A distributed photovoltaic power generation grid-connected anti-backflow control system and control method

Abstract

The present invention provides a distributed photovoltaic power generation grid-connected anti-backflow control system and control method, the system includes a load prediction and optimization component, a backflow detection and response component, and a control and optimization scheduling component. The method includes receiving inputs from multiple data sources such as power grid scheduling data, user electricity usage habits, and weather forecasts, and real-time prediction of load demand and power grid capacity in the future, providing a decision-making basis for the control and scheduling component; receiving the load demand and power grid capacity predicted by the load prediction and optimization component, as well as real-time power grid status data; when it is detected that a backflow is about to occur in the power grid, sending an instruction to adjust the photovoltaic output to the control and optimization scheduling component; receiving the predicted load demand and power grid capacity, and an instruction to adjust the photovoltaic output; optimizing the scheduling strategy of photovoltaic power generation, dynamically adjusting the output of photovoltaic power generation, and sending a control signal to the photovoltaic power generation equipment. The present invention can reduce power grid failures and losses caused by backflow.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic backflow prevention control, and in particular to a distributed photovoltaic power generation grid-connected backflow prevention control system and a control method. Background Art

[0002] Distributed photovoltaic power generation is a method of using solar photovoltaic technology to install distributed photovoltaic power generation systems near users, such as rooftops, parking lots, etc., to generate electricity and use electricity locally. This power generation method has the advantages of strong flexibility, low investment and short construction period, which can effectively reduce transmission losses and grid pressure and improve energy utilization efficiency. Grid-connected anti-reverse flow control refers to a control measure taken in distributed photovoltaic power generation systems to prevent the distributed photovoltaic power generation system from sending power to the grid in reverse when the power generation is greater than the load demand, causing grid voltage fluctuations or other safety hazards. Traditional methods include power control, voltage control, frequency control and reverse power protection. Although they can achieve anti-reverse flow control of photovoltaics, due to frequent disconnection, on the one hand, it greatly increases the cost of photovoltaic operation and maintenance, and on the other hand, it greatly affects the photovoltaic absorption rate, causing considerable losses to investors.

[0003] Prior art 1, application number: CN202410648108.5 discloses an optimization control method and system for industrial and commercial energy storage power stations, which collects power grid operation parameters, real-time active power of energy storage power stations and real-time active power of grid connection points; optimizes the real-time active power of grid connection points and real-time active power of energy storage power stations according to anti-backflow constraint conditions or anti-overcapacity constraint conditions of grid connection points, and takes the difference between the real-time active power optimization value of grid connection points and the real-time active power optimization value of energy storage power stations as the real-time equivalent load of grid connection points; based on time-of-use electricity prices, the lowest electricity price of the real-time equivalent load of the grid connection point is taken as the objective function; constructs rolling optimization variables based on the energy storage power station operation plan curve; constructs the constraint coefficient and constraint range of the rolling optimization variables based on the operating state constraint conditions of the energy storage power station, the anti-backflow constraint conditions of the grid connection point and the anti-overcapacity constraint conditions; constructs an energy storage power station operation benefit model; solves the energy storage power station operation benefit model, and optimizes and controls the industrial and commercial energy storage power stations. Although it is possible to optimize the control of industrial and commercial energy storage power stations without relying on load data, the anti-backflow control method usually adopts the rough cutting off of the photovoltaic anti-backflow switch. Although this method can achieve anti-backflow control, frequent cutting off will greatly increase the photovoltaic operation and maintenance costs. Frequent cutting off of photovoltaic anti-backflow switches not only increases the operation and maintenance costs, but also greatly affects the photovoltaic absorption rate, causing losses to investors. Anti-backflow and anti-overcapacity are achieved through optimized control methods, but in actual operation, there may be a lack of refined control means, resulting in less than ideal control effects. Anti-backflow and anti-overcapacity are mainly achieved by optimizing the real-time active power of the grid connection point and the real-time active power of the energy storage power station, but in actual operation, how to better match the grid capacity and load demand is still a challenge.

[0004] Prior art 2, application number: CN 202110409055.8 discloses a method and system for preventing backflow in high-voltage grid-connected photovoltaic power stations. The method for preventing backflow in high-voltage grid-connected photovoltaic power stations is as follows: detecting the buying or selling state of the photovoltaic power station on the high-voltage side of the photovoltaic power station, and when it is detected that the photovoltaic power station is in the selling state, reducing the power generation power of the photovoltaic power station; when it is detected that the photovoltaic power station is in the buying state and the buying power reaches or exceeds a% of the installed capacity of the photovoltaic power station, increasing the power generation power of the photovoltaic power station, when it is detected that the photovoltaic power station is in the buying state and the buying power is less than a% of the installed capacity of the photovoltaic power station, the power generation power of the photovoltaic power station remains unchanged, where a% is a preset ratio threshold. The high-voltage grid-connected anti-backflow system for photovoltaic power stations based on the above-mentioned method for preventing backflow in high-voltage grid-connected photovoltaic power stations includes an electric meter and an anti-backflow controller. Although it is easy to implement and has low cost, it has low cost and reliable control effect. However, by detecting the buying or selling status of the photovoltaic power station and adjusting the power generation accordingly to achieve anti-backflow control, frequent power adjustment and state switching may lead to increased operation and maintenance costs; frequent power adjustment and state switching not only increase the operation and maintenance costs, but may also affect the photovoltaic absorption rate and cause losses to investors; anti-backflow control is achieved by adjusting the power generation, but in actual operation, there may be a lack of refined control methods, resulting in less than ideal control effects; anti-backflow control is achieved by detecting the buying or selling status of electricity and adjusting the power generation, but in actual operation, how to better match the grid capacity and load demand remains a challenge.

