On-line decision optimization management system and management method for phase-change switch

Abstract

The invention discloses a commutation switch online decision optimization management system and management method, and relates to the technical field of commutation switch control management, and the system comprises a deep learning model module, an online decision optimization module, a power grid state prediction module, a commutation switch control module, a quality detection module and a feedback adjustment module. According to the on-line decision optimization management system for the phase-change switch, through intelligent and automatic decision optimization, the operation stability and efficiency of a power grid are improved; the system can dynamically adapt to the change of a power grid, carries out the training and optimization of a deep learning model through the real-time data and historical data, generates an optimal phase-change switch operation strategy, effectively reduces the fault rate and the electric energy loss, has the self-learning and optimization capability through the multi-dimensional information collection, real-time decision and feedback adjustment, and is suitable for large-scale popularization and application. Power grid load fluctuation can be quickly responded, and energy efficiency and safety are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of phase-changing switch control, and in particular to an online decision-making optimization management system and a management method for phase-changing switches. Background Art

[0002] With the continuous development of power systems and the advancement of intelligentization, the requirements for the efficiency, stability and reliability of power distribution are becoming increasingly higher. As one of the important equipment in the power system, the reasonable operation of phase-changing switches plays a key role in improving power quality, reducing losses and improving system stability. Traditional phase-changing switch operations mainly rely on empirical rules and simple algorithms. These methods are often difficult to achieve optimal decisions when faced with complex and changeable power system operating conditions; for example, under different load conditions, changes in grid topology, and distributed energy access.

[0003] Most traditional methods cannot accurately predict the optimal operation time and mode of phase-changing switches, which leads to a decline in power quality, increased losses and affected system stability. Moreover, with the expansion of the scale of the power grid and the complexity of the structure, the operation frequency and demand of phase-changing switches continue to increase, while the existing manual operation or automatic control based on fixed rules is difficult to adapt to the dynamically changing power grid operation environment, which can easily lead to voltage fluctuations, load imbalance or local overload problems, thereby affecting the stability and energy efficiency of the power grid. Moreover, the existing phase-changing switch control system cannot make full use of real-time power grid data for decision optimization, and lacks comprehensive analysis of multi-dimensional power grid status information such as voltage, current, and power, resulting in the operation strategy of the phase-changing switch often lagging behind and unable to respond quickly to sudden situations in the power grid. The traditional phase-changing switch operation strategy lacks flexibility and cannot be optimized according to the real-time load changes and historical data of the power grid, which can easily lead to a decline in power grid energy efficiency or frequent phase-changing operations, increasing the cost and risk of power grid operation. Summary of the invention

[0004] The present invention provides a phase-changing switch online decision-making optimization management system and management method, so as to realize efficient and accurate online decision-making optimization management of the phase-changing switch.

[0005] According to one aspect of the present invention, there is provided an online decision-making optimization management system for phase-changing switches, comprising:

[0006] A deep learning model module is used to make predictions based on the historical and real-time data of the power grid to obtain the corresponding commutation switch operation strategy;

[0007] The power grid state prediction module is used to make short-term and long-term predictions of the future state of the power grid based on the historical data and real-time data of the power grid, identify potential faults that may occur, and obtain the power grid state prediction results;

[0008] An online decision optimization module, used to generate an optimal decision instruction by combining the historical data of the power grid, the real-time data, the phase-changing switch operation strategy and the power grid state prediction result;

[0009] A phase-changing switch control module, used for operating the phase-changing switch based on the preferred decision instruction;

[0010] A quality detection module, used to perform real-time detection on the state of the power grid after the phase-changing switch operation to obtain detection information, and to evaluate the quality of the phase-changing switch operation based on the detection information and generate an evaluation result;

[0011] A feedback adjustment module is used to dynamically adjust and optimize the deep learning model module and the online decision optimization module based on the detection information and the evaluation results.

