times of the characteristic operation equipment, if the returned historical maintenance times are not empty, a corresponding number of differentiation indicators are obtained according to the historical maintenance times. Assume that the historical maintenance times of the characteristic operation equipment are n times, traverse the historical maintenance times, extract the differentiation indicators of each historical maintenance, and obtain the first differentiation indicator, the second differentiation indicator, ... the nth differentiation indicator. Among them, the first differentiation indicator is the differentiated recognition result of the first device twin model and the second device twin model under the first maintenance times, the second differentiation indicator is the differentiated recognition result of the first device twin model and the second device twin model under the second maintenance times, ..., and the nth differentiation indicator is the differentiated recognition result of the first device twin model and the second device twin model under the nth maintenance times. Then, the first differentiation indicator, the second differentiation indicator, ... the nth differentiation indicator are integrated to obtain n differentiation indicators.

[0059] After that, each maintenance time of the characteristic operation equipment is taken as a time node. At each time node, the differential indicators corresponding to the maintenance are mapped to obtain each time node and the differential indicators, and the discrete differential indicator data of the characteristic operation equipment is obtained. Subsequently, the regression analysis method is used to seek the function model with the best curve fit for the discrete differential indicator data. After multiple iterations, the optimal solution is determined to obtain the regression equation, and the regression equation is plotted to reflect the overall development trend of the equipment performance gradually decreasing with the number of maintenance times, and the performance degradation curve is obtained. Then, based on the collected equipment failure records and life data, the expert group pre-sets the threshold of the performance attenuation rate to obtain the preset decline rate. Next, the descending gradient of the performance degradation curve is detected to obtain the curve decline rate. When the curve decline rate is greater than or equal to the preset decline rate, the second warning information is automatically triggered and obtained, and the second warning information is pushed in the format of SMS and email. The relevant maintenance personnel warn the characteristic operation equipment of the continuous maintenance quality according to the second warning information to prompt that the performance of the equipment has declined rapidly after multiple repairs to prevent greater harm.

[0060] Furthermore, the embodiments of the present application include:

[0061] Obtaining a first equipment twin model of the nth maintenance, wherein the first equipment twin model of the nth maintenance is a digital twin simulation result of the characteristic operation equipment before the nth maintenance;

[0062] Determine an operation data set corresponding to the characteristic operation equipment after the nth maintenance;

[0063] Perform digital twin simulation on the operating data set corresponding to the nth maintenance of the characteristic operating equipment to determine the twin model of the second equipment for the nth maintenance;

[0064] Model differentiation monitoring is performed based on the first equipment twin model of the nth maintenance and the second equipment twin model of the nth maintenance to obtain the nth differentiation index.

[0065] In a preferred embodiment, when obtaining each differentiation indicator, first, obtain the first equipment twin model of the nth maintenance. The model is constructed based on the actual operation data of the equipment before the nth maintenance, and reflects the various performance indexes of the equipment in the normal working state before the nth maintenance. Then, determine the exact time node of the nth maintenance, query the continuous operation data after this time point, and obtain the corresponding operation data set after the nth maintenance, which truly reflects the latest performance and working condition level of the equipment after the nth maintenance is completed. Subsequently, import the operation data set after the completion of the nth maintenance into the digital twin platform, configure the parameter settings, establish a simulation model, and form the second digital twin model after the nth maintenance. Next, import multiple groups of simulation output results of the first twin model and the second twin model of the nth maintenance into the differentiation recognition network layer, perform model differentiation monitoring, and obtain the nth differentiation indicator.

[0066] In summary, the hydropower plant equipment monitoring and early warning method based on twin factories provided in the embodiment of the present application has the following technical effects:

[0067] By performing digital twin modeling on the equipment of the hydropower plant, a hydropower plant equipment model library is obtained, and a digital twin model of the equipment in a normal state is obtained as a reference for subsequent difference analysis. By collecting the real-time status of the equipment in the hydropower plant, characteristic operating equipment is obtained, wherein the characteristic operating equipment is the operating equipment after maintenance in the hydropower plant, and the equipment maintenance research object is determined. The first equipment twin model of the characteristic operating equipment is called according to the hydropower plant equipment model library to obtain a digital twin model in a normal state before maintenance. By collecting the current operating data of the characteristic operating equipment, the second equipment twin model of the characteristic operating equipment is obtained, and the digital twin model of the state after maintenance is obtained. Model differentiation identification is performed according to the first equipment twin model and the second equipment twin model, and differentiation indicators are output to evaluate and quantify the difference in equipment status before and after maintenance. When the differentiation indicator is not within the preset differentiation indicator threshold, the first warning information is obtained, and the current maintenance quality is warned according to the first warning information, so as to realize accurate monitoring and warning of the maintenance quality.

