Double-engine modeling transformer substation equipment model updating method and related equipment

Abstract

The invention relates to the technical field of model data processing, and discloses a double-engine modeling transformer substation equipment model updating method and device, and the method comprises the steps: obtaining the real-time operation original data of equipment in a transformer substation, and injecting a preset controllable periodic signal into the original data to obtain target data; respectively inputting the original data and the target data into a heterogeneous dual-simulation engine to obtain difference characteristics for the heterogeneous dual-simulation engine; and obtaining a periodical characteristic of the preset controllable periodic signal, and updating the heterogeneous dual-simulation engine according to a phase difference between the difference characteristic and the periodic characteristic. Through a parallel computing and cross validation mechanism of heterogeneous double simulation engines, the system can capture complementary characteristics of dynamic behaviors of equipment, and meanwhile, difference characteristics and phase difference analysis generated by active injection of controllable periodic signals are utilized, so that single model deviation is remarkably reduced, robustness of multi-source data disturbance is improved, and the method is suitable for large-scale popularization and application. And the simulation result better fits the actual state of the equipment.

Description

Technical Field

[0001] This invention relates to the field of model data processing technology, and in particular to a method for updating substation equipment models using dual-engine modeling and related equipment. Background Technology

[0002] Existing methods for updating substation equipment models typically rely on a single simulation engine, which makes it difficult to effectively capture the dynamic behavior characteristics of equipment under complex operating conditions and lacks robustness to multi-source disturbances such as temperature, current, and voltage. Traditional parameter update mechanisms are mostly static or passive adjustments, which cannot track changes in equipment status such as aging and load fluctuations in real time, leading to the accumulation of deviations between simulation results and actual operation, affecting the accuracy of early fault warnings and the reliability of the model.

[0003] An existing patent (publication number CN119397744A) discloses a method for updating a multi-parameter simulation model of a substation. This existing patent, employing a single-model static parameter adjustment mechanism, relies on manual, iterative modification of equipment parameters and comparison of simulation results, lacking real-time dynamic tracking capabilities. It does not introduce active excitation signals, resulting in insufficient sensitivity to abnormal responses and an inability to capture transient characteristics of the equipment. The parameter update process is offline and passive, making it difficult to cope with the coupled effects of multi-source disturbances such as temperature and current. Furthermore, fault location relies on posterior data analysis, failing to achieve early warning. Specifically, manually adjusting parameters for each device and restoring them to their original values ​​is inefficient and cannot respond in real-time to changes in equipment status (such as load fluctuations and aging). The single model cannot reduce bias through heterogeneous engine complementarity, exhibiting poor adaptability to multi-source data disturbances (such as the combined effects of temperature and current). Relying on output data anomalies to infer faulty equipment lacks the feature of actively injecting signals to amplify minor anomalies, resulting in low accuracy in early fault identification. It does not involve frequency domain calibration methods such as phase difference analysis, making it unable to track the dynamic response characteristics of the equipment (such as transient process lag issues). Summary of the Invention

[0004] This invention provides a substation equipment model update system with dual-engine modeling to solve existing technical problems, addressing issues such as insufficient dynamic tracking capability of a single model, poor robustness against multi-source disturbances, and low fault detection sensitivity.

[0005] To address the aforementioned technical problems, this invention provides a method for updating substation equipment models using dual-engine modeling, comprising: The raw data of real-time operation of equipment in the substation is acquired, and a preset controllable periodic signal is injected into the raw data to obtain the target data; The original data and the target data are respectively input into the heterogeneous dual simulation engine to obtain the difference features for the heterogeneous dual simulation engine; The periodic characteristics of the preset controllable periodic signal are obtained, and the heterogeneous dual simulation engine is updated according to the phase difference between the difference characteristics and the periodic characteristics.

[0006] Optionally, the step of inputting the original data and the target data into the heterogeneous dual simulation engine respectively to obtain the difference features for the heterogeneous dual simulation engine includes: The original data and the target data are respectively input into the first simulation engine in the heterogeneous dual simulation engine to obtain a first simulation result corresponding to the original data and a second simulation result corresponding to the target data; Based on the first simulation results and the second simulation results, the differential characteristics of the first simulation engine in the heterogeneous dual simulation engine are determined.

