Distributed power supply intelligent grid-connected evaluation scheduling method and system based on multi-source data fusion
By integrating data from distributed power sources and grid nodes and employing a multi-stage scheduling model, the problems of inaccurate assessment and delayed response in existing technologies are solved, enabling high-precision assessment and rapid response of grid status, thus ensuring grid stability and power quality.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-17
- Publication Date
- 2026-04-07
AI Technical Summary
Existing systems lack the ability to perform real-time fusion and in-depth analysis of multi-source data, resulting in inaccurate assessments of power grid operation status, particularly in the assessment of stability and power quality. Furthermore, the response is delayed, making it difficult to provide rapid and real-time policy support for multi-stage, cross-timescale scheduling decisions.
By collecting raw operating data from distributed power sources and grid nodes, preprocessing and data fusion are performed to generate a high-precision fused dataset. A multi-stage scheduling model is then used for grid status prediction and risk assessment. Scheduling strategies for the initialization, dynamic adjustment, and emergency response stages are constructed to achieve rapid response and accurate assessment of the grid.
It improves the accuracy of power grid assessment and the timeliness of response, reduces the deviation of stability and power quality assessment, and ensures that the power grid can take effective measures quickly in the face of sudden fluctuations, avoiding response lag.
Smart Images
Figure CN121813530A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to a distributed power intelligent grid-connected scheduling method, in particular to a distributed power intelligent grid-connected evaluation scheduling method and system based on multi-source data fusion. BACKGROUND
[0002] With the rapid development of renewable energy and distributed power, after the power grid connects a large number of distributed power, it faces the challenges of power quality and safety, and the efficient collaborative operation of distributed power and power grid becomes a key problem to ensure the stability of the power grid.
[0003] By analyzing the influence of large-scale distributed power access on multiple operation indexes of the distribution network, an evaluation system model is established, the influence factors of distributed power access on the distribution network are refined by using specific indexes, and the index layer data of the fuzzy evaluation model is mapped to the evaluation set by using the membership degree theory. The severity of the influence of distributed power access on the distribution network is comprehensively reflected through the evaluation results, and the risk level is evaluated by using the severity data, so as to prewarn the high-level risk.
[0004] Although this scheduling method can provide a basis for distributed power grid-connected operation control and operation inspection safety protection, it still has the following defects:
[0005] 1. The existing system has insufficient real-time fusion and deep analysis capabilities for multi-source data (such as weather information, load prediction, power generation output, etc.), which leads to deviations and inaccuracies in the evaluation of the power grid operation state, especially in the evaluation of stability and power quality, resulting in inaccurate evaluation.
[0006] 2. Due to the long data processing chain and complex calculation model, the existing system is difficult to provide fast and real-time strategy support for multi-stage and cross-time scale scheduling decisions, and the response is slow when facing power grid sudden fluctuations, resulting in response lag.
[0007] The information disclosed in this background section is intended only to increase an understanding of the general background of the application, and should not be construed as recognizing or admitting that this information constitutes prior art that is already known to those of ordinary skill in the art. SUMMARY
[0008] The purpose of the present application is to overcome the shortcomings of inaccurate evaluation and slow response in the prior art, and to provide a distributed power intelligent grid-connected evaluation scheduling method and system based on multi-source data fusion, which is accurate in evaluation and timely in response.
[0009] To achieve the above purpose, the technical solution of the present application is:
[0010] A method for intelligent grid-connected evaluation and scheduling of distributed power sources based on multi-source data fusion, the evaluation and scheduling method comprising the following steps:
[0011] S1. Collect raw operating data from distributed power sources and grid nodes, preprocess and fuse the raw operating data to generate a fused dataset;
[0012] S2. Based on the fused dataset, analyze the impact of distributed power generation on the grid operation status, obtain evaluation indicators, and combine the evaluation indicators with the grid operation status to make predictions, obtain short-term predicted values of the grid status, and combine the short-term predicted values with the evaluation indicators to evaluate the potential stability or safety risks of distributed power generation grid connection operation, and obtain the risk level of distributed power generation grid connection operation.
[0013] S3. Construct a multi-stage scheduling model, which includes an initialization stage, a dynamic adjustment stage, and an emergency response stage. The multi-stage scheduling model selects the current corresponding scheduling stage based on the current power grid operating status, evaluation indicators, and short-term predicted values of the power grid status, and generates the corresponding scheduling strategy.
[0014] S4. Dispatch the power grid according to the dispatch strategy, collect power grid response data to form feedback information, obtain the current power grid operating status based on the feedback information, and return to S2 to re-evaluate the grid connection based on the current power grid operating status.
[0015] In S1, the original operating data includes synchronous operating data from different power sources and grid nodes. The synchronous operating data includes voltage, current, active power, reactive power, and environmental parameters. The preprocessing step includes data cleaning, anomaly detection, and normalization of the original operating data, and outputting the fused dataset to the grid-connected evaluation module and the multi-stage scheduling module.