[0005] Prior art three, application number: CN 202010339040.4 discloses a grid-connected power generation anti-backflow system and method, computer equipment and storage medium, which monitors the electrical signals at the grid connection point, disconnects the grid connection point switch when a backflow occurs, and calculates the required electric energy of the user load by analyzing the real-time electrical signals at the property boundary point. The maximum amount of power generation units in the system is judged based on logical analysis, and the generated electric energy will not exceed the needs of the user load and cause a backflow phenomenon. Perform operations according to the analysis results, disconnect the collection lines of the redundant power generation units, and close the grid connection point switch to provide the required electric energy to the user. Although it is possible to ensure that the system's new energy power generation is consumed by users to the maximum extent without reverse flow, it can avoid the phenomenon of serious waste of power generation resources and improve the economic benefits of the system. However, by monitoring the electrical signals at the grid connection point, disconnecting the grid connection point switch when reverse flow occurs, and disconnecting the redundant power generation unit collection lines based on the analysis results, this frequent switching operation will lead to an increase in operation and maintenance costs; frequent switching operations not only increase operation and maintenance costs, but may also affect the photovoltaic absorption rate and cause losses to investors; anti-reverse flow control is achieved by disconnecting the grid connection point switch and the redundant power generation unit collection lines, but in actual operation, there may be a lack of refined control methods, resulting in less than ideal control effects; anti-reverse flow control is achieved by monitoring electrical signals and disconnecting the switch, but in actual operation, how to better match the grid capacity and load demand is still a challenge; reverse flow is avoided by disconnecting the collection lines of redundant power generation units, but this method may cause some power generation resources to be wasted.

[0006] At present, the existing technologies 1, 2 and 3 have problems such as high operation and maintenance costs due to frequent disconnection, low photovoltaic absorption rate, insufficient refinement of anti-backflow control, and poor matching between grid capacity and load demand. Therefore, the present invention provides a distributed photovoltaic power generation grid-connected anti-backflow control system and control method. Summary of the invention

[0007] In order to solve the above technical problems, the present invention provides a distributed photovoltaic power generation grid-connected anti-backflow control system, comprising:

[0008] The load forecasting and optimization component is responsible for receiving inputs from multiple data sources such as power grid dispatching data, user electricity consumption habits, and weather forecasts, and predicting load demand and power grid capacity in the future in real time, providing decision-making basis for the control and dispatching components;

[0009] The reverse flow detection and response component is responsible for receiving the load demand and grid capacity predicted by the load prediction and optimization component, as well as real-time grid status data; when it detects that a reverse flow is about to occur in the grid, it sends instructions to the control and optimization scheduling component to adjust the photovoltaic output;

[0010] The control and optimization scheduling component is responsible for receiving predicted load demand and grid capacity, as well as instructions for adjusting photovoltaic output; optimizing the scheduling strategy of photovoltaic power generation, dynamically adjusting the output of photovoltaic power generation, and sending control signals to photovoltaic power generation equipment.

[0011] Optional load forecasting and optimization components, including:

[0012] The data preprocessing module is responsible for acquiring data from multiple data sources, identifying and removing outliers and noise data, unifying data of different dimensions to the same scale, and obtaining preprocessed data;

[0013] The feature extraction module is responsible for extracting the load change curve over the past period of time to reflect the time series characteristics of the load; analyzing the user's electricity consumption habits to extract the electricity consumption characteristics during peak and off-peak hours; extracting temperature, humidity, wind speed and sunshine intensity meteorological parameters as important features affecting photovoltaic power generation; and selecting the most influential features for load forecasting and grid capacity forecasting from the time series features, electricity consumption features and important features to obtain the model training feature set;

[0014] The model building module is responsible for dividing the model training feature set into training set, validation set and test set, using the training set data to train the support vector machine, adjusting the support vector machine parameters through gradient descent to minimize the prediction error; using the validation set data to evaluate the performance of the support vector machine, and using the mean square error indicator to measure the prediction accuracy of the support vector machine; adjusting the support vector machine parameters and structure based on the evaluation results.

[0015] Optional, backflow detection and response components, including:

[0016] The reverse flow detection module is responsible for continuously monitoring the key parameters of the power grid, such as voltage, current and frequency, and identifying the typical characteristics of abnormal changes in the current direction of reverse flow in the power grid; combining load demand and power grid capacity forecast, it evaluates the probability and impact of reverse flow;

[0017] The reverse flow decision module is responsible for detecting that a reverse flow is about to occur. The reverse flow detection and response component will immediately issue a warning signal and start the decision-making mechanism; generate a reverse flow warning signal to notify the control and optimization scheduling component of the imminent reverse flow; combine the load demand and grid capacity forecast to evaluate the feasibility of adjusting the photovoltaic output and obtain the reverse flow decision result;

[0018] The instruction generation module is responsible for generating specific instructions for adjusting photovoltaic output according to the results of the backflow risk assessment, and sending the instructions to the control and optimization scheduling components in real time.

[0019] Optional, backflow detection module, including:

[0020] The probability evaluation submodule is responsible for constructing a dynamic Bayesian network model using load demand, grid capacity and reverse flow occurrence variables as nodes. Each node represents a random variable, and the connection between nodes represents the dependency relationship between variables. The parameters of the dynamic Bayesian network model are learned using historical variable data to determine the conditional probability distribution between nodes and calculate the probability distribution of reverse flow occurrence under the current load demand and grid capacity conditions.