[0007] Optionally, the step of inputting the original data and the target data into the heterogeneous dual simulation engine respectively to obtain the difference features for the heterogeneous dual simulation engine further includes: The original data and the target data are respectively input into the second simulation engine in the heterogeneous dual simulation engine to obtain a third simulation result corresponding to the original data and a fourth simulation result corresponding to the target data; Based on the third and fourth simulation results, the differential characteristics of the second simulation engine in the heterogeneous dual simulation engine are determined.

[0008] Optionally, the step of inputting the original data and the target data into the heterogeneous dual simulation engine respectively to obtain the difference features for the heterogeneous dual simulation engine further includes: The original data is jointly simulated by the first and second simulation engines in the heterogeneous dual simulation engine to obtain the fifth simulation result. The target data is jointly simulated by the first and second simulation engines in the heterogeneous dual simulation engine to obtain the sixth simulation result. Based on the fifth and sixth simulation results, the differences in characteristics when performing joint simulation for heterogeneous dual simulation engines are determined.

[0009] Optionally, the step of updating the heterogeneous dual simulation engine based on the phase difference between the difference feature and the periodic feature includes: The first simulation engine in the heterogeneous dual simulation engine is updated based on the phase difference between the periodic feature and the first difference feature, using the difference features of the first simulation engine as the first difference feature.

[0010] Optionally, the step of updating the heterogeneous dual simulation engine based on the phase difference between the difference feature and the periodic feature further includes: The difference features of the second simulation engine in the heterogeneous dual simulation engine are used as the second difference features, and the second simulation engine in the heterogeneous dual simulation engine is updated according to the phase difference between the periodic features and the second difference features.

[0011] Optionally, the step of updating the heterogeneous dual simulation engine based on the phase difference between the difference feature and the periodic feature further includes: The difference features during joint simulation of the heterogeneous dual simulation engines are used as the third difference features. The first and second simulation engines in the heterogeneous dual simulation engines are updated based on the phase difference between the periodic features and the third difference features.

[0012] Secondly, this application provides a substation equipment model update system with dual-engine modeling, including: The data processing module is used to acquire raw data of real-time operation of equipment in the substation and inject a preset controllable periodic signal into the raw data to obtain target data. The difference feature extraction module is used to input the original data and the target data into the heterogeneous dual simulation engine respectively to obtain the difference features for the heterogeneous dual simulation engine. The update module is used to obtain the periodic characteristics of the preset controllable periodic signal and update the heterogeneous dual simulation engine according to the phase difference between the difference characteristics and the periodic characteristics.

[0013] Thirdly, this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the substation equipment model update method of dual-engine modeling as described above.

[0014] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the substation equipment model update method for dual-engine modeling as described above. This invention provides a dual-engine modeling method for updating substation equipment models. Compared with existing technologies, this method achieves the following advantages: Through parallel computing and cross-validation mechanisms using heterogeneous dual simulation engines, the system can capture complementary characteristics of the dynamic behavior of equipment. Utilizing differential features and phase difference analysis generated by the active injection of controllable periodic signals, the system significantly reduces the bias of individual models and improves robustness to disturbances from multiple sources such as temperature, current, and voltage, making simulation results more closely reflect the actual state of the equipment. Based on the relationship between differential features and the phase difference of controllable periodic signals, the system dynamically adjusts the sensitivity parameters of each engine and the parameters of the joint model. This design transforms traditional static parameter updates into a real-time feedback closed loop, automatically matching the optimal parameter combination through mathematical relationships to ensure the model continuously tracks dynamic characteristics such as equipment aging and load changes. Active excitation with controllable periodic signals allows the system to amplify abnormal equipment responses. The differential analysis module extracts the implicit fault feature phase difference by comparing simulation results with and without applied signals. Combined with hierarchical verification of multi-engine joint differentials, this significantly improves the detection sensitivity and location accuracy of early faults. The heterogeneous dual-engine architecture supports task division and conquest; independent operation of a single engine is used for rapid local parameter tuning, while the joint engine is used for global state calibration. By combining the lightweight deployment of vectorization preprocessing and parameter update modules, the system reduces computational load while ensuring simulation accuracy, meeting the millisecond-level real-time monitoring requirements of substation equipment. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the operation of the substation equipment model update method using dual-engine modeling in this embodiment of the invention. Figure 2 This is a flowchart illustrating the process of updating the first simulation engine separately in an embodiment of the present invention; Figure 3 This is a flowchart illustrating the process of updating the second simulation engine separately in an embodiment of the present invention; Figure 4 This is a flowchart illustrating the updating of the first and second simulation engines during co-simulation in an embodiment of the present invention. Figure 5 This is a simulation diagram of the target data in an embodiment of the present invention; Figure 6 This is a simulation diagram of the original data in an embodiment of the present invention; Figure 7 This is a feature difference diagram for different data in an embodiment of the present invention; Figure 8 This is a schematic diagram of an adjustable periodic signal in an embodiment of the present invention; Figure 9 This is the overall state in the embodiments of the present invention. and differences Relationship diagram; Figure 10 This is the overall state in the embodiments of the present invention. and differences Relationship diagram; Figure 11 This is the overall state in the embodiments of the present invention. and differences Differences and characteristics Relationship diagram; Figure 12 This is a schematic diagram of the substation equipment model update system with dual-engine modeling in an embodiment of the present invention; Figure 13 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application; Figure 14 This is a schematic diagram of the structure of a computer-readable storage medium provided in an embodiment of this application. Detailed Implementation