[0016] The data fusion step includes performing time-series alignment and fusion on the preprocessed data to generate the fused dataset, and performing state estimation on the fused dataset to improve the accuracy of the evaluation results of the grid connection evaluation module.
[0017] In step S2, the evaluation indicators include voltage fluctuation, frequency offset, and power flow distribution. Predicting the future operating state of the power grid using these evaluation indicators and the grid's operating status includes:
[0018] ;
[0019] In the above formula, This is a short-term prediction of the voltage. The voltage is the current voltage, and η is the adjustment coefficient. The rate of change of voltage is used to obtain the predicted value of voltage fluctuation;
[0020] The current frequency and power flow are predicted using the Newton-Raphson method, and the predicted values of frequency shift and power flow distribution are obtained.
[0021] The risk levels of the intelligent grid connection of distributed power sources include voltage fluctuation risk level, frequency offset risk level and power flow distribution risk level.
[0022] The predicted voltage fluctuation is compared with the set voltage threshold to obtain the voltage fluctuation risk level;
[0023] The predicted frequency offset is compared with the set frequency threshold to obtain the frequency offset risk level;
[0024] The predicted value of the tidal current distribution is compared with the set tidal current threshold to obtain the tidal current distribution risk level.
[0025] In S3, the initialization phase is: when the distributed power source is connected to the grid for the first time, a default scheduling strategy for the initial phase is generated based on the evaluation indicators.
[0026] The dynamic adjustment phase involves obtaining weighting coefficients based on the risk level of distributed generation intelligent grid connection and the grid status during grid operation. The default scheduling strategy is then dynamically adjusted and corrected based on these weighting coefficients and the following objective function to obtain the scheduling strategy for the dynamic adjustment phase.
[0027] ;
[0028] In the above formula, J is the overall optimization objective of the scheduling strategy, which aims to balance the power error and the smoothness of control variable changes. The weighting coefficients are related to the risk level. As a weighting coefficient related to scheduling stability, For the target power value, For the current power, This is a scheduling control variable.
[0029] In S3, the emergency response phase is as follows: when the risk level exceeds the safe operation threshold, the power grid is judged to have entered a high-risk operation state, and a dispatch strategy for the emergency response phase is quickly generated according to the type of risk level that exceeds the safe operation threshold.
[0030] The multi-stage scheduling module further includes a policy switching mechanism, which is used to set the switching conditions and priorities between each scheduling stage, and to perform version management and coordination control of the scheduling policies of different stages.
[0031] A distributed power intelligent grid-connection evaluation and scheduling system based on multi-source data fusion is provided. The system is used to execute the distributed power intelligent grid-connection evaluation and scheduling method based on multi-source data fusion as described in Example 1. Specifically, it includes: a data acquisition and fusion module, a grid-connection evaluation module, a multi-stage scheduling module, and a multi-stage control module.
[0032] The data acquisition and fusion module is used to: acquire raw operating data of distributed power sources and grid nodes, including voltage, current, active power, reactive power and environmental parameters; preprocess and fuse the raw operating data to generate a fused dataset.
[0033] The grid connection assessment module is used to: analyze the impact of distributed power generation grid connection on the grid operation status based on the fused dataset, output assessment indicators, and make predictions by combining the assessment indicators and the grid operation status to obtain short-term predicted values of the grid status and the risk level of distributed power generation grid connection operation.
[0034] The multi-stage scheduling module is used to: generate a multi-stage scheduling strategy based on the evaluation indicators, the scheduling strategy including an initialization stage, a dynamic adjustment stage and an emergency response stage, and generate scheduling instructions based on the current scheduling strategy;
[0035] The multi-stage control module is used to: schedule the power grid according to the scheduling strategy, collect power grid response data to form feedback information, obtain the current power grid operating status based on the feedback information, and return to the grid connection evaluation module to re-evaluate based on the current power grid operating status.
[0036] The data acquisition and fusion module is used to: perform data cleaning, anomaly detection and normalization on the raw operating data, and output the fused dataset to the grid-connected evaluation module and the multi-stage scheduling module;
[0037] The raw operating data includes synchronous operating data from different power sources and grid nodes, including voltage, current, active power, reactive power, and environmental parameters. The preprocessing steps include data cleaning, anomaly detection, and normalization of the raw operating data, and outputting the fused dataset to the grid-connected evaluation module and the multi-stage scheduling module.
[0038] The preprocessed data is time-series aligned and fused to generate the fused dataset. The state of the fused dataset is then estimated to improve the accuracy of the evaluation results of the grid connection evaluation module.
[0039] In the grid connection assessment module, the assessment indicators include voltage fluctuation, frequency offset, and power flow distribution. Predicting the future operating status of the power grid based on these indicators and the grid's operating status includes:
[0040] ;
[0041] In the above formula, This is a short-term prediction of the voltage. The voltage is the current voltage, and η is the adjustment coefficient. The rate of change of voltage is used to obtain the predicted value of voltage fluctuation;
[0042] The current frequency and power flow are predicted using the Newton-Raphson method, and the predicted values of frequency shift and power flow distribution are obtained.