[0068] Embodiment 2

[0069] Based on the same inventive concept as the hydropower plant equipment monitoring and early warning method based on twin plants in the aforementioned embodiment, Figure 3 As shown, the embodiment of the present application provides a hydropower plant equipment monitoring and early warning system based on twin plants, the system comprising:

[0070] The equipment twin modeling module 11 is used to obtain a model library of the hydropower plant equipment by performing digital twin modeling on the equipment of the hydropower plant;

[0071] The characteristic operation equipment module 12 is used to obtain characteristic operation equipment by collecting the real-time status of the equipment in the hydropower plant, wherein the characteristic operation equipment is the operation equipment after maintenance in the hydropower plant;

[0072] A twin model retrieval module 13 is used to call the first equipment twin model of the characteristic operation equipment according to the hydropower plant equipment model library, wherein the first equipment twin model is a digital twin simulation result of the characteristic operation equipment before maintenance;

[0073] A maintenance twin model module 14 is used to obtain a second device twin model of the characteristic operation device by collecting current operation data of the characteristic operation device, wherein the second device twin model is a digital twin simulation result of the characteristic operation device after maintenance;

[0074] A differentiation test module 15 is used to perform model testing according to the first device twin model and the second device twin model, including differentiation testing of a random independent partition model and differentiation testing of a random continuous partition model, and output differentiation indicators, wherein the random independent partition model includes a model of an independent operating device randomly selected from the first device twin model, and a model of the corresponding independent device in the second device twin model, and the random continuous partition model is a model of at least two continuous devices with a continuous collaborative operating relationship randomly selected from the first device twin model, and a model of the corresponding continuous device in the second device twin model;

[0075] The maintenance quality early warning module 16 is used to obtain first early warning information when the differentiation index is not within a preset differentiation index threshold, and to issue an early warning on the current maintenance quality according to the first early warning information.

[0076] Furthermore, the embodiment of the present application also includes a device hierarchical management module, which includes the following execution steps:

[0077] Connect to the equipment maintenance management system to obtain the historical maintenance times of the characteristic operation equipment;

[0078] If the historical maintenance times of the characteristic operation equipment are returned as null, calling the first equipment twin model of the characteristic operation equipment according to the hydropower plant equipment model library;

[0079] Obtaining a third device twin model of the feature operation device, wherein the third device twin model is a digital twin simulation result when the feature operation device performs abnormally before maintenance;

[0080] The first device twin model and the third device twin model are used to obtain abnormality levels, and the warning information is managed in a graded manner according to the abnormality levels.

[0081] Furthermore, the device hierarchical management module further includes the following execution steps:

[0082] If the historical maintenance times of the characteristic operation equipment is not empty, obtain n differentiation indicators, where n is the return value of the historical maintenance times;

[0083] generating, based on the n differentiation indicators, a performance degradation curve of the operating equipment based on the maintenance identification feature;

[0084] The performance degradation curve of the equipment running based on the maintenance identification feature is subjected to degradation rate detection, and when the degradation rate is greater than or equal to a preset degradation rate, second warning information is obtained, and a warning is issued on the continuous maintenance quality according to the second warning information.

[0085] Furthermore, the device hierarchical management module further includes the following execution steps:

[0086] According to the historical maintenance times of the characteristic operation equipment, a differentiation index between two adjacent maintenance times is calculated to obtain a first differentiation index, a second differentiation index, ... an nth differentiation index;

[0087] Among them, the first differentiation index is the differentiated recognition result of the first device twin model and the second device twin model under the first maintenance number, the second differentiation index is the differentiated recognition result of the first device twin model and the second device twin model under the second maintenance number,…, the nth differentiation index is the differentiated recognition result of the first device twin model and the second device twin model under the nth maintenance number;

[0088] The first differentiation index, the second differentiation index, ... the nth differentiation index are integrated to obtain n differentiation indexes.