[0016] To make the technical solution of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0017] like Figure 1 As shown, a substation equipment model update method using dual-engine modeling includes: S110. Obtain the raw data of real-time operation of equipment in the substation, and inject a preset controllable periodic signal into the raw data to obtain the target data; For example, the collected initial data is vectorized to obtain raw data, and a preset controllable periodic signal is injected into the raw data to obtain target data. The raw data includes temperature data, current data, and voltage data. The preset controllable periodic signal is a sine wave or sine wave signal with adjustable amplitude and frequency.

[0018] S120. Input the original data and the target data into the heterogeneous dual simulation engine respectively to obtain the difference features for the heterogeneous dual simulation engine; In one possible implementation, the step of inputting the original data and the target data into the heterogeneous dual simulation engine respectively to obtain the difference features specific to the heterogeneous dual simulation engine includes: The original data and the target data are respectively input into the first simulation engine in the heterogeneous dual simulation engine to obtain a first simulation result corresponding to the original data and a second simulation result corresponding to the target data; Based on the first simulation results and the second simulation results, the differential characteristics of the first simulation engine in the heterogeneous dual simulation engine are determined.

[0019] For example, without applying a controllable periodic signal to the original data, the state of the substation equipment itself is simulated by the real-time simulation engine A to obtain simulation feature X1 (first simulation result); while with the application of a controllable periodic signal, the state of the substation equipment itself is simulated by the real-time simulation engine A (first simulation engine) to obtain simulation feature X2 (second simulation result), and the difference feature M1 of the real-time simulation engine A is obtained based on the difference feature between simulation feature X2 and simulation feature X1.

[0020] For example, such as Figure 2 As shown, without applying a controllable periodic signal to the original data, the state of the substation equipment itself is simulated by the real-time simulation engine A to obtain simulation feature X1; while with the application of a controllable periodic signal, the state of the substation equipment itself is simulated by the real-time simulation engine A to obtain simulation feature X2. The difference feature M1 of the real-time simulation engine A is obtained based on the difference between simulation feature X2 and simulation feature X1; the parameters of the real-time simulation engine A are adjusted based on the phase difference between the difference feature M1 and the periodic feature of the controllable periodic signal.

[0021] The state of one of the transformers, K, is simulated using real-time simulation engine A, specifically as follows: like Figure 5 As shown in the simulation diagram for the target data, the state of the substation equipment itself is simulated by the real-time simulation engine A to obtain simulation feature X2. like Figure 6 As shown in the simulation diagram based on the original data, the state of the substation equipment itself is simulated using the real-time simulation engine A to obtain simulation features X1.

[0022] The state of transformer K is then simulated using real-time simulation engine A for 100 iterations. Specifically: Sample data were collected from 100 transformers K, among which... This represents the deviation between the transformer K state simulated by the real-time simulation engine A and the actual transformer state; this deviation is represented by the ranking of any deviation among the other 100 sample data.

[0023] For example, if the deviation of a simulation result for transformer K in a certain sampling is better than that of the other 50 samples, then it can be said that... .

[0024] This indicates that the simulation deviation is better than 50% of the historical samples. Essentially, it is a normalization process: converting the absolute error into percentile ranking to avoid direct comparison of deviations of data with different dimensions (such as temperature / current), and to ensure that parameter adjustments focus on the relative performance improvement of the model in historical data.

[0025] in, This represents the phase difference between the difference feature M1 of the transformer K in 100 sample acquisitions and the periodic feature of the controllable periodic signal.