[0021] The impact assessment submodule is responsible for identifying key nodes and vulnerable links in the power grid, assessing the impact of reverse flow on the stability of the power grid structure; simulating the operating status of the equipment when reverse flow occurs, assessing the thermal stress and mechanical stress parameters of the equipment, and predicting the life and failure risk of the equipment; constructing a causal relationship map between each device and node in the power grid, and through the causal map, assessing the chain reaction that may be caused by the reverse flow, and predicting the potential fault diffusion path and impact range;

[0022] The comprehensive assessment submodule is responsible for dynamically adjusting the weights of probability assessment and impact assessment based on historical assessment results and actual operation feedback; and weightedly integrating the results of probability assessment and impact assessment to generate a backflow risk index.

[0023] Optional control and optimization scheduling components, including:

[0024] The result evaluation module is responsible for analyzing the predicted load demand and obtaining the peak and valley periods of electricity consumption of users in the future; analyzing the grid capacity and evaluating the grid's carrying capacity and potential risks;

[0025] The strategy formulation module is responsible for setting the maximum utilization rate of photovoltaic power generation, minimizing grid load fluctuations and reducing the risk of reverse flow; based on the settings and analysis results, the photovoltaic power generation scheduling strategy is formulated, including dynamic adjustment of photovoltaic output, priority setting and emergency plans;

[0026] The strategy execution module is responsible for sending control signals to photovoltaic power generation equipment and adjusting output according to the formulated scheduling strategy; real-time monitoring of the operating status of photovoltaic power generation equipment and the status of the power grid, and timely adjustment of the scheduling strategy.

[0027] Optional, policy making module, including:

[0028] The target setting submodule is responsible for adjusting the operating status of photovoltaic power generation equipment by optimizing the dispatching strategy; dynamically adjusting photovoltaic output to balance the grid load; dispatching photovoltaic power generation equipment to avoid grid reverse flow;

[0029] The dynamic adjustment submodule is responsible for obtaining the load demand and the current and future capacity of the power grid, dynamically adjusting the output of photovoltaic power generation equipment to adapt to changes in load demand; real-time monitoring of the operating status of photovoltaic power generation equipment, including the health status of the equipment and power generation efficiency; setting the priority of the equipment according to the operating status, geographical location and power generation efficiency factors of the equipment, and giving priority to dispatching high-priority equipment when the power grid load fluctuates;

[0030] The emergency plan submodule is responsible for automatically starting the backup power supply when the power grid fails or the equipment fails, and transferring the load from the fault area to other areas with normal operation.

[0031] Optional, dynamically adjust submodules, including:

[0032] The condition setting unit is responsible for setting the conditions for dynamically adjusting the start-up of photovoltaic power generation equipment, where the future capacity is greater than the current capacity; when the conditions are met, the real-time change in load demand is obtained;

[0033] The value acquisition unit is responsible for calling the output mapping table of the photovoltaic power generation equipment. The mapping table records the corresponding relationship between the real-time change of the load demand and the output, and calls the corresponding output value according to the real-time change; the mapping table is calculated based on the historical load demand and the output of the photovoltaic power generation equipment;

[0034] The adjustment sequence unit is responsible for forming an adjustment strategy based on the load demand and the corresponding output, and at the same time, adjusting the output sequence according to the priority of the photovoltaic power generation equipment.

[0035] Optional, value acquisition unit, including:

[0036] The numerical calculation subunit is responsible for calculating the load demand change at each time point based on the collected load demand data. The load demand change is calculated by the load demand difference between the previous and next time points. The output value corresponding to each load demand change is determined based on the collected output data. The output value is determined by the corresponding relationship in the historical data.

[0037] The table creation subunit is responsible for creating a table whose columns include load demand change and corresponding output value; the calculated load demand change and corresponding output value are filled into the table;

[0038] The data sorting subunit is responsible for sorting the data in the table from small to large according to the size of the load demand change; for the missing load demand change and output value, the interpolation method is used to supplement the data in the table in a personalized way.

[0039] Optional, data sorting subunit, including:

[0040] The grid division component is responsible for setting a coordinate system and dividing the load demand change area into multiple grids. Each grid contains a certain range of load demand change, which is used to preliminarily locate the position of the missing output value; obtain the coordinate longitude and latitude information of each grid, and determine the load demand change and output value within the grid;

[0041] The data correction component is responsible for obtaining information related to the output value through historical data records, which is used to preliminarily determine the candidate value of the missing output value; divide the second type of grid, and correct the load demand change and output value;

[0042] The interpolation calculation component is responsible for using the interpolation algorithm to solve the optimal missing data value based on the corrected load demand change and output value; and adjusting the grid size and related parameters according to the interpolation results.

[0043] The present invention provides a distributed photovoltaic power generation grid-connected anti-backflow control method, comprising the following steps:

[0044] Receive input from multiple data sources such as grid dispatch data, user electricity usage habits, and weather forecasts, and make real-time predictions of load demand and grid capacity in the future, providing decision-making basis for control and dispatch components;

[0045] Receive the load demand and grid capacity predicted by the load forecasting and optimization component, as well as real-time grid status data; when detecting that reverse flow is about to occur in the grid, send instructions to the control and optimization scheduling component to adjust the photovoltaic output;

[0046] Receive predicted load demand and grid capacity, and instructions for adjusting photovoltaic output; optimize photovoltaic power generation scheduling strategy, dynamically adjust photovoltaic power generation output, and send control signals to photovoltaic power generation equipment.