[0012] According to another aspect of the present invention, a phase-changing switch online decision optimization management method is provided, comprising:

[0013] Based on the historical data and real-time data of the power grid, the prediction is carried out to obtain the corresponding phase-changing switch operation strategy;

[0014] Based on the historical and real-time data of the power grid, short-term and long-term predictions are made on the future state of the power grid, and potential faults that may occur are identified to obtain the prediction results of the power grid state;

[0015] Generate an optimal decision instruction by combining the historical data and real-time data of the power grid, the phase-changing switch operation strategy and the power grid state prediction result;

[0016] Based on the preferred decision instruction, operating the phase-changing switch;

[0017] Performing real-time detection on the state of the power grid after the phase-changing switch operation and generating detection information, and evaluating the quality of the phase-changing switch operation based on the detection information and generating an evaluation result;

[0018] The phase-changing switch operation strategy and the preferred decision instruction are dynamically adjusted and optimized based on the detection information and the evaluation result.

[0019] The technical solution of the embodiment of the present invention improves the operational stability and efficiency of the power grid through intelligent and automated decision optimization; the system can dynamically adapt to changes in the power grid, use real-time data and historical data to train and optimize deep learning models, generate optimal phase-changing switch operation strategies, and effectively reduce failure rates and power losses; through multi-dimensional information collection, real-time decision-making and feedback adjustment, the system has the ability of self-learning and optimization, can quickly respond to grid load fluctuations, and improve energy efficiency and safety; in addition, the distributed architecture enables the system to adapt to the multi-regional management needs of large-scale power grids and ensure the coordination and reliability of the overall power grid; the system also supports historical data backtracking and manual intervention, which enhances the transparency and flexibility of operations and significantly reduces operational risks. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 This is an architecture diagram of an online decision-making optimization management system for phase-changing switches provided in the first embodiment of the present invention;

[0022] Figure 2 A diagram of a distributed deployment architecture provided in Embodiment 1 of the present invention;

[0023] Figure 3 This is an architectural diagram of a user interface module provided in Embodiment 1 of the present invention;

[0024] Figure 4 This is a flow chart of an online decision-making optimization management method for a phase-changing switch provided in the second embodiment of the present invention. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0027] Embodiment 1

[0028] Figure 1 The following is an architecture diagram of an online decision-making optimization management system for a phase-changing switch provided in the first embodiment of the present invention. An online decision-making optimization management system for a phase-changing switch includes:

[0029] The deep learning model module is used to make predictions based on the historical data and real-time data of the power grid to obtain the phase-changing switch operation strategy corresponding to the phase-changing switch.

[0030] Specifically, the deep learning model module can be trained based on the historical data and real-time data of the power grid to construct a deep learning model that can automatically adapt to the dynamic changes of the power grid; the deep learning model continuously optimizes the operation strategy by learning the impact of the phase-changing switch operation on the power grid state, and finally the historical data and real-time data of the power grid can be input into the model for prediction to obtain the phase-changing switch operation strategy corresponding to the phase-changing switch.

[0031] The grid state prediction module is used to make short-term and long-term predictions of the future state of the grid based on the historical data and real-time data of the grid, identify potential faults that may occur, and obtain grid state prediction results.

[0032] Specifically, the grid state prediction module makes short-term and long-term predictions on the future state of the grid based on the historical data and real-time data of the grid, identifies possible grid instability factors or potential failures, and provides the online decision-making optimization module with prediction results of grid operation, namely, grid state prediction results.

[0033] The online decision optimization module is used to generate an optimal decision instruction by combining the historical data and real-time data of the power grid, the phase-changing switch operation strategy and the power grid state prediction result.

[0034] Specifically, in order to further optimize the phase-changing switch operation strategy predicted by the deep learning model module, the real-time data of the power grid, historical data, the phase-changing switch operation strategy and the power grid status prediction results can be combined to dynamically generate optimal decision instructions to ensure the stable operation of the power grid under load changes and fault conditions.

[0035] In an embodiment of the present invention, the phase-changing switch online decision optimization management system may also include a dynamic information acquisition module, which can collect real-time data from sensors in the power grid, and the real-time data includes voltage, current, power, and temperature, and provide the real-time data to the deep learning model module and the online decision optimization module as a basis for decision-making.

[0036] The phase-changing switch control module is used to operate the phase-changing switch based on the preferred decision instructions output by the online decision optimization module to ensure that the power grid can adapt to different operating conditions and load changes.

[0037] The quality detection module is used to perform real-time detection of the power grid state after the phase-changing switch operation, evaluate the quality of the phase-changing switch operation, and generate corresponding detection information and evaluation results.