[0026] like Figure 9 As shown, the overall state of transformer K and differences If we establish a mathematical model for the relationship between them, then we have: (Formula 1); In Formula 1 above, Used to control differential characteristics The changing trend and the overall state of transformer K Approximate constant, Used to control differential characteristics The initial size and the overall state of transformer K An approximate constant. And by Figure 9 When the data in the middle can be determined, when , At this point, the overall state of the simulated transformer K by simulation engine A tends to be the same as the actual state. Therefore, at this time: ; In the formula, This represents the state simulated solely by a sub-model based on real-time simulation engine A.

[0027] In one possible implementation, the step of inputting the original data and the target data into the heterogeneous dual simulation engine respectively to obtain the difference features for the heterogeneous dual simulation engine further includes: The original data and the target data are respectively input into the second simulation engine in the heterogeneous dual simulation engine to obtain a third simulation result corresponding to the original data and a fourth simulation result corresponding to the target data; Based on the third and fourth simulation results, the differential characteristics of the second simulation engine in the heterogeneous dual simulation engine are determined.

[0028] For example, without applying a controllable periodic signal to the original data, the state of the substation equipment itself is simulated by the real-time simulation engine B (the second simulation engine) to obtain simulation feature Y1 (the third simulation result); while after applying a controllable periodic signal, the state of the substation equipment itself is simulated by the real-time simulation engine B to obtain simulation feature Y2 (the fourth simulation result), and the difference feature M2 of the real-time simulation engine B is obtained based on the difference between simulation feature Y2 and simulation feature Y1.

[0029] For example, such as Figure 3As shown, without applying a controllable periodic signal to the original data, the state of the substation equipment itself is simulated by the real-time simulation engine B to obtain simulation feature Y1; while with the application of a controllable periodic signal, the state of the substation equipment itself is simulated by the real-time simulation engine B to obtain simulation feature Y2. The difference feature M2 of the real-time simulation engine B is obtained based on the difference between simulation feature Y2 and simulation feature Y1; the parameters of the real-time simulation engine B are adjusted based on the phase difference between the difference feature M2 and the periodic feature of the controllable periodic signal.

[0030] Similarly, the state of transformer K is simulated 100 times using real-time simulation engine B. Specifically, the simulation also includes: Sample data were collected from 100 transformers K, among which... This represents the deviation between the transformer K state simulated by the real-time simulation engine B and the actual transformer state; this deviation is represented by the ranking of any deviation among the other 100 sample data.

[0031] For example, if the deviation of a simulation result for transformer K in a certain sampling is better than that of the other 50 samples, then it can be said that... .

[0032] in, This represents the phase difference between the difference feature M2 of the transformer K in 100 sample acquisitions and the periodic feature of the controllable periodic signal.

[0033] Among them, the difference feature M2 represents the difference feature of the transformer K in 100 sample collections.

[0034] like Figure 10 As shown, the overall state of transformer K and differences If we establish a mathematical model for the relationship between them, then we have: (Formula 2); In formula 2 above, Used to control differential characteristics The changing trend and the overall state of transformer K Approximate constant, Used to control differential characteristics The initial size and the overall state of transformer K An approximate constant. And by Figure 10 When the data in the middle can be determined, when , At this point, the overall state of the simulated transformer K in simulation engine B tends to be the same as the actual state. Therefore, at this time: ; In the formula, This represents the state simulated solely by a sub-model based on the real-time simulation engine B.

[0035] In one possible implementation, the step of inputting the original data and the target data into the heterogeneous dual simulation engine respectively to obtain the difference features for the heterogeneous dual simulation engine further includes: The original data is jointly simulated by the first and second simulation engines in the heterogeneous dual simulation engine to obtain the fifth simulation result. The target data is jointly simulated by the first and second simulation engines in the heterogeneous dual simulation engine to obtain the sixth simulation result. Based on the fifth and sixth simulation results, the differences in characteristics when performing joint simulation for heterogeneous dual simulation engines are determined.

[0036] For example, without applying a controllable periodic signal to the original data, the state of the substation equipment itself is simulated simultaneously through real-time simulation engine A and real-time simulation engine B, and simulation feature Z1 (the fifth simulation result) is obtained; while with the application of a controllable periodic signal, the state of the substation equipment itself is simulated simultaneously through real-time simulation engine A and real-time simulation engine B, and simulation feature Z2 (the sixth simulation result) is obtained. The difference feature M3 of the joint difference between real-time simulation engine A and real-time simulation engine B is obtained based on the difference between simulation feature Z2 and simulation feature Z1.