[0043] The risk levels of the intelligent grid connection of distributed power sources include voltage fluctuation risk level, frequency offset risk level and power flow distribution risk level.
[0044] The predicted voltage fluctuation is compared with the set voltage threshold to obtain the voltage fluctuation risk level;
[0045] The predicted frequency offset is compared with the set frequency threshold to obtain the frequency offset risk level;
[0046] The predicted value of the tidal current distribution is compared with the set tidal current threshold to obtain the tidal current distribution risk level.
[0047] In the multi-stage scheduling module, the initialization stage is when the distributed power source is first connected to the grid, a default scheduling strategy for the initial stage is generated based on the evaluation indicators.
[0048] The dynamic adjustment phase involves obtaining weighting coefficients based on the risk level of distributed generation intelligent grid connection and the grid status during grid operation. The default scheduling strategy is then dynamically adjusted and corrected based on these weighting coefficients and the following objective function to obtain the scheduling strategy for the dynamic adjustment phase.
[0049] ;
[0050] In the above formula, J is the overall optimization objective of the scheduling strategy, which aims to balance the power error and the smoothness of control variable changes. The weighting coefficients are related to the risk level. As a weighting coefficient related to scheduling stability, For the target power value, For the current power, This is a scheduling control variable.
[0051] In the multi-stage scheduling module, the emergency response stage is as follows: when the risk level exceeds the safe operation threshold, the power grid is judged to have entered a high-risk operation state, and a scheduling strategy for the emergency response stage is quickly generated according to the type of risk level that exceeds the safe operation threshold.
[0052] The multi-stage scheduling module further includes a policy switching mechanism, which is used to set the switching conditions and priorities between each scheduling stage, and to perform version management and coordination control of the scheduling policies of different stages.
[0053] Compared with the prior art, the beneficial effects of the present invention are as follows:
[0054] 1. In this invention, a distributed power source intelligent grid-connected evaluation and scheduling method based on multi-source data fusion, the collected raw operating data is cleaned, anomaly detected, and normalized to remove noise and abnormal data, ensuring data quality. The preprocessed data is then time-aligned and fused to generate a high-precision fused dataset. This fused dataset contains synchronous operating data from different power sources and grid nodes, reducing bias and inaccuracy in the evaluation of stability, power quality, etc. Therefore, this design can perform grid-connected evaluation and scheduling using a high-precision fused dataset, effectively improving the accuracy of the evaluation.
[0055] 2. In the intelligent grid-connected assessment and scheduling method for distributed power sources based on multi-source data fusion of this invention, during the data acquisition and fusion stage, high-quality fused datasets are generated by cleaning, anomaly detection, time-series alignment, and state estimation of multi-source raw data such as voltage, current, power, and environmental parameters. Based on these fused datasets, short-term predicted values of the grid state are obtained. Grid-connected assessment and scheduling are then performed using these short-term predicted values, avoiding response lag. Therefore, this design can effectively improve response timeliness by performing grid-connected assessment and scheduling using short-term predicted values. Attached Figure Description
[0056] Figure 1 This is a flowchart of the method described in this invention.
[0057] Figure 2 This is a structural diagram of the system described in this invention.
[0058] Figure 3 This is a structural diagram of the device described in this invention. Detailed Implementation
[0059] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0060] Example 1:
[0061] See Figure 1 A method for intelligent grid-connected evaluation and scheduling of distributed power sources based on multi-source data fusion, the evaluation and scheduling method comprising the following steps:
[0062] S1. Collect raw operating data from distributed power sources and grid nodes, preprocess and fuse the raw operating data to generate a fused dataset;
[0063] S2. Based on the fused dataset, analyze the impact of distributed power generation on the grid operation status, obtain evaluation indicators, and combine the evaluation indicators with the grid operation status to make predictions, obtain short-term predicted values of the grid status, and combine the short-term predicted values with the evaluation indicators to evaluate the potential stability or safety risks of distributed power generation grid connection operation, and obtain the risk level of distributed power generation grid connection operation.
[0064] S3. Construct a multi-stage scheduling model, which includes an initialization stage, a dynamic adjustment stage, and an emergency response stage. The multi-stage scheduling model selects the current corresponding scheduling stage based on the current power grid operating status, evaluation indicators, and short-term predicted values of the power grid status, and generates the corresponding scheduling strategy.
[0065] S4. Dispatch the power grid according to the dispatch strategy, collect power grid response data to form feedback information, obtain the current power grid operating status based on the feedback information, and return to S2 to re-evaluate based on the current power grid operating status.
[0066] In S1, the original operating data includes synchronous operating data from different power sources and grid nodes. The synchronous operating data includes voltage, current, active power, reactive power, and environmental parameters. The preprocessing step includes data cleaning, anomaly detection, and normalization of the original operating data, and outputting the fused dataset to the grid-connected evaluation module and the multi-stage scheduling module.
