Artificial Intelligence-Based Distribution Network Real-World Testing System and Control Method

Abstract

This invention discloses an artificial intelligence-based distribution network simulation testing system and control method, comprising the following steps: training an artificial intelligence model based on historical data, and generating a control scheme simulating a target distribution operation scenario according to the topology of the distribution network simulation testing platform; analyzing and controlling the operation of each functional cabinet and controllable device in the distribution network simulation testing platform according to the control scheme to generate the target distribution operation scenario; collecting actual operating data generated by the distribution network simulation testing platform when operating the target distribution operation scenario; inputting the data into the trained artificial intelligence model to evaluate the degree of conformity between the actual operating data and the expected characteristics of the target distribution operation scenario; and the system executing this method. This invention achieves intelligent and high-fidelity generation of test scenarios through data-driven artificial intelligence learning and closed-loop optimization.

Description

Technical Field

[0001] This invention relates to the field of power system testing, and more specifically to a distribution network full-scale testing system and control method based on artificial intelligence. Background Technology

[0002] A true-world distribution network testing system is a crucial infrastructure for power system research, testing, and verification. By constructing a physical environment encompassing real distribution lines, transformers, switches, circuit breakers, reactive power compensation devices, fault simulation equipment, and adjustable loads, it can reproduce the actual operating conditions and fault scenarios of the power grid within a controlled laboratory. This system is primarily used for performance testing of distribution automation terminals (DTUs / FTUs), verification of protection and control strategies, grid connection detection of new distribution equipment (such as smart circuit breakers and SVG), fault cause reproduction and analysis, and empirical research on distribution network optimization algorithms (such as voltage and reactive power control, and load transfer).

[0003] Currently, the mainstream technical solutions for constructing and controlling real-world testing scenarios in power distribution networks mainly include the following: Control methods based on pre-written scripts / rules: This is the most traditional and widely used method. Test engineers pre-write detailed test scripts or set a series of control rules according to test requirements. These scripts precisely specify the sequence of actions and parameters of various devices (such as switches, load cells, and fault generators) at specific points in time. This method can achieve deterministic scenario reproduction, but it relies heavily on the engineer's prior knowledge of power grid operation and fault characteristics. The core problem is that manually written scripts are difficult to simulate the complex and uncertain dynamic processes in a real power grid, such as random load fluctuations, intermittent changes in distributed power generation output, and the coupling and concurrency of multiple events. This results in the generated test scenarios being "too idealized" or "patterned," with significant differences from the rich statistical characteristics and randomness of a real power grid, thus reducing the field applicability of the test conclusions.

[0004] Control methods based on simple playback of historical data: To enhance the realism of the scenario, some systems incorporate historical operating data (such as voltage, current, active power, and reactive power curves) obtained from production management systems (e.g., SCADA, DMS). This method directly uses the recorded data as the control target, driving the adjustable power supply and load of the test platform to track the data. This method is an improvement over pure script-based approaches, as it can reflect some of the actual operating trajectories. However, its essence is "open-loop playback," which has significant limitations: First, it can only reproduce a single scenario corresponding to existing data and cannot generate "unknown" scenarios that have not occurred but may occur (e.g., extreme fault combinations), resulting in limited test coverage. Second, simple playback cannot adapt to the differences between the physical structure of the test platform and the actual power grid, which may lead to control commands failing to execute or distorted execution results, lacking closed-loop adaptive adjustments to the platform's own characteristics.

[0005] Lack of intelligent evaluation and self-optimization control loop: Existing full-scale testing systems typically employ an open-loop or weakly closed-loop control process of "setup-execution-data collection-manual analysis." After the system executes a preset scenario, engineers need to analyze the collected test data and manually determine whether the scenario simulation is realistic and whether the test objectives have been achieved. If not, engineers must modify the control parameters based on experience and retest. This process is time-consuming, labor-intensive, highly dependent on expert experience, and inefficient. More importantly, due to the lack of a quantitative, automated "realism" evaluation standard and intelligent optimization engine, the system cannot automatically and continuously optimize its scenario generation strategy, making it difficult to achieve autonomous evolution and improvement in the fidelity of the test scenarios.