[0037] For example, such as Figure 4 As shown, without applying a controllable periodic signal to the original data, the state of the substation equipment itself is simulated simultaneously through real-time simulation engine A and real-time simulation engine B to obtain simulation feature Z1; while with the application of a controllable periodic signal, the state of the substation equipment itself is simulated simultaneously through real-time simulation engine A and real-time simulation engine B to obtain simulation feature Z2.

[0038] The difference feature M3 between the simulation feature Z2 and the simulation feature Z1 is obtained by combining the difference feature M3 of the real-time simulation engine A and the real-time simulation engine B. The parameters of the combined real-time simulation engine A and the real-time simulation engine B are adjusted by the phase difference between the difference feature M3 and the periodic feature of the controllable periodic signal.

[0039] Similarly, the state of transformer K is simulated 100 times using real-time simulation engine A and real-time simulation engine B. Specifically, the following steps are also taken: Sample data were collected from 100 transformers K, among which... This represents the deviation between the transformer K state simulated by real-time simulation engine A and real-time simulation engine B and the actual state of the transformer; this deviation is represented by the ranking of any deviation among the other 100 sample data.

[0040] For example, if the deviation of a simulation result for transformer K in a certain sampling is better than that of the other 50 samples, then it can be said that... .

[0041] Overall state of transformer K The differential feature M3, which is combined with real-time simulation engine A and real-time simulation engine B, consists of differential features. Differences and characteristics This is represented as follows: For example, differential feature M3 = differential feature M1. Difference characteristics ).like Figure 11 As shown, by establishing a mathematical model, we have: (Formula 3); In formula 3 above, This is used to control the deviation between the simulated state of transformer K and the actual state of the model jointly generated by real-time simulation engines A and B. When At this point, the deviation between the simulated state of transformer K by the model jointly created by real-time simulation engine A and real-time simulation engine B and the actual state approaches 0. Therefore, at this time: ; In the formula, This represents the state simulated solely based on the joint model of real-time simulation engine A and real-time simulation engine B.

[0042] Here, M3 is not simply a superposition of M1 and M2, but rather an independent difference characteristic of the dual-engine joint simulation results. The mathematical relationship between M3 and M1 / M2 in Formula 3 (as shown in media / image21.wmf) suggests that it is achieved through a weighted fusion of the dual-engine differences (e.g., M3 = ...). ), weight Global optimization is achieved through dynamic adjustment of phase difference.

[0043] parameter and parameters The regression coefficients are established using statistical sample data (100 transformer tests): when the phase difference between the difference feature M and the actual state m of the equipment satisfies a specific mathematical relationship, the model parameters are considered to have been calibrated.

[0044] S130. Obtain the periodicity characteristics of the preset controllable periodic signal, and update the heterogeneous dual simulation engine according to the phase difference between the difference characteristics and the periodic characteristics.

[0045] In one possible implementation, the step of updating the heterogeneous dual simulation engine based on the phase difference between the difference feature and the periodic feature includes: The first simulation engine in the heterogeneous dual simulation engine is updated based on the phase difference between the periodic feature and the first difference feature, using the difference features of the first simulation engine as the first difference feature.

[0046] For example, adjusting the sensitivity parameters of real-time simulation engine A , ; In one possible implementation, the step of updating the heterogeneous dual simulation engine based on the phase difference between the difference feature and the periodic feature further includes: The difference features of the second simulation engine in the heterogeneous dual simulation engine are used as the second difference features, and the second simulation engine in the heterogeneous dual simulation engine is updated according to the phase difference between the periodic features and the second difference features.

[0047] For example, adjusting the sensitivity parameters of the real-time simulation engine B. , ; In one possible implementation, the step of updating the heterogeneous dual simulation engine based on the phase difference between the difference feature and the periodic feature further includes: The difference features during joint simulation of the heterogeneous dual simulation engines are used as the third difference features. The first and second simulation engines in the heterogeneous dual simulation engines are updated based on the phase difference between the periodic features and the third difference features.

[0048] For example, adjusting the parameters of the model sensitivity after combining real-time simulation engine A and real-time simulation engine B. .

[0049] Parameters (such as) , The phase difference analysis is essentially a frequency domain response calibration. For example, when the phase difference between M1 and the excitation signal increases, it indicates that the model's dynamic response is lagging, and increasing the phase difference in this case... Formula 1 can improve the simulation engine's ability to track high-frequency disturbances, enabling... It more closely reflects the actual transient process.