[0047] The load prediction and optimization component of the present invention can predict the load demand and grid capacity in the future in real time through big data analysis and machine learning algorithms, improve the accuracy and timeliness of the prediction; integrate multiple data sources such as grid dispatching data, user electricity consumption habits and weather forecasts to ensure the comprehensiveness and reliability of the prediction results. The significance achieved: provide accurate decision-making basis for the control and dispatching components, reduce human intervention, and improve the automation level of distributed photovoltaic power generation systems; optimize the configuration of grid resources through accurate load forecasting, improve the absorption rate of photovoltaic power generation, and reduce the phenomenon of abandoned light. The reverse flow detection and response component can quickly detect the situation of the imminent occurrence of reverse flow through real-time grid status data, and improve the response speed of the distributed photovoltaic power generation system; when it is detected that the reverse flow is about to occur, it can quickly send adjustment instructions to the control and optimization dispatching component to achieve dynamic adjustment of photovoltaic output and avoid the occurrence of reverse flow. The significance achieved: through rapid response and dynamic adjustment, ensure the stable operation of the power grid, reduce grid failures and losses caused by reverse flow; through real-time monitoring and rapid response, improve the reliability and safety of distributed photovoltaic power generation systems, and reduce operation and maintenance costs. The control and optimization dispatching component optimizes the dispatching strategy of photovoltaic power generation according to the predicted load demand and grid capacity, as well as the adjustment instructions of the reverse flow detection and response component, and realizes dynamic adjustment of photovoltaic output; by sending control signals to photovoltaic power generation equipment, it realizes precise control of photovoltaic power generation output and ensures the efficient operation of distributed photovoltaic power generation system. The significance achieved: Through optimizing the dispatching strategy and precise control, the absorption rate of photovoltaic power generation is improved, the phenomenon of abandoned light is reduced, and the overall efficiency of distributed photovoltaic power generation system is improved; through automated and intelligent dispatching strategies, human intervention is reduced, operation and maintenance costs are reduced, and the economic benefits of distributed photovoltaic power generation system are improved.

[0048] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0049] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0051] Figure 1 This is a block diagram of a distributed photovoltaic power generation grid-connected anti-backflow control system in Example 1 of the present invention;

[0052] Figure 2This is a block diagram of the load prediction and optimization components in Embodiment 2 of the present invention;

[0053] Figure 3 This is a block diagram of a backflow detection and response component in Example 3 of the present invention;

[0054] Figure 4 This is a block diagram of a backflow detection module in Example 4 of the present invention;

[0055] Figure 5 This is a block diagram of the control and optimization scheduling components in Example 5 of the present invention;

[0056] Figure 6 This is a block diagram of a strategy formulation module in Embodiment 6 of the present invention;

[0057] Figure 7 This is a block diagram of a dynamic adjustment submodule in Embodiment 7 of the present invention;

[0058] Figure 8 This is a block diagram of a numerical value acquisition unit in Embodiment 8 of the present invention;

[0059] Fig. 9 This is a block diagram of a data sorting subunit in Embodiment 9 of the present invention;

[0060] Fig.10 This is a flow chart of the anti-backflow control method for distributed photovoltaic power generation grid-connected in Example 10 of the present invention. DETAILED DESCRIPTION

[0061] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0062] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the embodiments of the present application. The singular forms of "a", "said" and "the" used in the embodiments of the present application are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in this article refers to and includes any or all possible combinations of one or more associated listed items.

[0063] When the following description relates to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are only examples of devices and methods consistent with some aspects of the present application. In the description of the present application, it should be understood that the terms "first", "second", "third", etc. are only used to distinguish similar objects, and do not have to be used to describe a specific order or sequence, nor can they be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances.

[0064] Example 1: Figure 1 As shown, an embodiment of the present invention provides a distributed photovoltaic power generation grid-connected anti-backflow control system, comprising:

[0065] The load forecasting and optimization component is responsible for receiving inputs from multiple data sources such as power grid dispatching data, user electricity consumption habits, and weather forecasts, and predicts load demand and power grid capacity in the future in real time, providing decision-making basis for the control and dispatching components;

[0066] The reverse flow detection and response component is responsible for receiving the load demand and grid capacity predicted by the load prediction and optimization component, as well as real-time grid status data; when it detects that a reverse flow is about to occur in the grid, it sends instructions to the control and optimization scheduling component to adjust the photovoltaic output;

[0067] The control and optimization scheduling component is responsible for receiving predicted load demand and grid capacity, as well as instructions for adjusting photovoltaic output; optimizing the scheduling strategy of photovoltaic power generation, dynamically adjusting the output of photovoltaic power generation, and sending control signals to photovoltaic power generation equipment.