[0038] The feedback adjustment module can dynamically adjust and optimize the deep learning model module and the power grid status prediction module based on the detection information and evaluation results provided by the quality detection module, thereby continuously improving the decision-making accuracy and response speed of the system.

[0039] Through the collaborative work of the above modules, the system can realize online decision-making optimization of power grid phase-changing switches and improve the operation stability and efficiency of the power grid.

[0040] It should be further explained that, in the specific implementation process, through the adaptive optimization of the deep learning model, the system can identify the nonlinear dynamic characteristics of the power grid, continuously improve the operation strategy, and ensure the generation of the optimal phase-changing switch operation strategy under different load conditions; this not only improves the accuracy of the phase-changing switch operation, but also reduces the necessity of human intervention.

[0041] In an embodiment of the present invention, the deep learning model module includes:

[0042] The data preprocessing unit is used to clean, denoise and normalize the historical data and real-time data in the power grid, and convert the historical data and real-time data into a standardized input data set. The model training unit is used to train the neural network model based on the standardized input data set to obtain a deep learning model for predicting the phase-changing switch operation strategy. The neural network can capture the nonlinear dynamic characteristics of the power grid operation in the standardized input data set, and optimize the deep learning model of the phase-changing switch operation through repeated training.

[0043] The model update unit is used to regularly update the deep learning model based on real-time feedback data and changes in the power grid status to ensure that the model can continue to adapt to the dynamic changes of the power grid.

[0044] The model verification unit is used to verify the trained deep learning model: the accuracy of the deep learning model's operational decisions under different grid conditions is evaluated through grid simulation and historical data backtesting, the verification results are obtained, and the deep learning model parameters are adjusted based on the verification results; through the deep learning model module, the system can adaptively optimize the phase-changing switch operation and improve the reliability and safety of grid operation.

[0045] It should be further explained that, in the specific implementation process, through the deep learning model's learning of the historical data and real-time data of the power grid, the system can adaptively optimize the phase-changing switch operation, automatically adapt to the dynamic changes of the power grid, improve the intelligence and accuracy of the operation strategy, and realize the system's function of dynamically adapting to changes in the power grid.

[0046] In an embodiment of the present invention, the power grid state prediction module includes:

[0047] The historical data analysis unit is used to analyze the historical data in the power grid, identify the periodic laws and abnormal conditions in the operation of the power grid, and provide a basis for predicting the future state of the power grid.

[0048] A state prediction model is used to predict the future trend of power grid operation and obtain the power grid state prediction result; wherein the state prediction model is a short-term and long-term power grid state prediction model trained based on the historical data, and the future trend of power grid operation includes voltage fluctuation, load change and potential faults.

[0049] The prediction correction unit is used to correct the state prediction model according to the actual operation state of the power grid and the power grid state prediction result.

[0050] The grid state prediction module helps the online decision optimization module to make more accurate operational decisions and improve the safety and reliability of grid operation by accurately predicting the future state of the grid.

[0051] It should be further explained that, in the specific implementation process, the system uses the state prediction model in the power grid state prediction module to make short-term and long-term predictions on the future power grid state, identify potential faults or unstable factors in the power grid in advance, improve the safety and stability of power grid operation, and realize power grid state prediction and prevention.

[0052] In an embodiment of the present invention, the online decision optimization module includes:

[0053] The optimization algorithm unit is used to combine historical data, real-time data, commutation switch operation strategy and power grid state prediction results, and generate candidate commutation switch operation strategy by using reinforcement learning algorithm. The candidate commutation switch operation strategy can be a strategy obtained by optimizing the commutation switch operation strategy output by the deep learning model module by using reinforcement learning algorithm.

[0054] The real-time decision-making unit is used to update the candidate commutation switch operation strategy in real time according to real-time data when the grid load fluctuates.

[0055] A decision evaluation unit is used to evaluate each generated candidate phase-changing switch operation decision, analyze the grid stability, energy efficiency and safety after the candidate phase-changing switch operation strategy is operated, and select the candidate phase-changing switch operation strategy whose grid stability, energy efficiency and safety meet the preset evaluation criteria as the preferred decision instruction, and transmit the preferred decision instruction to the phase-changing switch control module.

[0056] Through the online decision optimization module, the system can quickly generate optimal decision instructions in a complex power grid operating environment.