[0050] This application discloses a dual-engine modeling method for updating substation equipment models. It acquires raw data of real-time operation of equipment in the substation and injects a preset controllable periodic signal into this raw data to obtain target data. The raw data and target data are then input into a heterogeneous dual-simulation engine to obtain difference features specific to the engine. The periodic features of the preset controllable periodic signal are also obtained, and the heterogeneous dual-simulation engine is updated based on the phase difference between these difference features and the periodic features. Through the parallel computing and cross-validation mechanism of the heterogeneous dual-simulation engine, the system can capture complementary features of the equipment's dynamic behavior. Utilizing the difference features and phase difference analysis generated by the actively injected controllable periodic signal significantly reduces the bias of a single model and improves robustness to disturbances from multiple sources such as temperature, current, and voltage, making the simulation results more closely reflect the actual state of the equipment. Based on the relationship between the difference features and the phase difference of the controllable periodic signal, the system dynamically adjusts the sensitivity parameters of each engine and the joint model parameters. This design transforms traditional static parameter updates into a real-time feedback closed loop, automatically matching the optimal parameter combination through mathematical relationships to ensure that the model continuously tracks dynamic characteristics such as equipment aging and load changes. By actively exciting controllable periodic signals, the system can amplify abnormal equipment responses. The difference analysis module extracts the implicit fault characteristic phase difference by comparing simulation results with and without applied signals. Combined with hierarchical verification of differences using multiple engines, it significantly improves the detection sensitivity and location accuracy of early faults. A heterogeneous dual-engine architecture supports task divide-and-conquer; single-engine independent computation is used for rapid local parameter tuning, while the joint engine is used for global state calibration. Combined with the lightweight deployment of vectorized preprocessing and parameter update modules, the system reduces computational load while ensuring simulation accuracy, meeting the millisecond-level real-time monitoring requirements of substation equipment.

[0051] In one possible implementation, the heterogeneous dual simulation engine includes a real-time simulation engine A and a real-time simulation engine B; the heterogeneous dual simulation engine simulates the state of the substation equipment itself based on the raw data of the real-time operation of the equipment in the substation; the structure of the real-time simulation engine A and the real-time simulation engine B is a heterogeneous dual-model architecture.

[0052] For example, if raw data is acquired for transformer K in a substation and a simulation model is constructed, and this model is based on a dual-engine dynamic behavior simulation model, then the sub-model based on real-time simulation engine A is: ; The sub-model based on real-time simulation engine B is: ; The model based on the combination of real-time simulation engine A and real-time simulation engine B is as follows: ; In the above formula, m represents the overall state of transformer K in the substation (if the input raw data is the current signal at several key points in the line, then the output simulation result is the overall load of transformer K). , This refers to the state simulated solely based on a sub-model of real-time simulation engine A or solely based on a sub-model of real-time simulation engine B.

[0053] This indicates the different characteristics of the transformer K (such as adding a controllable periodic signal to the input current signal and not adding a controllable periodic signal). This indicates the difference in characteristics of the transformer K (such as the difference between adding and not adding a controllable periodic signal to the input current signal).

[0054] in, , This refers to the parameters used to adjust the sensitivity of the real-time simulation engine A. , This refers to the parameters used to adjust the sensitivity of the real-time simulation engine B. This represents the parameter used to adjust the sensitivity of the model after real-time simulation engine A and real-time simulation engine B are combined.

[0055] like Figure 7 As shown, M1 and M2 are not directly measured values, but rather analytical results of the engine's output with or without a periodic signal applied. Taking M1 as an example: subtracting X1 (without signal simulation) from X2 (with signal simulation) yields the waveform difference, which is then compared with... Figure 8 The original excitation signals shown are compared in phase; the phase difference reflects the model's response hysteresis characteristics and is used to adjust the parameters. , .

[0056] In one possible implementation, such as Figure 12 As shown, this application provides a substation equipment model update system with dual-engine modeling, including: The data processing module 201 is used to acquire raw data of real-time operation of equipment in the substation, and inject a preset controllable periodic signal into the raw data to obtain target data; The difference feature extraction module 202 is used to input the original data and the target data into the heterogeneous dual simulation engine respectively to obtain the difference features for the heterogeneous dual simulation engine. The update module 203 is used to obtain the periodic characteristics of the preset controllable periodic signal and update the heterogeneous dual simulation engine according to the phase difference between the difference characteristics and the periodic characteristics.