[0068] The working principle and beneficial effects of the above technical solution are as follows: the load prediction and optimization component of this embodiment receives inputs from multiple data sources such as power grid dispatching data, user electricity usage habits and weather forecasts, and predicts the load demand and power grid capacity in the future in real time, providing decision-making basis for the control and dispatching component; the reverse flow detection and response component receives the load demand and power grid capacity predicted by the load prediction and optimization component, as well as real-time power grid status data; when it is detected that reverse flow is about to occur in the power grid, it sends an instruction to adjust the photovoltaic output to the control and optimization dispatching component; the control and optimization dispatching component receives the predicted load demand and power grid capacity, and the instruction to adjust the photovoltaic output; optimizes the dispatching strategy of photovoltaic power generation, dynamically adjusts the output of photovoltaic power generation, and sends control signals to photovoltaic power generation equipment. The load prediction and optimization component of the above solution can predict the load demand and power grid capacity in the future in real time through big data analysis and machine learning algorithms, improve the accuracy and timeliness of the prediction; integrate multiple data sources such as power grid dispatching data, user electricity usage habits and weather forecasts to ensure the comprehensiveness and reliability of the prediction results. Significance achieved: Provide accurate decision-making basis for control and dispatch components, reduce human intervention, and improve the automation level of distributed photovoltaic power generation systems; optimize the configuration of power grid resources through accurate load forecasting, improve the absorption rate of photovoltaic power generation, and reduce the phenomenon of abandoned light. The reverse flow detection and response component can quickly detect the situation of impending reverse flow through real-time power grid status data, and improve the response speed of distributed photovoltaic power generation systems; when detecting that reverse flow is about to occur, it can quickly send adjustment instructions to the control and optimization dispatch component to achieve dynamic adjustment of photovoltaic output and avoid the occurrence of reverse flow. Significance achieved: Through rapid response and dynamic adjustment, ensure the stable operation of the power grid, reduce power grid failures and losses caused by reverse flow; through real-time monitoring and rapid response, improve the reliability and safety of distributed photovoltaic power generation systems, and reduce operation and maintenance costs. The control and optimization dispatch component optimizes the dispatch strategy of photovoltaic power generation according to the predicted load demand and power grid capacity, as well as the adjustment instructions of the reverse flow detection and response component, and realizes dynamic adjustment of photovoltaic output; by sending control signals to photovoltaic power generation equipment, it realizes precise control of photovoltaic power generation output and ensures the efficient operation of distributed photovoltaic power generation systems. Significance achieved: By optimizing scheduling strategies and precise control, the absorption rate of photovoltaic power generation can be improved, the phenomenon of abandoned light can be reduced, and the overall efficiency of distributed photovoltaic power generation systems can be improved; through automated and intelligent scheduling strategies, human intervention can be reduced, operation and maintenance costs can be reduced, and the economic benefits of distributed photovoltaic power generation systems can be improved.

[0069] In summary, this embodiment improves the absorption rate of photovoltaic power generation, reduces the phenomenon of abandoned light, and improves the economic benefits of photovoltaic power generation through accurate load forecasting and dynamic adjustment; through real-time monitoring and rapid response, it ensures the stable operation of the power grid and reduces power grid failures and losses caused by reverse flow; through the collaborative work of multiple components, it improves the automation level of distributed photovoltaic power generation systems, reduces human intervention, and improves the reliability and safety of distributed photovoltaic power generation systems; through optimized scheduling strategies and precise control, it reduces operation and maintenance costs and improves the economic benefits of distributed photovoltaic power generation systems. This embodiment can achieve efficient, stable, and economical operation, bring higher economic benefits to investors, and provide strong guarantees for the stable operation of the power grid.

[0070] Example 2: Figure 2 As shown, based on Example 1, the load forecasting and optimization component provided by the embodiment of the present invention includes:

[0071] The data preprocessing module is responsible for acquiring data from multiple data sources, identifying and removing outliers and noise data, unifying data of different dimensions to the same scale, and obtaining preprocessed data;

[0072] The feature extraction module is responsible for extracting the load change curve over the past period of time to reflect the time series characteristics of the load; analyzing the user's electricity consumption habits to extract the electricity consumption characteristics during peak and off-peak hours; extracting meteorological parameters such as temperature, humidity, wind speed and sunshine intensity as important features affecting photovoltaic power generation; and selecting the most influential features for load forecasting and grid capacity forecasting from the time series features, electricity consumption features and important features to obtain the model training feature set;

[0073] The model building module is responsible for dividing the model training feature set into training set, validation set and test set, using the training set data to train the support vector machine, adjusting the support vector machine parameters through gradient descent to minimize the prediction error; using the validation set data to evaluate the performance of the support vector machine, and using the mean square error indicator to measure the prediction accuracy of the support vector machine; adjusting the support vector machine parameters and structure based on the evaluation results;

[0074] Among them, the equation for adjusting the support vector machine parameters is

[0075] Using a kernel function Calculate the similarity between training set data;

[0076] The original problem is transformed into a dual problem. The original problem refers to directly using the linear inner product to calculate the similarity between the training set data before introducing the kernel technique and the loss function, and trying to find an optimal hyperplane to separate the training set data of different categories; the goal of the dual problem is to minimize the following function:

[0077]

[0078] In the formula, is the dual variable vector, It is i The labels of the training set data, is the kernel function; No. j Labels of training set data; is the total number of training set data; is a dual variable;

[0079] Adjusting the dual variable via gradient descent , it is necessary to calculate the gradient of the dual problem and update the dual variables. The specific steps are as follows:

[0080] For the dual variable , the gradient is:

[0081]

[0082] In the formula, is the loss function;

[0083] Dual Variable The update formula is:

[0084]

[0085] In the formula, is the learning rate, which controls the step size of each parameter update;

[0086] After updating the dual variables, the weight vector can be calculated by the following formula and the bias term b :

[0087]

[0088]

[0089] In the formula, is the number of support vectors, is the set of support vectors.