[0089] Furthermore, the device hierarchical management module further includes the following execution steps:

[0090] Obtaining a first equipment twin model of the nth maintenance, wherein the first equipment twin model of the nth maintenance is a digital twin simulation result of the characteristic operation equipment before the nth maintenance;

[0091] Determine an operation data set corresponding to the characteristic operation equipment after the nth maintenance;

[0092] Perform digital twin simulation on the operating data set corresponding to the nth maintenance of the characteristic operating equipment to determine the twin model of the second equipment for the nth maintenance;

[0093] Model differentiation monitoring is performed based on the first equipment twin model of the nth maintenance and the second equipment twin model of the nth maintenance to obtain the nth differentiation index.

[0094] Furthermore, the differential testing module 15 includes the following execution steps:

[0095] Establishing a differentiated identification network layer, wherein the differentiated identification network layer includes a digital twin network, and the digital twin network includes a first data input channel and a second data input channel;

[0096] Connecting the first data input channel to the first device twin model to receive a device performance test data set from the first device twin model;

[0097] Connecting the second data input channel to the second device twin model to receive a device performance test data set from the second device twin model;

[0098] A similarity analysis is performed based on the first data input channel and the second data input channel to obtain a differentiation index.

[0099] Furthermore, the differential testing module 15 also includes the following execution steps:

[0100] According to the first data input channel and the second data input channel, the energy consumption change characteristics of the equipment before and after the maintenance of the characteristic operation equipment, the equipment operation sensitivity characteristics, the equipment recovery rate characteristics and the equipment start-stop power characteristics are obtained;

[0101] A similarity analysis is performed based on the equipment energy consumption change characteristics, equipment operation sensitivity characteristics, equipment recovery rate characteristics, and equipment start-stop power characteristics to obtain differentiation indicators.

[0102] Any step of the method described above can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor to implement any method in the embodiments of the present application, without any unnecessary restrictions.

[0103] Furthermore, the first or second mentioned above may not only represent an order relationship, but may also represent a specific concept, and / or refer to multiple elements that can be selected individually or in full. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A hydropower plant equipment monitoring and early warning method based on twin plants, characterized in that: The method comprises: By conducting digital twin modeling on the equipment in the hydropower plant, a model library of the hydropower plant equipment is obtained; By collecting the real-time status of the equipment in the hydropower plant, characteristic operating equipment is obtained, wherein the characteristic operating equipment is the operating equipment after maintenance in the hydropower plant; Calling a first equipment twin model of the characteristic operation equipment according to the hydropower plant equipment model library, wherein the first equipment twin model is a digital twin simulation result of the characteristic operation equipment before maintenance; By collecting current operation data of the characteristic operation device, a second device twin model of the characteristic operation device is obtained, wherein the second device twin model is a digital twin simulation result of the characteristic operation device after maintenance; Perform model testing according to the first device twin model and the second device twin model, including differentiation testing of random independent partition models and differentiation testing of random continuous partition models, and output differentiation indicators, wherein the random independent partition model includes a model of an independent operating device randomly selected from the first device twin model, and a model of the corresponding independent device in the second device twin model, and the random continuous partition model is a model of at least two continuous devices with a continuous collaborative operating relationship randomly selected from the first device twin model, and a model of the corresponding continuous device in the second device twin model; When the differentiation index is not within the preset differentiation index threshold, obtaining first warning information, and issuing a warning on the current maintenance quality according to the first warning information; Connect to the equipment maintenance management system to obtain the historical maintenance times of the characteristic operation equipment; If the historical maintenance times of the characteristic operation equipment are returned as null, calling the first equipment twin model of the characteristic operation equipment according to the hydropower plant equipment model library; Obtaining a third device twin model of the feature operation device, wherein the third device twin model is a digital twin simulation result when the feature operation device performs abnormally before maintenance; The first device twin model and the third device twin model are used to obtain abnormality levels, and the warning information is managed in a graded manner according to the abnormality levels.

2. The method according to claim 1, characterized in that The method further comprises: If the historical maintenance times of the characteristic operation equipment is not empty, obtain n differentiation indicators, where n is the return value of the historical maintenance times; generating, based on the n differentiation indicators, a performance degradation curve of the operating equipment based on the maintenance identification feature; The performance degradation curve of the equipment running based on the maintenance identification feature is subjected to degradation rate detection, and when the degradation rate is greater than or equal to a preset degradation rate, second warning information is obtained, and a warning is issued on the continuous maintenance quality according to the second warning information.