[0057] It should be further explained that, in the specific implementation process, the online decision optimization module combines the historical data, real-time data, phase-changing switch operation strategy and grid state prediction results to dynamically generate the optimal decision instructions to ensure the stable operation of the grid under load changes and fault conditions; this module uses a reinforcement learning algorithm, which not only improves the accuracy of decision-making, but also enhances the real-time nature of decision-making, thereby realizing the real-time nature of decision optimization.

[0058] In an optional implementation, the dynamic information collection module may include:

[0059] The multi-dimensional sensor network deploys voltage sensors, current sensors, power sensors, and temperature sensors at each key node of the power grid to collect the operating status information of the power grid in real time and collect the real-time data corresponding to the sensors.

[0060] The data acquisition unit is used to obtain real-time data from the sensor network and transmit the real-time data to the deep learning model module and the online decision optimization module through wireless communication.

[0061] The data integration unit is used to integrate, calibrate and verify the real-time data from various sensors to ensure the accuracy and consistency of the real-time data of the sensors and provide high-quality data input for subsequent deep learning model training and online decision-making.

[0062] Through the dynamic information collection module, the system can fully and accurately grasp the real-time operating status of the power grid and provide data support for phase-changing switch operation decisions.

[0063] It should be further explained that, in the specific implementation process, the dynamic information acquisition module obtains voltage, current, power, temperature and other data in real time through the sensor network, and integrates and calibrates them to ensure the accuracy and consistency of the data, provide high-quality data support for deep learning models and optimized decisions, and realize comprehensive information collection functions.

[0064] In an embodiment of the present invention, the phase-changing switch control module includes:

[0065] A control instruction generating unit is used to use the preferred decision instruction as the control instruction of the phase-changing switch, and control the phase-changing switch to perform operations based on the control instruction; a switch status monitoring unit is used to monitor the operating status of the phase-changing switch in real time; and an emergency processing unit is used to trigger emergency processing measures if an abnormal situation is detected during the operation of the phase-changing switch.

[0066] Specifically, the control instruction generation unit can use the preferred decision instruction output by the online decision optimization module as the control instruction of the phase-changing switch, and send the control instruction to the phase-changing switch, and the phase-changing switch can perform corresponding operations according to the instruction. The switch status monitoring unit can monitor the operating status of the phase-changing switch in real time to ensure that the phase-changing switch performs the operation correctly according to the control instruction. The emergency processing unit is used to trigger emergency processing measures when an abnormal situation is detected during the operation of the phase-changing switch to ensure the safe operation of the power grid.

[0067] Through the phase-changing switch control module, the system can perform phase-changing operations safely and accurately and adapt to changes in different grid operating conditions.

[0068] It should be further explained that, in the specific implementation process, through the phase-changing switch control module, the system can accurately perform phase-changing operations according to the preferred decision instructions, and when an abnormality is detected, emergency measures can be triggered in time to ensure the safe operation of the power grid and realize the real-time quality detection and feedback optimization functions of the system.

[0069] The quality detection module includes: a power grid operation status detection unit, which is used to perform real-time detection of the operation status of the power grid after the phase-changing switch is operated, and record the changes in the power grid operation data, and then generate detection information; wherein the power grid operation data includes key indicators such as voltage, current, power, etc., and the detection information includes the stability, efficiency and troubleshooting information of the power grid operation; an operation effect evaluation unit, which evaluates the quality of the phase-changing switch operation based on the power grid status detection result, and generates an evaluation result, and the evaluation result includes the stability of the power grid, the degree of improvement of the power factor and the energy efficiency improvement; a fault diagnosis unit, which is used to diagnose faults occurring during the phase-changing switch operation when the power grid status is abnormal, and generate a fault report; through the quality detection module, the system can evaluate the effect of the phase-changing switch operation in real time to ensure the effectiveness of the operation strategy and the stability of the power grid.

[0070] It should be further explained that, in the specific implementation process, the quality detection module evaluates the grid state after the phase-changing switch operation in real time, and dynamically adjusts and optimizes the deep learning model and decision instructions through the feedback adjustment module to ensure that the decision-making accuracy and response speed of the system continue to improve; through the grid state prediction and quality detection modules, the system can identify potential faults in the grid in advance, take preventive measures, and reduce the failure rate and grid losses caused by improper phase-changing switch operation, thereby improving the overall stability and safety of the grid.