[0090] The working principle and beneficial effects of the above technical solution are as follows: the data preprocessing module of this embodiment obtains data from multiple data sources, identifies and removes outliers and noise data, unifies data of different dimensions to the same scale, and obtains preprocessed data; the feature extraction module extracts the load change curve over the past period of time to reflect the time series characteristics of the load; analyzes the user's electricity consumption habits, and extracts the electricity consumption characteristics during peak hours and valley hours; extracts meteorological parameters such as temperature, humidity, wind speed and sunshine intensity as important features affecting photovoltaic power generation; from the time series characteristics, electricity consumption characteristics and important characteristics, selects the most influential features for load forecasting and grid capacity forecasting to obtain a model training feature set; the model construction module divides the model training feature set into a training set, a validation set and a test set, uses the training set data to train the support vector machine, adjusts the support vector machine parameters by gradient descent, and minimizes the prediction error; uses the validation set data to evaluate the performance of the support vector machine, and uses the mean square error indicator to measure the prediction accuracy of the support vector machine; according to the evaluation results, adjusts the support vector machine parameters and structure. The data preprocessing module of the above scheme obtains data from multiple data sources, identifies and removes outliers and noise data, unifies data of different dimensions to the same scale, and ensures data quality and consistency. Significance: It provides a clean and standardized data foundation for feature extraction and model training, reduces the impact of data noise on model performance, and improves the accuracy and reliability of prediction. The feature extraction module extracts the load change curve over the past period of time, analyzes the user's electricity usage habits, extracts important features such as meteorological parameters, and selects the most influential features for load forecasting and grid capacity forecasting. Significance: By deeply analyzing the time series characteristics of the load and the user's electricity usage habits, and considering meteorological factors, it can more accurately capture the laws and influencing factors of load changes, provide a high-quality feature set for model training, and enhance the model's prediction ability. Model construction module: The model training feature set is divided into a training set, a validation set, and a test set, and training is performed using a support vector machine. The parameters are adjusted by gradient descent to minimize the prediction error, evaluate the performance of the model, and adjust the parameters and structure. Significance: Through the scientific model training and verification process, the generalization ability and prediction accuracy of the model are guaranteed, which can effectively respond to the load changes in the actual power grid, provide reliable technical support for the optimal scheduling of the power grid, and improve the operation efficiency and stability of the power grid.

[0091] In summary, the load prediction and optimization component of this embodiment realizes accurate prediction of load changes and optimal configuration of power grid capacity through the organic combination of three modules: data preprocessing, feature extraction and model construction, providing strong technical support for the intelligent management and efficient operation of the power grid.

[0092] Example 3: Figure 3As shown, based on Example 1, the backflow detection and response component provided in the embodiment of the present invention includes:

[0093] The reverse flow detection module is responsible for continuously monitoring the key parameters of the power grid, such as voltage, current and frequency, and identifying typical characteristics such as abnormal changes in the current direction of reverse flow in the power grid; combining load demand and power grid capacity forecast, it evaluates the probability and impact of reverse flow;

[0094] The reverse flow decision module is responsible for detecting that a reverse flow is about to occur. The reverse flow detection and response component will immediately issue a warning signal and start the decision-making mechanism; generate a reverse flow warning signal to notify the control and optimization scheduling component of the imminent reverse flow; combine the load demand and grid capacity forecast to evaluate the feasibility of adjusting the photovoltaic output and obtain the reverse flow decision result;

[0095] The instruction generation module is responsible for generating specific instructions for adjusting photovoltaic output according to the results of the backflow risk assessment, and sending the instructions to the control and optimization scheduling components in real time.

[0096] The working principle and beneficial effects of the above technical solution are as follows: the reverse flow detection module of this embodiment continuously monitors the key parameters such as voltage, current and frequency of the power grid, identifies the typical characteristics such as abnormal changes in the current direction of the reverse flow in the power grid; combines the load demand and the power grid capacity forecast to evaluate the probability and impact of the reverse flow; when the reverse flow decision module detects that the reverse flow is about to occur, the reverse flow detection and response component will immediately issue an early warning signal and start the decision-making mechanism; generates a reverse flow early warning signal to notify the control and optimization scheduling component that the reverse flow is about to occur; combines the load demand and the power grid capacity forecast to evaluate the feasibility of adjusting the photovoltaic output and obtain the reverse flow decision result; the instruction generation module generates a specific instruction to adjust the photovoltaic output according to the reverse flow risk assessment result, and sends the instruction to the control and optimization scheduling component in real time. The reverse flow detection module of the above solution continuously monitors the key parameters such as voltage, current and frequency of the power grid to ensure the real-time and accuracy of the data; through feature extraction technology, it identifies the typical characteristics of the reverse flow in the power grid, such as abnormal changes in the current direction; combines the load demand and the power grid capacity forecast to evaluate the probability and impact of the reverse flow, and provides a scientific basis for decision-making. Significance achieved: Through real-time monitoring and feature recognition, early warning can be given before reverse flow occurs to avoid damage to the power grid caused by reverse flow; accurate data support is provided for decision-making and control to ensure that the distributed photovoltaic power generation system can respond to reverse flow risks in a timely manner. When the reverse flow decision module detects that reverse flow is about to occur, it immediately issues a warning signal to notify the control and optimization scheduling component that reverse flow is about to occur; combined with load demand and grid capacity forecast, the feasibility of adjusting photovoltaic output is evaluated to obtain the reverse flow decision result. Significance achieved: Through warning signals and decision-making mechanisms, it is possible to quickly respond to reverse flow risks and reduce the possibility of reverse flow; combined with load demand and grid capacity forecast, the scheduling strategy of photovoltaic power generation is optimized to ensure the stable operation of the power grid. The instruction generation module generates specific instructions for adjusting photovoltaic output based on the reverse flow risk assessment results, such as reducing photovoltaic power generation output or suspending the operation of some photovoltaic equipment; through a reliable communication mechanism, the instructions are sent to the control and optimization scheduling component in real time to ensure the timeliness and accuracy of the instructions. Significance achieved: By generating specific adjustment instructions, the output of photovoltaic power generation equipment can be accurately controlled to avoid the occurrence of backflow; by sending instructions in real time, it is ensured that the distributed photovoltaic power generation system can respond to the backflow risk in a timely manner and ensure the safety and stability of the power grid.