[0006] In summary, existing power distribution network full-scale testing systems have the following main technical shortcomings in scenario construction: Insufficient realism and diversity of scenarios: Relying on manual presets or simple data replay, it is difficult to generate complex and random scenarios that fully match the statistical regularity, temporal correlation and uncertainty of real massive operating data. In particular, the ability to simulate scenarios such as rare faults and complex disturbances is weak.

[0007] Limited test coverage and generation efficiency: It cannot automatically create new scenarios beyond the historical record. The generation of test cases relies heavily on manual design, has a low degree of automation, and is difficult to meet the need for efficient traversal testing of a massive number of possible scenarios.

[0008] Lack of adaptive and self-optimizing capabilities: The control process has a low level of intelligence and lacks the closed-loop capability to automatically evaluate and optimize control strategies based on output results, which makes it impossible for the system to autonomously improve the realism of its scenario simulation.

[0009] Therefore, there is an urgent need for a real-world testing system and control method that can automatically and intelligently generate high-fidelity, diverse power distribution network operation and fault scenarios. Summary of the Invention

[0010] The technical problem to be solved by this invention is to provide an artificial intelligence-based distribution network simulation test system and control method to address the above-mentioned problems in the prior art. Through data-driven artificial intelligence learning and closed-loop optimization, the test scenario is generated intelligently and with high fidelity.

[0011] To solve the above-mentioned technical problems, the technical solution adopted by the method of the present invention includes the following steps: S1, obtain real historical operation data from the actual power distribution production business system; S2, The artificial intelligence model is trained based on the real historical operating data, and a control scheme for simulating the target power distribution operation scenario is generated according to the topology of the power distribution network real-type test platform. S3, The control scheme is sent to the analysis and control system of the real test platform. The analysis and control system controls the operation of each functional cabinet and controllable equipment in the power distribution network real test platform according to the control scheme to generate the target power distribution operation scenario. S4, Collect the actual operating data generated by the power distribution network real-type test platform when running the target power distribution operation scenario; S5, input the actual operating data into the trained artificial intelligence model to evaluate the degree of conformity between the actual operating data and the expected characteristics of the target power distribution operation scenario; S6. If the compliance level meets the preset requirements, the current control scheme and operating status are saved; if the compliance level does not meet the preset requirements, the artificial intelligence model corrects the control scheme based on the evaluation results and returns to step S3.

[0012] As a further improvement to the method of this invention, the distribution network full-scale test platform includes a 10 kV section and a distribution substation section; the objects controlled in step S3 include: The 10 kV section includes power supply cabinets, switchgear cabinets, fault cabinets, and impedance cabinets. Distribution transformers, distribution function cabinets, and adjustable load function cabinets in the distribution area; The distribution cabinet in the transformer area includes one or more of the following: intelligent distribution box, reactive power compensation cabinet, fault simulation cabinet, and line impedance simulation cabinet.

[0013] As a further improvement to the method of this invention, the adjustable load function cabinet of the distribution area is an adjustable RLC load cabinet.

[0014] As a further improvement to the method of this invention, in step S2, the learning objective of the artificial intelligence model includes the statistical distribution and time-series correlation of voltage, current, power, load characteristics, and fault waveform characteristics of the real power distribution network.

[0015] As a further improvement to the method of this invention, in step S5, the evaluation of the degree of conformity includes: comparing the feature distribution of the actual running data with the feature distribution of the corresponding scene type in the real historical running data in terms of similarity.

[0016] This invention also provides an artificial intelligence-based distribution network simulation test system for implementing the aforementioned artificial intelligence-based distribution network simulation test control method, comprising: The data access module is used to connect to the actual power distribution production business system to obtain real historical operation data; The artificial intelligence module is used to receive and learn the real historical operating data, generate and output a control scheme for controlling the distribution network real-type test platform to simulate the target operating scenario; The analysis and control module is communicatively connected to the artificial intelligence module, and is used to receive the control scheme and generate specific equipment control instructions based on the scheme; The full-scale testing platform includes a primary electrical connection network and a secondary communication connection network. The primary electrical connection network contains multiple functional cabinets and controllable devices, which are used to generate physical electrical scenarios under the instructions of the analysis and control module. The secondary communication connection network is used to connect the functional cabinets, controllable devices and the analysis and control module. The artificial intelligence module is also used to receive the actual operating data generated by the real-world testing platform during operation, and to evaluate the conformity of the data with the expected goals and make adjustments accordingly.