[0071] In an embodiment of the present invention, the phase-changing switch online decision optimization management system may further include a feedback adjustment module, which includes: a real-time feedback unit, which is used to obtain power grid operation data and detection information from the quality detection module, and transmit the power grid operation data and detection information to the deep learning model module and the online decision optimization module; a refresh unit, which is used to refresh and adjust the deep learning model based on real-time feedback data, and the refresh adjustment includes adjusting the weights, parameters and neural network structure of the deep learning model to adapt to the operation changes of the power grid; an optimization strategy adjustment unit, which is used to correct the preferred decision instructions of the online decision optimization module according to the real-time feedback data to ensure that the operation strategy of the phase-changing switch can be continuously optimized to meet the real-time needs of the power grid operation; a decision backtracking unit, which is used to perform backtracking analysis on the executed preferred decision instructions, and generate improvement suggestions and optimization directions for the operation strategy in combination with real-time feedback data and real-time power grid status, so as to further improve the accuracy of future decisions.

[0072] Through the feedback adjustment module, the system can continuously optimize itself and dynamically adapt to changes in the operating status of the power grid, thereby improving the accuracy of phase-changing switch operations and the overall stability of power grid operation.

[0073] like Figure 2The figure shows a distributed deployment architecture diagram provided in the first embodiment of the present invention. The online decision optimization management system for phase-changing switches supports distributed deployment, specifically including: a distributed deep learning model, which is used to independently model and optimize the operating status of different power grid areas, and generate preferred decision instructions in different power grid areas; a regionalized decision unit, which is used to independently generate operating instructions for phase-changing switches according to the operating conditions and load conditions of different power grid areas, to ensure that the phase-changing switch operation in each area can optimize the stability and efficiency of the power grid in a targeted manner; a central control unit, which is used to summarize and coordinate the preferred decision instructions, i.e., operating instructions, of the phase-changing switches in each area, to ensure the overall operational consistency and system security of the power grid, and to avoid local operations affecting the global power grid operation.

[0074] Through the distributed deployment architecture, the system can adapt to the multi-regional management needs of large-scale power grids, and achieve regional optimized control while ensuring overall stability.

[0075] It should be further explained that, in the specific implementation process, the system supports distributed deployment, and can independently optimize the phase-changing switch operation according to the actual operating status of different power grid areas, realize regional management, ensure the coordination and consistency of the overall power grid, and realize distributed management and flexible deployment of the system.

[0076] The phase-changing switch online decision optimization management system also includes a user interface module, which is used by the system operator to monitor, adjust and intervene in the operation of the phase-changing switch in real time, including: a visual monitoring unit, which is used to display the real-time operation status of the power grid, the phase-changing switch operation record, the decision optimization process and the power grid prediction results, and provide real-time data feedback to the operator through a graphical interface; a manual intervention unit, which is used by the operator to manually intervene in the phase-changing operation of the system through the interface, including temporarily modifying the operation strategy and forcibly adjusting the operation instructions of the phase-changing switch to deal with emergencies or urgent needs; a historical record query unit, which is used to provide a historical data query function for the phase-changing operation. The operator can view the past operation records, decision-making process and system feedback through the historical record query unit, evaluate the decision effect of the preferred decision instructions of the online decision optimization management system, and provide a basis for future optimization. Through the user interface module, the system can improve the operator's controllability and transparency of the power grid phase-changing switch operation, and enhance the flexibility and reliability of the actual application of the system.

[0077] It should be further explained that, in the specific implementation process, the user interface module provides operators with real-time data feedback, visual grid monitoring, decision-making process display and historical query functions, supports manual intervention, enhances the controllability of the system and the flexibility of operation, and enhances operator control and visual monitoring. Figure 3 , which is an architecture diagram of a user interface module provided in Embodiment 1 of the present invention.

[0078] The technical solution of the embodiment of the present invention provides an online decision-making optimization management system for phase-changing switches, which includes a deep learning model module, an online decision-making optimization module, a dynamic information collection module, a power grid state prediction module, a phase-changing switch control module, a quality detection module, and a feedback adjustment module. Through intelligent and automated decision optimization, the operation stability and efficiency of the power grid are improved; the system can dynamically adapt to changes in the power grid, use real-time data and historical data to train and optimize the deep learning model, generate the optimal phase-changing switch operation strategy, effectively reduce the failure rate and power loss, and through multi-dimensional information collection, real-time decision-making and feedback adjustment, the system has the ability of self-learning and optimization, can quickly respond to power grid load fluctuations, and improve energy efficiency and safety.