[0097] In summary, through real-time monitoring, feature recognition and rapid decision-making, the distributed photovoltaic power generation system in this embodiment can effectively warn and take response measures before the reverse flow occurs, so as to avoid the damage to the power grid caused by the reverse flow phenomenon; combined with the load demand and power grid capacity forecast, the distributed photovoltaic power generation system can optimize the scheduling strategy of photovoltaic power generation and ensure the stable operation of the power grid; through precise control and real-time response, the distributed photovoltaic power generation system can effectively ensure the safety and stability of the power grid and improve the reliability and economy of the distributed photovoltaic power generation system. In short, the various modules of the reverse flow detection and response component jointly ensure the safe, stable and efficient operation of the distributed photovoltaic power generation grid-connected anti-reverse flow control distributed photovoltaic power generation system.

[0098] Example 4: Figure 4 As shown, based on Example 3, the backflow detection module provided in this embodiment of the present invention includes:

[0099] The probability evaluation submodule is responsible for constructing a dynamic Bayesian network model using variables such as load demand, grid capacity, and reverse flow as nodes. Each node represents a random variable, and the connection between nodes represents the dependency relationship between variables. The parameters of the dynamic Bayesian network model are learned using historical variable data to determine the conditional probability distribution between nodes and calculate the probability distribution of reverse flow under the current load demand and grid capacity conditions.

[0100] The impact assessment submodule is responsible for identifying key nodes and vulnerable links in the power grid, assessing the impact of reverse flow on the stability of the power grid structure; simulating the operating status of the equipment when reverse flow occurs, assessing parameters such as thermal stress and mechanical stress of the equipment, and predicting the life and failure risk of the equipment; constructing a causal relationship map between each device and node in the power grid, and through the causal map, assessing the chain reaction that may be caused by the reverse flow, and predicting the potential fault diffusion path and impact range;

[0101] The comprehensive assessment submodule is responsible for dynamically adjusting the weights of probability assessment and impact assessment based on historical assessment results and actual operation feedback; and weightedly integrating the results of probability assessment and impact assessment to generate a backflow risk index.

[0102] The working principle and beneficial effects of the above technical solution are as follows: the probability assessment submodule of this embodiment takes variables such as load demand, grid capacity and reverse flow occurrence as nodes to construct a dynamic Bayesian network model, where each node represents a random variable, and the connection between nodes represents the dependency relationship between variables; the parameters of the dynamic Bayesian network model are learned using historical variable data to determine the conditional probability distribution between nodes, and calculate the probability distribution of reverse flow occurrence under the current load demand and grid capacity conditions; the impact assessment submodule identifies key nodes and vulnerable links in the grid, and assesses the impact of reverse flow on the stability of the grid structure; simulates the operating status of the equipment when reverse flow occurs, assesses parameters such as thermal stress and mechanical stress of the equipment, and predicts the life and failure risk of the equipment; constructs a causal relationship map between each device and node in the grid, and through the causal map, assesses the chain reaction that may be caused by the reverse flow, and predicts the potential fault diffusion path and impact range; the comprehensive assessment submodule dynamically adjusts the weights of probability assessment and impact assessment based on historical assessment results and actual operation feedback; the results of probability assessment and impact assessment are weighted and integrated to generate a reverse flow risk index. The probability assessment submodule of the above scheme can capture the dynamic characteristics of load demand and grid capacity changing over time by constructing a dynamic Bayesian network model, so as to more accurately describe the probability of reverse flow; the model parameters are learned by using historical data to determine the conditional probability distribution between nodes, thereby improving the prediction accuracy of the model; in real-time operation, the probability distribution of reverse flow under the current load demand and grid capacity conditions can be quickly calculated to provide data support for real-time decision-making. Significance achieved: Through the dynamic Bayesian network model, the probability of reverse flow can be predicted more accurately, providing reliable data support for decision-making; real-time probability inference can timely reflect the changes in the grid state, help grid managers respond quickly and reduce potential risks. The impact assessment submodule identifies the key nodes and vulnerable links in the grid, evaluates the impact of reverse flow on the stability of the grid structure, and helps managers understand the weak links of the grid; simulates the operating status of the equipment when reverse flow occurs, evaluates the parameters such as thermal stress and mechanical stress of the equipment, predicts the life and failure risk of the equipment, and discovers potential problems in advance; constructs a causal relationship map between the equipment and nodes in the grid, evaluates the chain reaction that may be caused by reverse flow, predicts the potential fault diffusion path and impact range, and helps managers formulate preventive measures. Significance achieved: Through multi-scale evaluation methods, the impact of reverse flow on the power grid can be comprehensively evaluated to ensure the comprehensiveness and accuracy of the evaluation results; through physical model simulation and cause-and-effect diagram analysis, potential fault risks can be discovered in advance, helping power grid managers to take timely measures to reduce the possibility of faults.The comprehensive evaluation submodule uses reinforcement learning methods to dynamically adjust the weights of probability evaluation and impact assessment according to historical evaluation results and actual operation feedback to improve the accuracy of comprehensive evaluation; weighted synthesis of the results of probability evaluation and impact assessment generates a reverse flow risk index to help managers intuitively understand the risk status of the power grid; based on the reverse flow risk index, early warning signals are generated in real time, and decision support is provided to help managers respond quickly and ensure the safe and stable operation of the power grid. Significance achieved: Through adaptive weight adjustment, the weights of evaluation indicators can be dynamically adjusted according to real-time operation conditions to ensure the accuracy and real-time nature of the evaluation results; through comprehensive evaluation, a reverse flow risk index can be generated to help managers fully understand the risk status of the power grid and provide a scientific basis for decision-making; through real-time early warning and decision support, it can help power grid managers respond quickly, reduce potential risks, and ensure the safe and stable operation of the power grid.