[0067] The data fusion step includes performing time-series alignment and fusion on the preprocessed data to generate the fused dataset, and performing state estimation on the fused dataset to improve the accuracy of the evaluation results of the grid connection evaluation module.
[0068] The accuracy criterion depends on the degree of conformity between the assessment metrics output by the assessment module and the actual operating state of the power grid. Specifically, state estimation optimizes the dataset quality through time alignment and data fusion, reducing noise and errors, thereby improving the matching degree between assessment metrics (such as deviations in voltage and power predictions) and the actual behavior of the power grid.
[0069] When accuracy is insufficient, in S4, feedback information is generated by continuously collecting grid response data. This information is then combined with the current grid operating status to determine the dispatching phase and adjust the strategy. If the assessment accuracy is insufficient (e.g., the predicted value deviates significantly from the actual state), the feedback mechanism will trigger a dynamic adjustment of the dispatching strategy or a switch to the emergency response phase to ensure grid stability and grid connection security.
[0070] By integrating datasets, the impact of distributed generation on grid operation is analyzed, assessment indicators are output, and risk levels are generated based on these indicators and grid operation predictions. Risk levels are indicators used to assess the potential stability or safety risks of distributed generation operation connected to the grid.
[0071] In S2, the evaluation indicators include voltage fluctuation, frequency offset, and power flow distribution. The power system state estimation is completed based on the fused dataset using the weighted least squares method to obtain the node voltage, phase angle, and power flow distribution of all nodes in the system. The voltage fluctuation and frequency offset of the power grid are obtained based on the node voltage and phase angle distribution.
[0072] Predicting the future operating status of the power grid by evaluating indicators and the grid's operating status includes:
[0073] ;
[0074] In the above formula, This is a short-term prediction of the voltage. The voltage is the current voltage, and η is the adjustment coefficient. The rate of change of voltage is used to obtain the predicted value of voltage fluctuation;
[0075] The current frequency and power flow are predicted using the Newton-Raphson method, and the predicted values of frequency shift and power flow distribution are obtained.
[0076] The risk levels of the intelligent grid connection of distributed power sources include voltage fluctuation risk level, frequency offset risk level and power flow distribution risk level.
[0077] The predicted voltage fluctuation is compared with the set voltage threshold to obtain the voltage fluctuation risk level;
[0078] The predicted frequency offset is compared with the set frequency threshold to obtain the frequency offset risk level;
[0079] The predicted value of the tidal current distribution is compared with the set tidal current threshold to obtain the tidal current distribution risk level;
[0080] The voltage fluctuation is used to determine whether there is a voltage over-limit or voltage instability problem;
[0081] The frequency offset is used to determine whether there is a frequency over-limit or frequency instability problem;
[0082] The power flow distribution is used to determine whether there are line overload, equipment overload, or angle-of-attack instability problems.
[0083] In S3, the initialization phase is: when the distributed power source is connected to the grid for the first time, a default scheduling strategy for the initial phase is generated based on the evaluation indicators.
[0084] The dynamic adjustment phase involves obtaining weighting coefficients based on the risk level of distributed generation intelligent grid connection and the grid status during grid operation. The default scheduling strategy is then dynamically adjusted and corrected based on these weighting coefficients and the following objective function to obtain the scheduling strategy for the dynamic adjustment phase.
[0085] ;
[0086] In the above formula, J is the overall optimization objective of the scheduling strategy, which aims to balance the power error and the smoothness of control variable changes. The weighting coefficients are related to the risk level. As a weighting coefficient related to scheduling stability, For the target power value, For the current power, For scheduling and control;
[0087] The weight coefficients are obtained by querying the weight table based on the current risk level and power grid status. The weight table is set according to the risk level and power grid status of historical projects.
[0088] The Based on the risk level, the aforementioned Based on the current power grid operating status, This is used to encourage scheduling strategies to better correct power errors and increase the adjustment range of scheduling control variables. Used to limit the rate of change of dispatch control quantities to avoid drastic power grid fluctuations;
[0089] In practical applications, when the power grid faces high risks (such as severe load fluctuations, grid frequency deviations, etc.), The value of will increase, causing the scheduling strategy to pay more attention to the correction of power errors, for example, when the grid load fluctuates greatly. An increase in will prompt the system to adjust its power more aggressively in order to reduce power error; This is related to the stationarity of the scheduling strategy, which controls the scheduling amount. The rate of change is adjusted to avoid over-adjustment that could cause grid oscillations. For example, when the grid is operating relatively smoothly, When the value is small, the system adjusts more gently; however, when the power grid condition changes drastically, The value of can be increased to limit the rate of change of the scheduling quantity and maintain the stability of the system.
[0090] Generate specific control instructions based on the current scheduling strategy.
[0091] In S3, the emergency response phase is as follows: when the risk level exceeds the safe operation threshold, the power grid is judged to have entered a high-risk operation state, and a dispatch strategy for the emergency response phase is quickly generated according to the type of risk level that exceeds the safe operation threshold.