3. The method according to claim 2, characterized in that If the historical maintenance times of the characteristic operation equipment are not empty, obtain n differentiation indicators, including: According to the historical maintenance times of the characteristic operation equipment, a differentiation index between two adjacent maintenance times is calculated to obtain a first differentiation index, a second differentiation index, ... an nth differentiation index; Among them, the first differentiation index is the differentiated recognition result of the first device twin model and the second device twin model under the first maintenance number, the second differentiation index is the differentiated recognition result of the first device twin model and the second device twin model under the second maintenance number,…, the nth differentiation index is the differentiated recognition result of the first device twin model and the second device twin model under the nth maintenance number; The first differentiation index, the second differentiation index, ... the nth differentiation index are integrated to obtain n differentiation indexes.

4. The method according to claim 3, characterized in that Methods for obtaining each differentiation indicator include: Obtaining a first equipment twin model of the nth maintenance, wherein the first equipment twin model of the nth maintenance is a digital twin simulation result of the characteristic operation equipment before the nth maintenance; Determine an operation data set corresponding to the characteristic operation equipment after the nth maintenance; Perform digital twin simulation on the operating data set corresponding to the nth maintenance of the characteristic operating equipment to determine the twin model of the second equipment for the nth maintenance; Model differentiation monitoring is performed based on the first equipment twin model of the nth maintenance and the second equipment twin model of the nth maintenance to obtain the nth differentiation index.

5. The method according to claim 1, characterized in that Performing model differentiation identification according to the first device twin model and the second device twin model, the method comprising: Establishing a differentiated identification network layer, wherein the differentiated identification network layer includes a digital twin network, and the digital twin network includes a first data input channel and a second data input channel; Connecting the first data input channel to the first device twin model to receive a device performance test data set from the first device twin model; Connecting the second data input channel to the second device twin model to receive a device performance test data set from the second device twin model; A similarity analysis is performed based on the first data input channel and the second data input channel to obtain a differentiation index.

6. The method according to claim 5, characterized in that Performing similarity analysis based on the first data input channel and the second data input channel to obtain a differentiation index, the method further includes: According to the first data input channel and the second data input channel, the energy consumption change characteristics of the equipment before and after the maintenance of the characteristic operation equipment, the equipment operation sensitivity characteristics, the equipment recovery rate characteristics and the equipment start-stop power characteristics are obtained; A similarity analysis is performed based on the equipment energy consumption change characteristics, equipment operation sensitivity characteristics, equipment recovery rate characteristics, and equipment start-stop power characteristics to obtain differentiation indicators.

7. The hydropower plant equipment monitoring and early warning system based on twin plants is characterized by: A method for monitoring and early warning of hydropower plant equipment based on twin plants according to any one of claims 1 to 6, the method comprising: An equipment twin modeling module, wherein the equipment twin modeling module is used to obtain a hydropower plant equipment model library by performing digital twin modeling on the equipment of the hydropower plant; A characteristic operation equipment module, wherein the characteristic operation equipment module is used to acquire characteristic operation equipment by collecting the real-time status of the equipment in the hydropower plant, wherein the characteristic operation equipment is the operation equipment after maintenance in the hydropower plant; A twin model retrieval module, wherein the twin model retrieval module is used to call the first equipment twin model of the characteristic operation equipment according to the hydropower plant equipment model library, wherein the first equipment twin model is a digital twin simulation result of the characteristic operation equipment before maintenance; A maintenance twin model module, wherein the maintenance twin model module is used to obtain a second device twin model of the feature operation device by collecting current operation data of the feature operation device, wherein the second device twin model is a digital twin simulation result of the feature operation device after maintenance; A differentiation test module, which is used to perform model testing based on the first device twin model and the second device twin model, including differentiation testing of random independent partition models and differentiation testing of random continuous partition models, and output differentiation indicators, wherein the random independent partition model includes a model of an independent operating device randomly selected from the first device twin model, and a model of the corresponding independent device in the second device twin model, and the random continuous partition model is a model of at least two continuous devices with a continuous collaborative operating relationship randomly selected from the first device twin model, and a model of the corresponding continuous device in the second device twin model; A maintenance quality warning module, wherein the maintenance quality warning module is used to obtain first warning information when the differentiation index is not within a preset differentiation index threshold, and to issue a warning on the current maintenance quality according to the first warning information.

Images