[0057] Through parallel computing and cross-validation mechanisms using heterogeneous dual simulation engines, the system can capture complementary characteristics of the dynamic behavior of equipment. Utilizing differential features and phase difference analysis generated by the active injection of controllable periodic signals, the system significantly reduces the bias of individual models and improves robustness to disturbances from multiple sources such as temperature, current, and voltage, making simulation results more closely reflect the actual state of the equipment. Based on the relationship between differential features and the phase difference of controllable periodic signals, the system dynamically adjusts the sensitivity parameters of each engine and the parameters of the joint model. This design transforms traditional static parameter updates into a real-time feedback closed loop, automatically matching the optimal parameter combination through mathematical relationships to ensure the model continuously tracks dynamic characteristics such as equipment aging and load changes. Active excitation with controllable periodic signals allows the system to amplify abnormal equipment responses. The differential analysis module extracts the implicit fault feature phase difference by comparing simulation results with and without applied signals. Combined with hierarchical verification of multi-engine joint differentials, this significantly improves the detection sensitivity and location accuracy of early faults. The heterogeneous dual-engine architecture supports task division and conquest; independent operation of a single engine is used for rapid local parameter tuning, while the joint engine is used for global state calibration. By combining the lightweight deployment of vectorization preprocessing and parameter update modules, the system reduces computational load while ensuring simulation accuracy, meeting the millisecond-level real-time monitoring requirements of substation equipment.

[0058] In one possible implementation, such as Figure 13 As shown, this application embodiment provides a terminal device 300, including: a memory 310, a processor 320, and a first computer program 311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the first computer program 311, it implements the steps of a substation equipment model update method with dual-engine modeling.

[0059] Through parallel computing and cross-validation mechanisms using heterogeneous dual simulation engines, the system can capture complementary characteristics of the dynamic behavior of equipment. Utilizing differential features and phase difference analysis generated by the active injection of controllable periodic signals, the system significantly reduces the bias of individual models and improves robustness to disturbances from multiple sources such as temperature, current, and voltage, making simulation results more closely reflect the actual state of the equipment. Based on the relationship between differential features and the phase difference of controllable periodic signals, the system dynamically adjusts the sensitivity parameters of each engine and the parameters of the joint model. This design transforms traditional static parameter updates into a real-time feedback closed loop, automatically matching the optimal parameter combination through mathematical relationships to ensure the model continuously tracks dynamic characteristics such as equipment aging and load changes. Active excitation with controllable periodic signals allows the system to amplify abnormal equipment responses. The differential analysis module extracts the implicit fault feature phase difference by comparing simulation results with and without applied signals. Combined with hierarchical verification of multi-engine joint differentials, this significantly improves the detection sensitivity and location accuracy of early faults. The heterogeneous dual-engine architecture supports task division and conquest; independent operation of a single engine is used for rapid local parameter tuning, while the joint engine is used for global state calibration. By combining the lightweight deployment of vectorization preprocessing and parameter update modules, the system reduces computational load while ensuring simulation accuracy, meeting the millisecond-level real-time monitoring requirements of substation equipment.

[0060] In one possible implementation, such as Figure 14 As shown, this application embodiment provides a computer-readable storage medium 400, on which a second computer program 411 is stored. When the second computer program 411 is executed by a processor, it implements the steps of a substation equipment model update method for dual-engine modeling.

[0061] Through parallel computing and cross-validation mechanisms using heterogeneous dual simulation engines, the system can capture complementary characteristics of the dynamic behavior of equipment. Utilizing differential features and phase difference analysis generated by the active injection of controllable periodic signals, the system significantly reduces the bias of individual models and improves robustness to disturbances from multiple sources such as temperature, current, and voltage, making simulation results more closely reflect the actual state of the equipment. Based on the relationship between differential features and the phase difference of controllable periodic signals, the system dynamically adjusts the sensitivity parameters of each engine and the parameters of the joint model. This design transforms traditional static parameter updates into a real-time feedback closed loop, automatically matching the optimal parameter combination through mathematical relationships to ensure the model continuously tracks dynamic characteristics such as equipment aging and load changes. Active excitation with controllable periodic signals allows the system to amplify abnormal equipment responses. The differential analysis module extracts the implicit fault feature phase difference by comparing simulation results with and without applied signals. Combined with hierarchical verification of multi-engine joint differentials, this significantly improves the detection sensitivity and location accuracy of early faults. The heterogeneous dual-engine architecture supports task division and conquest; independent operation of a single engine is used for rapid local parameter tuning, while the joint engine is used for global state calibration. By combining the lightweight deployment of vectorization preprocessing and parameter update modules, the system reduces computational load while ensuring simulation accuracy, meeting the millisecond-level real-time monitoring requirements of substation equipment.