[0092] The multi-stage scheduling module further includes a policy switching mechanism, which is used to set the switching conditions and priorities between each scheduling stage, and to perform version management and coordination control of the scheduling policies of different stages.
[0093] In S4, from the initial stage to the dynamic adjustment stage: when the power grid enters the normal operation stage and the power error is within the allowable range, the system switches from the initial strategy to the dynamic optimization stage to optimize power scheduling;
[0094] From dynamic adjustment phase to emergency response phase: When the power grid status is detected to deviate significantly from the predetermined target, such as high-frequency fluctuations, excessive power fluctuations, or other high-risk situations, the system automatically switches to the emergency response phase to take rapid adjustment measures to avoid power grid instability.
[0095] Phase rollback: If the power grid's operating status is restored during the emergency response phase, the system will assess the grid's stability and decide whether to roll back to the dynamic adjustment phase or the initial phase.
[0096] For example, during the dynamic adjustment phase, when the grid load fluctuates significantly, the power allocation or voltage setting will be adjusted based on real-time voltage and power data, and specific dispatch instructions will be issued to distributed power sources.
[0097] These instructions guide grid equipment to make adaptive adjustments to achieve power balance or voltage stability;
[0098] During the emergency response phase, the system will generate more stringent control commands, such as forcibly starting or stopping some power sources or adjusting the power factor, in order to quickly respond to abnormal grid conditions.
[0099] Example 2:
[0100] See Figure 2 A distributed power intelligent grid-connection evaluation and scheduling system based on multi-source data fusion is provided. The system is used to execute the distributed power intelligent grid-connection evaluation and scheduling method based on multi-source data fusion as described in Example 1. Specifically, it includes: a data acquisition and fusion module, a grid-connection evaluation module, a multi-stage scheduling module, and a multi-stage control module.
[0101] The data acquisition and fusion module is used to: acquire raw operating data of distributed power sources and grid nodes, including voltage, current, active power, reactive power and environmental parameters; preprocess and fuse the raw operating data to generate a fused dataset.
[0102] The grid connection assessment module is used to: analyze the impact of distributed power generation grid connection on the grid operation status based on the fused dataset, output assessment indicators, and make predictions by combining the assessment indicators and the grid operation status to obtain short-term predicted values of the grid status and the risk level of distributed power generation grid connection operation.
[0103] The multi-stage scheduling module is used to: generate a multi-stage scheduling strategy based on the evaluation indicators, the scheduling strategy including an initialization stage, a dynamic adjustment stage and an emergency response stage, and generate scheduling instructions based on the current scheduling strategy;
[0104] The multi-stage control module is used to: schedule the power grid according to the scheduling strategy, collect power grid response data to form feedback information, obtain the current power grid operating status based on the feedback information, and return to the grid connection evaluation module to re-evaluate based on the current power grid operating status.
[0105] The data acquisition and fusion module is used to: perform data cleaning, anomaly detection and normalization on the raw operating data, and output the fused dataset to the grid-connected evaluation module and the multi-stage scheduling module;
[0106] The raw operating data includes synchronous operating data from different power sources and grid nodes, including voltage, current, active power, reactive power, and environmental parameters. The preprocessing steps include data cleaning, anomaly detection, and normalization of the raw operating data, and outputting the fused dataset to the grid-connected evaluation module and the multi-stage scheduling module.
[0107] The preprocessed data is time-series aligned and fused to generate the fused dataset. The state of the fused dataset is then estimated to improve the accuracy of the evaluation results of the grid connection evaluation module.
[0108] In the grid connection assessment module, the assessment indicators include voltage fluctuation, frequency offset, and power flow distribution. Predicting the future operating status of the power grid based on these indicators and the grid's operating status includes:
[0109] ;
[0110] In the above formula, This is a short-term prediction of the voltage. The voltage is the current voltage, and η is the adjustment coefficient. The rate of change of voltage is used to obtain the predicted value of voltage fluctuation;
[0111] The current frequency and power flow are predicted using the Newton-Raphson method, and the predicted values of frequency shift and power flow distribution are obtained.
[0112] The risk levels of the intelligent grid connection of distributed power sources include voltage fluctuation risk level, frequency offset risk level and power flow distribution risk level.
[0113] The predicted voltage fluctuation is compared with the set voltage threshold to obtain the voltage fluctuation risk level;
[0114] The predicted frequency offset is compared with the set frequency threshold to obtain the frequency offset risk level;
[0115] The predicted value of the tidal current distribution is compared with the set tidal current threshold to obtain the tidal current distribution risk level.
[0116] In the multi-stage scheduling module, the initialization stage is when the distributed power source is first connected to the grid, a default scheduling strategy for the initial stage is generated based on the evaluation indicators.
[0117] The dynamic adjustment phase involves obtaining weighting coefficients based on the risk level of distributed generation intelligent grid connection and the grid status during grid operation. The default scheduling strategy is then dynamically adjusted and corrected based on these weighting coefficients and the following objective function to obtain the scheduling strategy for the dynamic adjustment phase.