[0017] As a further improvement to the system of this invention, the true-type test platform includes a 10 kV section and a transformer substation section; The primary electrical connection network of the 10 kV section includes a power supply function cabinet, a switchgear cabinet, a fault function cabinet, and an impedance function cabinet connected in sequence. The primary electrical connection network of the transformer substation is connected to the 10 kV substation via a high-voltage fuse, and includes a distribution transformer, as well as at least one primary transformer substation distribution function cabinet and a transformer substation adjustable load function cabinet connected in parallel.

[0018] As a further improvement to the system of this invention, the distribution cabinet of the transformer area includes one or more of the following: intelligent circuit breaker, reactive power compensation device, static var generator, and phase-changing switch.

[0019] The present invention also provides an artificial intelligence-based distribution network simulation test device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the above-mentioned artificial intelligence-based distribution network simulation test control method.

[0020] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described artificial intelligence-based distribution network simulation test control method.

[0021] Compared with the prior art, the advantages of the present invention are as follows: This invention improves the efficiency of real-world power distribution network testing by introducing artificial intelligence learning based on actual production data and an intelligent closed loop of "generation-execution-evaluation-optimization". It can automatically generate complex operation and fault scenarios that are highly realistic in terms of statistical characteristics and dynamic behavior, completely eliminating the reliance on manual preset experience. It realizes the automation, intelligence and continuous self-optimization of test scenario construction, thereby greatly enhancing the reliability, coverage and engineering practical value of test results. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating an embodiment of the present invention.

[0023] Figure 2 This is a schematic diagram of the system structure according to an embodiment of the present invention. Detailed Implementation

[0024] The present invention will be further described below with reference to the accompanying drawings and specific preferred embodiments, but this does not limit the scope of protection of the present invention.

[0025] The technical solution adopted in this embodiment is as follows: Figure 1 As shown, it includes the following steps: S1, obtain real historical operation data from the actual power distribution production business system; S2, The artificial intelligence model is trained based on the real historical operating data, and a control scheme for simulating the target power distribution operation scenario is generated according to the topology of the power distribution network real-type test platform. S3, The control scheme is sent to the analysis and control system of the real test platform. The analysis and control system controls the operation of each functional cabinet and controllable equipment in the power distribution network real test platform according to the control scheme to generate the target power distribution operation scenario. S4, Collect the actual operating data generated by the power distribution network real-type test platform when running the target power distribution operation scenario; S5, input the actual operating data into the trained artificial intelligence model to evaluate the degree of conformity between the actual operating data and the expected characteristics of the target power distribution operation scenario; S6. If the compliance level meets the preset requirements, the current control scheme and operating status are saved; if the compliance level does not meet the preset requirements, the artificial intelligence model corrects the control scheme based on the evaluation results and returns to step S3.

[0026] In a specific application example, the power distribution network full-scale test platform includes a 10 kV section and a power distribution substation section; the objects controlled in step S3 include: The 10 kV section includes power supply cabinets, switchgear cabinets, fault cabinets, and impedance cabinets. Distribution transformers, distribution function cabinets, and adjustable load function cabinets in the distribution area; The distribution cabinet in the transformer area includes one or more of the following: intelligent distribution box, reactive power compensation cabinet, fault simulation cabinet, and line impedance simulation cabinet.

[0027] In a specific application example, the adjustable load function cabinet for the distribution area is an adjustable RLC load cabinet.

[0028] In a specific application example, in step S2, the learning objectives of the artificial intelligence model include the statistical distribution and time-series correlation patterns of voltage, current, power, load characteristics, and fault waveform features in real power distribution network operation.

[0029] In a specific application example, step S5, the assessment of the degree of conformity includes: comparing the feature distribution of the actual running data with the feature distribution of the corresponding scene type in the real historical running data to determine the similarity.