[0062] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for updating substation equipment models using dual-engine modeling, characterized in that, include: The raw data of real-time operation of equipment in the substation is acquired, and a preset controllable periodic signal is injected into the raw data to obtain the target data; The original data and the target data are respectively input into the heterogeneous dual simulation engine to obtain the difference features for the heterogeneous dual simulation engine; The periodic characteristics of the preset controllable periodic signal are obtained, and the heterogeneous dual simulation engine is updated according to the phase difference between the difference characteristics and the periodic characteristics.

2. The substation equipment model update method based on dual-engine modeling according to claim 1, characterized in that, The step of inputting the original data and the target data into the heterogeneous dual simulation engine respectively to obtain the difference features for the heterogeneous dual simulation engine includes: The original data and the target data are respectively input into the first simulation engine in the heterogeneous dual simulation engine to obtain a first simulation result corresponding to the original data and a second simulation result corresponding to the target data; Based on the first simulation results and the second simulation results, the differential characteristics of the first simulation engine in the heterogeneous dual simulation engine are determined.

3. The substation equipment model update method based on dual-engine modeling according to claim 1, characterized in that, The step of inputting the original data and the target data into the heterogeneous dual simulation engine respectively to obtain the difference features for the heterogeneous dual simulation engine further includes: The original data and the target data are respectively input into the second simulation engine in the heterogeneous dual simulation engine to obtain a third simulation result corresponding to the original data and a fourth simulation result corresponding to the target data; Based on the third and fourth simulation results, the differential characteristics of the second simulation engine in the heterogeneous dual simulation engine are determined.

4. The substation equipment model update method based on dual-engine modeling according to claim 1, characterized in that, The step of inputting the original data and the target data into the heterogeneous dual simulation engine respectively to obtain the difference features for the heterogeneous dual simulation engine further includes: The original data is jointly simulated by the first and second simulation engines in the heterogeneous dual simulation engine to obtain the fifth simulation result. The target data is jointly simulated by the first and second simulation engines in the heterogeneous dual simulation engine to obtain the sixth simulation result. Based on the fifth and sixth simulation results, the differences in characteristics when performing joint simulation for heterogeneous dual simulation engines are determined.

5. The substation equipment model update method based on dual-engine modeling according to claim 2, characterized in that, The step of updating the heterogeneous dual simulation engine based on the phase difference between the difference features and the periodic features includes: The first simulation engine in the heterogeneous dual simulation engine is updated based on the phase difference between the periodic feature and the first difference feature, using the difference features of the first simulation engine as the first difference feature.

6. The substation equipment model update method based on dual-engine modeling according to claim 3, characterized in that, The step of updating the heterogeneous dual simulation engine based on the phase difference between the difference features and the periodic features further includes: The difference features of the second simulation engine in the heterogeneous dual simulation engine are used as the second difference features, and the second simulation engine in the heterogeneous dual simulation engine is updated according to the phase difference between the periodic features and the second difference features.

7. The substation equipment model update method based on dual-engine modeling according to claim 4, characterized in that, The step of updating the heterogeneous dual simulation engine based on the phase difference between the difference features and the periodic features further includes: The difference features during joint simulation of the heterogeneous dual simulation engines are used as the third difference features. The first and second simulation engines in the heterogeneous dual simulation engines are updated based on the phase difference between the periodic features and the third difference features.

8. A substation equipment model update system with dual-engine modeling, characterized in that, include: The data processing module is used to acquire raw data of real-time operation of equipment in the substation and inject a preset controllable periodic signal into the raw data to obtain target data. The difference feature extraction module is used to input the original data and the target data into the heterogeneous dual simulation engine respectively to obtain the difference features for the heterogeneous dual simulation engine. The update module is used to obtain the periodic characteristics of the preset controllable periodic signal and update the heterogeneous dual simulation engine according to the phase difference between the difference characteristics and the periodic characteristics.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the substation equipment model update method for dual-engine modeling as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the substation equipment model update method for dual-engine modeling as described in any one of claims 1 to 7.

Images