[0118] ;
[0119] In the above formula, J is the overall optimization objective of the scheduling strategy, which aims to balance the power error and the smoothness of control variable changes. The weighting coefficients are related to the risk level. As a weighting coefficient related to scheduling stability, For the target power value, For the current power, For scheduling and control;
[0120] The weight coefficients are obtained by querying the weight table based on the current risk level and power grid status. The weight table is set according to the risk level and power grid status of historical projects.
[0121] Generate specific control instructions based on the current scheduling strategy.
[0122] In the multi-stage scheduling module, the emergency response stage is as follows: when the risk level exceeds the safe operation threshold, the power grid is judged to have entered a high-risk operation state, and a scheduling strategy for the emergency response stage is quickly generated according to the type of risk level that exceeds the safe operation threshold.
[0123] The multi-stage scheduling module further includes a policy switching mechanism, which is used to set the switching conditions and priorities between each scheduling stage, and to perform version management and coordination control of the scheduling policies of different stages.
[0124] Example 3:
[0125] Example 3 is basically the same as Example 2, except that:
[0126] This system also includes a communication and security module, which is responsible for ensuring efficient and secure data transmission between the various modules of the system. The main function of this module is to safeguard data security through encryption and authentication mechanisms, and to implement redundancy mechanisms to ensure system stability under various operating conditions.
[0127] First, regarding data transmission, the communication and security modules use secure communication protocols such as TLS (Transport Layer Security) or VPN (Virtual Private Network) to ensure that data transmitted between modules within the system is not stolen or tampered with by unauthorized third parties. During data transmission, all sensitive data (such as control commands, power grid status, feedback information, etc.) is encrypted to ensure the security of information transmission.
[0128] To enhance data transmission reliability and prevent single points of failure, the communication and security module is also equipped with a redundancy mechanism. When a communication channel fails, the redundancy mechanism can quickly switch to a backup channel, ensuring continuous data transmission. For example, if the primary network connection fails, the system can automatically switch to the backup link to avoid system downtime or data loss, ensuring the stability of the power grid operation.
[0129] Regarding identity authentication, the communication and security module integrates a multi-factor authentication mechanism. Data interaction between modules requires authentication to ensure that only authorized modules can access or modify data. The system uses digital certificates, two-factor authentication, and other methods to ensure that every module involved in data transmission and processing is verified, preventing malicious attackers from impersonating others to perform unauthorized operations.
[0130] Furthermore, the communication and security module can monitor the status and security of data transmission in real time. Upon detecting anomalies, it will immediately take appropriate measures, such as interrupting data transmission, activating backup channels, or notifying the administrator for handling. Through these security measures, the communication and security module provides efficient and reliable support to the system, ensuring that data remains protected in complex power grid operating environments and preventing power grid malfunctions due to data leakage or tampering.
[0131] Overall, the communication and security module provides the necessary security for the entire system, ensuring the integrity, confidentiality, and reliability of data during transmission, thereby enabling the efficient and stable operation of this invention.
[0132] In some embodiments, the system further includes a human-computer interaction module, which in this invention primarily provides users with an intuitive and convenient interface and operating methods, enabling users to monitor the power grid's operating status in real time, view dispatching strategies, and input and adjust control commands to the system. This module provides users with comprehensive interactive functions through a visual interface and control input unit.
[0133] Firstly, regarding the visualization interface, the system graphically displays the power grid's operating status, evaluation indicators, and dispatch strategies. Users can view real-time trends in power grid data such as voltage, frequency, and power, and key evaluation indicators (such as voltage fluctuations and frequency deviations) are displayed through intuitive charts and graphs. Furthermore, the system uses maps or distributed power source location markers to help users intuitively understand the operating status and position of each power grid node, enabling more accurate decision-making. The visualization interface dynamically adjusts its display content according to different dispatch stages, allowing users to access the current power grid's specific operating status and dispatch strategies at any time.
[0134] Secondly, regarding the control input unit, users can input control commands or adjust existing dispatch strategies through this unit. When the system detects grid anomalies or load fluctuations, users can quickly adjust the dispatch strategy based on feedback from the visual interface. For example, users can adjust dispatch strategy parameters using buttons or sliders on the interface, or immediately respond to grid operation needs by selecting preset emergency strategies. The system will update the dispatch strategy in real time based on the user-input control commands and execute corresponding operations through multi-stage control modules.
[0135] Regarding policy adjustment requests, the control input unit supports users in making personalized modifications and adjustments to existing scheduling policies. Users can choose to modify the scheduling phase, adjust power allocation, or optimize the load balance of the power grid.
[0136] All input commands are immediately transmitted to the stage control module or multi-stage scheduling module, triggering the corresponding adjustment mechanism. To ensure stable system operation, the control input unit also performs real-time verification before execution, ensuring that the input adjustment commands comply with the system's operating rules and complete the task without affecting power grid safety.