[0030] Compared with the prior art, the technical solution of this embodiment has the following significant technical advantages: 1. Achieved high-fidelity and high-diversity generation of test scenarios: Traditional methods rely on manual scripts or simple replays, making it difficult to simulate the complex random characteristics of real power grids. This method utilizes artificial intelligence to deeply mine and learn from massive amounts of real historical operating data, enabling it to capture deep-seated patterns such as load fluctuations, fault characteristics, and time-series correlations. Therefore, the system can automatically generate test scenarios that highly match the actual situation in terms of probability distribution, fluctuation patterns, and event combinations, including rare or extreme "edge cases," solving the core problems of "distortion" and "singularity" in test scenarios.

[0031] 2. Automated and intelligent transformation of the testing process: This method liberates test engineers from the tedious and repetitive task of scenario design. Users only need to set the test objective (such as "simulating a single-phase high-resistance fault in a low-resistance grounding system"), and the artificial intelligence module can automatically generate a complete control scheme and coordinate the execution of all equipment. This significantly lowers the testing threshold and reduces reliance on expert experience, greatly improving the efficiency of scenario construction and test execution, making it possible to conduct rapid and batch testing on a massive number of possible operating conditions.

[0032] 3. A smart closed loop of "perception-decision-execution-evaluation-optimization" is constructed: This is the core of this method. The system can not only automatically execute tests, but also collect output data in real time through sensors and feed it back to the artificial intelligence module for automatic evaluation. If the simulation results do not match the learned real features, the artificial intelligence module will autonomously correct the control scheme and start a new round of testing until the preset fidelity requirements are met. This closed loop enables the system to have the ability to learn and correct itself, and the quality of the test scenarios can be continuously improved with iteration, realizing the "autonomous evolution" of the real-world testing system.

[0033] 4. Enhanced validity and engineering value of test results: Because the test scenarios are driven by real data and validated through closed-loop optimization, the performance of power distribution equipment, the action logic of protection algorithms, and the effectiveness of control strategies verified under this environment have extremely high credibility and field portability. This greatly reduces the unknown risks caused by insufficient testing when applying new products and algorithms in actual power grids, providing strong empirical support for the reliable implementation of new power distribution network technologies, and has significant practical engineering value.

[0034] This embodiment also includes an artificial intelligence-based distribution network full-scale testing system, used to implement the aforementioned artificial intelligence-based distribution network full-scale testing control method, such as... Figure 2 As shown, it includes: The data access module is used to connect to the actual power distribution production business system to obtain real historical operation data; The artificial intelligence module is used to receive and learn the real historical operating data, generate and output a control scheme for controlling the distribution network real-type test platform to simulate the target operating scenario; The analysis and control module is communicatively connected to the artificial intelligence module, and is used to receive the control scheme and generate specific equipment control instructions based on the scheme; The full-scale testing platform includes a primary electrical connection network and a secondary communication connection network. The primary electrical connection network contains multiple functional cabinets and controllable devices, which are used to generate physical electrical scenarios under the instructions of the analysis and control module. The secondary communication connection network is used to connect the functional cabinets, controllable devices and the analysis and control module. The artificial intelligence module is also used to receive the actual operating data generated by the real-world testing platform during operation, and to evaluate the conformity of the data with the expected goals and make adjustments accordingly.

[0035] In specific application examples, the real-type test platform includes a 10 kV section and a transformer substation section; The primary electrical connection network of the 10 kV section includes a power supply function cabinet, a switchgear cabinet, a fault function cabinet, and an impedance function cabinet connected in sequence. The primary electrical connection network of the transformer substation is connected to the 10 kV substation via a high-voltage fuse, and includes a distribution transformer, as well as at least one primary transformer substation distribution function cabinet and a transformer substation adjustable load function cabinet connected in parallel.

[0036] In specific application examples, the distribution cabinet of the transformer area includes one or more of the following: intelligent circuit breaker, reactive power compensation device, static var generator, and phase-changing switch.

[0037] This embodiment also includes an artificial intelligence-based distribution network simulation test device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described artificial intelligence-based distribution network simulation test control method.

[0038] This embodiment also includes a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the above-described artificial intelligence-based distribution network real-world test control method.