[0079] Embodiment 2

[0080] Figure 4 A flowchart of an online decision-making optimization management method for a phase-changing switch provided in Embodiment 2 of the present invention is provided. The method is applied to an online decision-making optimization management system for a phase-changing switch. The online decision-making optimization management system for a phase-changing switch includes: a deep learning model module, an online decision-making optimization module, a dynamic information acquisition module, a power grid state prediction module, a phase-changing switch control module, a quality detection module, and a feedback adjustment module; wherein the method includes:

[0081] S210: Predict based on historical data and real-time data of the power grid to obtain a phase-changing switch operation strategy corresponding to the phase-changing switch.

[0082] Specifically, training can be carried out based on the historical data and real-time data of the power grid to build a deep learning model that can automatically adapt to the dynamic changes of the power grid; the deep learning model continuously optimizes the operation strategy by learning the impact of the phase-changing switch operation on the power grid state. Ultimately, the historical data and real-time data of the power grid can be input into the model for prediction, and the phase-changing switch operation strategy corresponding to the phase-changing switch can be obtained.

[0083] S220: Based on the historical data and real-time data of the power grid, short-term and long-term predictions are made on the future state of the power grid, and potential faults that may occur are identified to obtain a power grid state prediction result.

[0084] Specifically, short-term and long-term predictions can be made on the future state of the power grid based on the historical and real-time data of the power grid, possible grid instability factors or potential failures can be identified, and the prediction results of the power grid operation, i.e., the grid state prediction results, can be provided to the online decision-making optimization module.

[0085] S230, generating a preferred decision instruction by combining the historical data and real-time data of the power grid, the phase-changing switch operation strategy and the power grid state prediction result.

[0086] Specifically, in order to further optimize the phase-changing switch operation strategy predicted by the deep learning model module, the real-time data of the power grid, historical data, the phase-changing switch operation strategy and the power grid status prediction results can be combined to dynamically generate optimal decision instructions to ensure the stable operation of the power grid under load changes and fault conditions.

[0087] Through the collaborative work of the above modules, the system can realize online decision-making optimization of power grid phase-changing switches and improve the operation stability and efficiency of the power grid.

[0088] On the basis of the above scheme, the preferred decision instruction is generated by combining the historical data, real-time data, the phase-changing switch operation strategy and the grid state prediction result of the power grid, which may include: combining the historical data, the real-time data, the phase-changing switch operation strategy and the grid state prediction result, using a reinforcement learning algorithm to generate candidate phase-changing switch operation strategies; when the grid load fluctuates, updating the candidate phase-changing switch operation strategy in real time according to the real-time data; evaluating each of the generated candidate phase-changing switch operation decisions, analyzing the grid stability, energy efficiency and safety after the candidate phase-changing switch operation strategy is operated, and selecting the candidate phase-changing switch operation strategy whose grid stability, energy efficiency and safety meet the preset evaluation criteria as the preferred decision instruction.

[0089] S240. Based on the preferred decision instruction, operate the phase-changing switch.

[0090] Specifically, the preferred decision instruction can be sent to the phase-changing switch, and the phase-changing switch can perform corresponding operations according to the instruction. In addition, the operating status of the phase-changing switch can be monitored in real time to ensure that the phase-changing switch performs the operation correctly according to the control instruction. When an abnormal situation is detected during the operation of the phase-changing switch, emergency treatment measures are triggered to ensure the safe operation of the power grid.

[0091] By controlling the commutation switches and financial control, the system can safely and accurately perform commutation operations and adapt to changes in different grid operating conditions. The system can accurately perform commutation operations according to the optimal decision instructions, and when abnormalities are detected, emergency measures are triggered in a timely manner to ensure the safe operation of the grid and realize the system's real-time quality detection and feedback optimization functions.

[0092] S250, performing real-time detection on the grid state after the phase-changing switch operation and generating detection information, and evaluating the quality of the phase-changing switch operation based on the detection information and generating an evaluation result.