[0137] The human-computer interaction module provided in this embodiment offers users a flexible and highly operable interface. Through intuitive data display and convenient control input methods, users can make timely and effective decisions during grid connection and dispatch, thereby improving the system's response speed and operational efficiency.
[0138] Example 4:
[0139] See Figure 3A distributed power intelligent grid-connected evaluation and scheduling device based on multi-source data fusion includes a memory and a processor. The memory is used to store computer program code and transmit the computer program code to the processor.
[0140] The processor is configured to execute, according to instructions in the computer program code, the distributed power intelligent grid-connection evaluation and scheduling method based on multi-source data fusion as described in Embodiment 1.
[0141] Example 5:
[0142] A computer program product includes a computer program that is executed by a processor as described in Example 1, which is a distributed power intelligent grid-connected evaluation and scheduling method based on multi-source data fusion.
[0143] The above description is only a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. Any equivalent modifications or changes made by those skilled in the art based on the content disclosed in the present invention should be included within the scope of protection set forth in the claims.
Claims
1. A method for intelligent grid-connected evaluation and scheduling of distributed power sources based on multi-source data fusion, characterized in that, The evaluation and scheduling method includes the following steps: S1. Collect raw operating data from distributed power sources and grid nodes, preprocess and fuse the raw operating data to generate a fused dataset; S2. Based on the fused dataset, analyze the impact of distributed power generation on the grid operation status, obtain evaluation indicators, and combine the evaluation indicators with the grid operation status to make predictions, obtain short-term predicted values of the grid status, and combine the short-term predicted values with the evaluation indicators to evaluate the potential stability or safety risks of distributed power generation grid connection operation, and obtain the risk level of distributed power generation grid connection operation. S3. Construct a multi-stage scheduling model, which includes an initialization stage, a dynamic adjustment stage, and an emergency response stage. The multi-stage scheduling model selects the current corresponding scheduling stage based on the current power grid operating status, evaluation indicators, and short-term predicted values of the power grid status, and generates the corresponding scheduling strategy. S4. Dispatch the power grid according to the dispatch strategy, collect power grid response data to form feedback information, obtain the current power grid operating status based on the feedback information, and return to S2 to re-evaluate based on the current power grid operating status.
2. The distributed power intelligent grid-connection evaluation and scheduling method based on multi-source data fusion according to claim 1, characterized in that, In S1, the original operating data includes synchronous operating data from different power sources and grid nodes. The synchronous operating data includes voltage, current, active power, reactive power, and environmental parameters. The preprocessing step includes data cleaning, anomaly detection, and normalization of the original operating data, and outputting the fused dataset to the grid-connected evaluation module and the multi-stage scheduling module. The data fusion step includes performing time-series alignment and fusion on the preprocessed data to generate the fused dataset.
3. The method for intelligent grid-connected evaluation and scheduling of distributed power sources based on multi-source data fusion according to claim 1, characterized in that, In step S2, the evaluation indicators include voltage fluctuation, frequency offset, and power flow distribution. Predicting the future operating state of the power grid using these evaluation indicators and the grid's operating status includes: ； In the above formula, This is a short-term prediction of the voltage. The voltage is the current voltage, and η is the adjustment coefficient. The rate of change of voltage is used to obtain the predicted value of voltage fluctuation; The current frequency and power flow are predicted using the Newton-Raphson method, and the predicted values of frequency shift and power flow distribution are obtained respectively. The risk levels of the intelligent grid connection of distributed power sources include voltage fluctuation risk level, frequency offset risk level and power flow distribution risk level. The predicted voltage fluctuation is compared with the set voltage threshold to obtain the voltage fluctuation risk level; The predicted frequency offset is compared with the set frequency threshold to obtain the frequency offset risk level; The predicted value of the tidal current distribution is compared with the set tidal current threshold to obtain the tidal current distribution risk level.
4. The distributed power intelligent grid-connection evaluation and scheduling method based on multi-source data fusion according to claim 1, characterized in that, In S3, the initialization phase is: when the distributed power source is connected to the grid for the first time, a default scheduling strategy for the initial phase is generated based on the evaluation indicators. The dynamic adjustment phase involves obtaining weighting coefficients based on the risk level of distributed generation intelligent grid connection and the grid status during grid operation. The default scheduling strategy is then dynamically adjusted and corrected based on these weighting coefficients and the following objective function to obtain the scheduling strategy for the dynamic adjustment phase. ； In the above formula, J is the overall optimization objective of the scheduling strategy, which aims to balance the power error and the smoothness of control variable changes. The weighting coefficients are related to the risk level. As a weighting coefficient for scheduling stability, the Based on the risk level, the aforementioned Based on the current power grid operating status, For the target power value, For the current power, This is a scheduling control variable.