[0039] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-readable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the process. Figure 1 One or more processes and / or boxes Figure 1The computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0040] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A power distribution network simulation test control method based on artificial intelligence, characterized in that, Includes the following steps: S1, obtain real historical operation data from the actual power distribution production business system; S2, The artificial intelligence model is trained based on the real historical operating data, and a control scheme for simulating the target power distribution operation scenario is generated according to the topology of the power distribution network real-type test platform. S3, The control scheme is sent to the analysis and control system of the real test platform. The analysis and control system controls the operation of each functional cabinet and controllable equipment in the power distribution network real test platform according to the control scheme to generate the target power distribution operation scenario. S4, Collect the actual operating data generated by the power distribution network real-type test platform when running the target power distribution operation scenario; S5, input the actual operating data into the trained artificial intelligence model to evaluate the degree of conformity between the actual operating data and the expected characteristics of the target power distribution operation scenario; S6. If the compliance level meets the preset requirements, the current control scheme and operating status are saved; if the compliance level does not meet the preset requirements, the artificial intelligence model corrects the control scheme based on the evaluation results and returns to step S3.

2. The artificial intelligence-based distribution network simulation test control method according to claim 1, characterized in that, The power distribution network full-scale test platform includes a 10 kV section and a power distribution substation section; the objects controlled in step S3 include: The 10 kV section includes power supply cabinets, switchgear cabinets, fault cabinets, and impedance cabinets. Distribution transformers, distribution function cabinets, and adjustable load function cabinets in the distribution area; The distribution cabinet in the transformer area includes one or more of the following: intelligent distribution box, reactive power compensation cabinet, fault simulation cabinet, and line impedance simulation cabinet.

3. The artificial intelligence-based distribution network simulation test control method according to claim 2, characterized in that, The adjustable load function cabinet in the distribution area is an adjustable RLC load cabinet.

4. The artificial intelligence-based distribution network simulation test control method according to claim 1, characterized in that, In step S2, the learning objectives of the artificial intelligence model include the statistical distribution and temporal correlation patterns of voltage, current, power, load characteristics, and fault waveform features in real power distribution network operation.

5. The artificial intelligence-based distribution network simulation test control method according to claim 1, characterized in that, In step S5, the assessment of the degree of conformity includes: comparing the feature distribution of the actual operating data with the feature distribution of the corresponding scene type in the real historical operating data to determine the similarity.

6. A power distribution network simulation test system based on artificial intelligence, used to implement the control method according to any one of claims 1-5, characterized in that, include: The data access module is used to connect to the actual power distribution production business system to obtain real historical operation data; The artificial intelligence module is used to receive and learn the real historical operating data, generate and output a control scheme for controlling the distribution network real-type test platform to simulate the target operating scenario; The analysis and control module is communicatively connected to the artificial intelligence module, and is used to receive the control scheme and generate specific equipment control instructions based on the scheme; The full-scale testing platform includes a primary electrical connection network and a secondary communication connection network. The primary electrical connection network contains multiple functional cabinets and controllable devices, which are used to generate physical electrical scenarios under the instructions of the analysis and control module. The secondary communication connection network is used to connect the functional cabinets, controllable devices and the analysis and control module. The artificial intelligence module is also used to receive the actual operating data generated by the real-world testing platform during operation, and to evaluate the conformity of the data with the expected goals and make adjustments accordingly.

7. The artificial intelligence-based distribution network simulation testing system according to claim 6, characterized in that, The full-scale test platform includes a 10 kV section and a transformer substation section; The primary electrical connection network of the 10 kV section includes a power supply function cabinet, a switchgear cabinet, a fault function cabinet, and an impedance function cabinet connected in sequence. The primary electrical connection network of the transformer substation is connected to the 10 kV substation via a high-voltage fuse, and includes a distribution transformer, as well as at least one primary transformer substation distribution function cabinet and a transformer substation adjustable load function cabinet connected in parallel.

8. The artificial intelligence-based distribution network simulation testing system according to claim 7, characterized in that, The distribution cabinet of the transformer area includes one or more of the following: intelligent circuit breaker, reactive power compensation device, static var generator, and phase-changing switch.

9. A power distribution network simulation testing device based on artificial intelligence, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the artificial intelligence-based distribution network simulation test control method as described in any one of claims 1 to 5.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the artificial intelligence-based distribution network simulation test control method as described in any one of claims 1 to 5.

Images