[0093] During the specific implementation process, the grid status after the phase-changing switch operation can be evaluated in real time, and the deep learning model and decision-making instructions can be dynamically adjusted and optimized to ensure that the decision-making accuracy and response speed of the system continue to improve; through grid status prediction and quality detection, the system can identify potential faults in the grid in advance, take preventive measures, and reduce the failure rate and grid losses caused by improper phase-changing switch operation, thereby improving the overall stability and safety of the grid.

[0094] S260: Dynamically adjust and optimize the phase-changing switch operation strategy and the preferred decision instruction based on the detection information and the evaluation result.

[0095] Specifically, based on the detection information and evaluation results, the phase-changing switch operation strategy and the preferred decision instructions are dynamically adjusted and optimized, which can dynamically adapt to the changes in the operating state of the power grid, improve the accuracy of the phase-changing switch operation and the overall stability of the power grid operation. The technical solution of the embodiment of the present invention provides an online decision-making optimization management system for phase-changing switches, which includes a deep learning model module, an online decision-making optimization module, a dynamic information acquisition module, a power grid state prediction module, a phase-changing switch control module, a quality detection module and a feedback adjustment module. Through intelligent and automated decision optimization, the operating stability and efficiency of the power grid are improved; the system can dynamically adapt to changes in the power grid, use real-time data and historical data to train and optimize the deep learning model, generate the optimal phase-changing switch operation strategy, effectively reduce the failure rate and power loss, and through multi-dimensional information collection, real-time decision-making and feedback adjustment, the system has the ability of self-learning and optimization, and can quickly respond to power grid load fluctuations and improve energy efficiency and safety.

[0096] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0097] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A phase-changing switch online decision-making optimization management system, characterized in that: include: A deep learning model module is used to make predictions based on the historical and real-time data of the power grid to obtain the corresponding commutation switch operation strategy; The power grid state prediction module is used to make short-term and long-term predictions of the future state of the power grid based on the historical data and real-time data of the power grid, identify potential faults that may occur, and obtain the power grid state prediction results; An online decision optimization module, used to generate an optimal decision instruction by combining the historical data of the power grid, the real-time data, the phase-changing switch operation strategy and the power grid state prediction result; A phase-changing switch control module, used for operating the phase-changing switch based on the preferred decision instruction; A quality detection module, used to perform real-time detection on the state of the power grid after the phase-changing switch operation to obtain detection information, and to evaluate the quality of the phase-changing switch operation based on the detection information and generate an evaluation result; A feedback adjustment module is used to dynamically adjust and optimize the deep learning model module and the online decision optimization module based on the detection information and the evaluation results.

2. The system according to claim 1, characterized in that The deep learning model module includes: A data preprocessing unit, used to clean, denoise and normalize the historical data and the real-time data, and uniformly convert the historical data and the real-time data into a standardized input data set; A model training unit, used for training a neural network model based on the standardized input data set to obtain a deep learning model for predicting the commutation switch operation strategy; A model updating unit, configured to update the deep learning model based on real-time feedback data and changes in power grid status; A model verification unit is used to verify the trained deep learning model: the operational decision accuracy of the deep learning model under different power grid conditions is evaluated through power grid simulation and historical data backtesting, verification results are obtained, and the deep learning model parameters are adjusted based on the verification results.

3. The system according to claim 1, characterized in that The power grid state prediction module comprises: A historical data analysis unit, used to analyze the historical data in the power grid, identify periodic laws and abnormal conditions in the operation of the power grid, and provide a basis for predicting the future state of the power grid; A state prediction model is used to predict the future trend of the power grid operation and obtain the power grid state prediction result; wherein the state prediction model is a short-term and long-term power grid state prediction model obtained by training based on the historical data; The prediction correction unit is used to correct the state prediction model according to the actual operation state of the power grid and the power grid state prediction result.

4. The system according to claim 1, characterized in that The online decision optimization module includes: an optimization algorithm unit, configured to generate a candidate phase-changing switch operation strategy by using a reinforcement learning algorithm in combination with the historical data, the real-time data, the phase-changing switch operation strategy and the grid state prediction result; A real-time decision-making unit, used for updating the candidate commutation switch operation strategy in real time according to real-time data when the grid load fluctuates; A decision evaluation unit is used to evaluate each generated candidate phase-changing switch operation decision, analyze the grid stability, energy efficiency and safety after the candidate phase-changing switch operation strategy is operated, and select the candidate phase-changing switch operation strategy whose grid stability, energy efficiency and safety meet the preset evaluation criteria as the preferred decision instruction, and transmit the preferred decision instruction to the phase-changing switch control module.