5. The distributed power intelligent grid-connection evaluation and scheduling method based on multi-source data fusion according to claim 4, characterized in that, In S3, the emergency response phase is as follows: when the risk level exceeds the safe operation threshold, the power grid is judged to have entered a high-risk operation state, and a dispatch strategy for the emergency response phase is quickly generated according to the type of risk level that exceeds the safe operation threshold. The multi-stage scheduling module further includes a policy switching mechanism, which is used to set the switching conditions and priorities between each scheduling stage, and to perform version management and coordination control of the scheduling policies of different stages.
6. A distributed power intelligent grid-connection evaluation and scheduling system based on multi-source data fusion, characterized in that, The system is used to execute the distributed power intelligent grid-connected evaluation and scheduling method based on multi-source data fusion as described in any one of claims 1 to 5, specifically including: a data acquisition and fusion module, a grid-connected evaluation module, a multi-stage scheduling module, and a multi-stage control module; The data acquisition and fusion module is used to: acquire raw operating data of distributed power sources and grid nodes, including voltage, current, active power, reactive power and environmental parameters; preprocess and fuse the raw operating data to generate a fused dataset. The grid connection assessment module is used to: analyze the impact of distributed power generation grid connection on the grid operation status based on the fused dataset, output assessment indicators, and make predictions by combining the assessment indicators and the grid operation status to obtain short-term predicted values of the grid status and the risk level of distributed power generation grid connection operation. The multi-stage scheduling module is used to: generate a multi-stage scheduling strategy based on the evaluation indicators, the scheduling strategy including an initialization stage, a dynamic adjustment stage and an emergency response stage, and generate scheduling instructions based on the current scheduling strategy; The multi-stage control module is used to: schedule the power grid according to the scheduling strategy, collect power grid response data to form feedback information, obtain the current power grid operating status based on the feedback information, and return to the grid connection evaluation module to re-evaluate based on the current power grid operating status.
7. A distributed power intelligent grid-connection evaluation and scheduling system based on multi-source data fusion as described in claim 6. The data acquisition and fusion module is used to: perform data cleaning, anomaly detection and normalization on the raw operating data, and output the fused dataset to the grid-connected evaluation module and the multi-stage scheduling module; The raw operating data includes synchronous operating data from different power sources and grid nodes, including voltage, current, active power, reactive power, and environmental parameters. The preprocessing steps include cleaning, detecting anomalies, and normalizing the raw operating data, and outputting the fused dataset to the grid-connected evaluation module and the multi-stage scheduling module. The preprocessed data is time-series aligned and fused to generate the fused dataset.
8. A distributed power intelligent grid-connection evaluation and scheduling system based on multi-source data fusion as described in claim 6. In the grid connection assessment module, the assessment indicators include voltage fluctuation, frequency offset, and power flow distribution. Predicting the future operating status of the power grid based on these indicators and the grid's operating status includes: ； In the above formula, This is a short-term prediction of the voltage. The voltage is the current voltage, and η is the adjustment coefficient. The rate of change of voltage is used to obtain the predicted value of voltage fluctuation; The current frequency and power flow are predicted using the Newton-Raphson method, and the predicted values of frequency shift and power flow distribution are obtained. The risk levels of the intelligent grid connection of distributed power sources include voltage fluctuation risk level, frequency offset risk level and power flow distribution risk level. The predicted voltage fluctuation is compared with the set voltage threshold to obtain the voltage fluctuation risk level; The predicted frequency offset is compared with the set frequency threshold to obtain the frequency offset risk level; The predicted value of the tidal current distribution is compared with the set tidal current threshold to obtain the tidal current distribution risk level.
9. A distributed power intelligent grid-connection evaluation and scheduling system based on multi-source data fusion as described in claim 6. In the multi-stage scheduling module, the initialization stage is when the distributed power source is first connected to the grid, a default scheduling strategy for the initial stage is generated based on the evaluation indicators. The dynamic adjustment phase involves obtaining weighting coefficients based on the risk level of distributed generation intelligent grid connection and the grid status during grid operation. The default scheduling strategy is then dynamically adjusted and corrected based on these weighting coefficients and the following objective function to obtain the scheduling strategy for the dynamic adjustment phase. ； In the above formula, J is the overall optimization objective of the scheduling strategy, which aims to balance the power error and the smoothness of control variable changes. The weighting coefficients are related to the risk level. As a weighting coefficient related to scheduling stability, For the target power value, For the current power, For scheduling control; The weight coefficients are obtained by querying the weight table based on the current risk level and power grid status. The weight table is set according to the risk level and power grid status of historical projects. Generate specific control instructions based on the current scheduling strategy.
10. A distributed power intelligent grid-connection evaluation and scheduling system based on multi-source data fusion as described in claim 6. In S3, the emergency response phase is as follows: when the risk level exceeds the safe operation threshold, the power grid is judged to have entered a high-risk operation state, and a dispatch strategy for the emergency response phase is quickly generated according to the type of risk level that exceeds the safe operation threshold. The multi-stage scheduling module further includes a policy switching mechanism, which is used to set the switching conditions and priorities between each scheduling stage, and to perform version management and coordination control of the scheduling policies of different stages.