5. The system according to claim 1, characterized in that The phase-changing switch control module comprises: a control instruction generating unit, configured to use the preferred decision instruction as a control instruction of the phase-changing switch, and control the phase-changing switch to perform an operation based on the control instruction; A switch status monitoring unit, used for monitoring the operating status of the phase-changing switch in real time; The emergency processing unit is used to trigger emergency processing measures if an abnormal situation is detected during the operation of the phase-changing switch.

6. The system according to claim 1, characterized in that The quality detection module comprises: A power grid operation status detection unit, used to detect the power grid operation status data in real time after the phase-changing switch performs an operation, and generate detection information; wherein the power grid operation data includes voltage, current, and power, and the detection information includes the stability, efficiency, and troubleshooting information of the power grid operation; An operation effect evaluation unit, used to evaluate the quality of the phase-changing switch operation based on the detection information and generate an evaluation result; wherein the evaluation result includes the stability of the power grid, the degree of improvement of the power factor and the improvement of energy efficiency; The fault diagnosis unit is used to diagnose the faults occurring during the operation of the phase-changing switch and generate a fault report when the power grid is in an abnormal state.

7. The system according to claim 6, characterized in that It also includes a feedback adjustment module, which includes: A real-time feedback unit, used for transmitting the power grid operation status data and the detection information to the deep learning model module and the online decision optimization module; A refresh unit, configured to refresh and adjust the deep learning model according to the power grid operation status data and the detection information, wherein the refresh and adjustment includes adjusting the weights, parameters and neural network structure of the deep learning model; An optimization strategy adjustment unit, configured to modify the preferred decision instruction of the online decision optimization module according to the power grid operation status data and the detection information; The decision backtracking unit is used to perform backtracking analysis on the executed preferred decision instructions, and generate improvement suggestions and optimization directions for operation strategies based on real-time feedback data and real-time power grid status.

8. The system according to claim 1, characterized in that The phase-changing switch online decision optimization management system supports distributed deployment, specifically including: Distributed deep learning model, used to independently model and optimize the operating status of different power grid areas, and generate optimal decision instructions in different power grid areas; Regional decision-making unit, used to independently generate operation instructions for phase-changing switches according to the operating conditions and load conditions of different power grid areas; The central control unit is used to aggregate and coordinate the optimal decision instructions of the phase-changing switches in each area.

9. A phase-changing switch online decision optimization management method, characterized in that: include: Based on the historical data and real-time data of the power grid, the prediction is carried out to obtain the corresponding phase-changing switch operation strategy; Based on the historical and real-time data of the power grid, short-term and long-term predictions are made on the future state of the power grid, and potential faults that may occur are identified to obtain the prediction results of the power grid state; Generate an optimal decision instruction by combining the historical data and real-time data of the power grid, the phase-changing switch operation strategy and the power grid state prediction result; Based on the preferred decision instruction, operating the phase-changing switch; Performing real-time detection on the state of the power grid after the phase-changing switch operation and generating detection information, and evaluating the quality of the phase-changing switch operation based on the detection information and generating an evaluation result; The phase-changing switch operation strategy and the preferred decision instruction are dynamically adjusted and optimized based on the detection information and the evaluation result.

10. The method according to claim 9, characterized in that The combining of the historical data and real-time data of the power grid, the phase-changing switch operation strategy and the power grid state prediction result to generate the optimal decision instruction includes: Combining the historical data, the real-time data, the phase-changing switch operation strategy and the grid state prediction result, generating a candidate phase-changing switch operation strategy using a reinforcement learning algorithm; When the grid load fluctuates, the candidate commutation switch operation strategy is updated in real time according to the real-time data; Each generated candidate commutation switch operation decision is evaluated, the grid stability, energy efficiency and safety after the candidate commutation switch operation strategy is operated are analyzed, and the candidate commutation switch operation strategy whose grid stability, energy efficiency and safety meet the preset evaluation criteria is selected as the preferred decision instruction.

Images