[0103] In summary, the backflow detection module of this embodiment can comprehensively and accurately evaluate the probability and impact of backflow, provide a reliable decision-making basis for power grid managers, and ensure the safe and stable operation of the power grid.

[0104] Example 5: Figure 5 As shown, based on Example 1, the control and optimization scheduling component provided by the embodiment of the present invention includes:

[0105] The result evaluation module is responsible for analyzing the predicted load demand and obtaining the peak and valley periods of electricity consumption of users in the future; analyzing the grid capacity and evaluating the grid's carrying capacity and potential risks;

[0106] The strategy formulation module is responsible for setting the maximum utilization rate of photovoltaic power generation, minimizing grid load fluctuations and reducing the risk of reverse flow. Based on the settings and analysis results, the scheduling strategy of photovoltaic power generation is formulated, including dynamic adjustment of photovoltaic output, priority setting and emergency plans.

[0107] The strategy execution module is responsible for sending control signals to photovoltaic power generation equipment and adjusting output according to the formulated scheduling strategy; real-time monitoring of the operating status of photovoltaic power generation equipment and the status of the power grid, and timely adjustment of the scheduling strategy.

[0108] The working principle and beneficial effects of the above technical solution are as follows: the result evaluation module of this embodiment analyzes the predicted load demand and obtains the peak and valley periods of user electricity consumption in the future; analyzes the grid capacity and evaluates the grid's carrying capacity and potential risks; the strategy formulation module sets the maximum utilization rate of photovoltaic power generation, minimizes grid load fluctuations and reduces the risk of reverse flow, etc.; based on the setting and analysis results, formulates the scheduling strategy of photovoltaic power generation, including dynamic adjustment of photovoltaic output, priority setting and emergency plan; the strategy execution module sends a control signal to the photovoltaic power generation equipment according to the formulated scheduling strategy to adjust the output; monitors the operating status and grid status of the photovoltaic power generation equipment in real time, and adjusts the scheduling strategy in time. The above scheme accurately obtains the peak and valley periods of user electricity consumption in the future by analyzing the predicted load demand; evaluates the grid's carrying capacity and potential risks, and identifies the grid's weak links and potential fault points. The significance achieved: By accurately predicting load demand and grid capacity, optimizing resource allocation, and improving the utilization rate of photovoltaic power generation; identifying the potential risks of the grid in advance, formulating corresponding emergency plans, and reducing grid failures and reverse flow risks. The strategy formulation module sets goals such as maximizing the utilization rate of photovoltaic power generation, minimizing grid load fluctuations, and reducing the risk of reverse flow; based on the settings and analysis results, formulates the scheduling strategy for photovoltaic power generation, including dynamic adjustment of photovoltaic output, priority setting, and emergency plans. Significance achieved: By formulating a reasonable scheduling strategy, ensure the efficient operation of photovoltaic power generation equipment under different time periods and load conditions; by dynamically adjusting photovoltaic output and priority setting, maintain the stable operation of the power grid and reduce load fluctuations and reverse flow risks. The strategy execution module sends control signals to photovoltaic power generation equipment according to the formulated scheduling strategy to adjust its output; monitor the operating status of photovoltaic power generation equipment and the status of the power grid in real time, and adjust the scheduling strategy in time. Significance achieved: Through real-time monitoring and feedback, ensure the timely implementation of the scheduling strategy, respond to emergencies, and maintain the stable operation of the power grid; dynamically adjust the scheduling strategy according to the power grid status and the operation of photovoltaic power generation equipment to improve the flexibility and adaptability of distributed photovoltaic power generation systems.

[0109] In summary, this embodiment improves the overall efficiency of the distributed photovoltaic power generation system and maximizes the utilization rate of photovoltaic power generation through the coordinated work of various modules; through optimizing the scheduling strategy and real-time monitoring, it ensures the stable operation of the power grid and reduces the risk of load fluctuation and reverse flow; through optimizing resource allocation and efficient scheduling, it improves the user's power experience and ensures the reliability and stability of power supply. Together, we can achieve the optimized operation of the distributed photovoltaic power generation system, improve the efficiency and stability of the distributed photovoltaic power generation system, and ensure the safety of the power grid and the user's power experience.

[0110] Example 6: Figure 6 As shown, based on Example 5, the policy formulation module provided in this embodiment of the present invention includes:

[0111] The target setting submodule is responsible for adjusting the operating status of photovoltaic power generation equipment by optimizing the dispatching strategy; dynamically adjusting photovoltaic output to balance the grid load; dispatching photovoltaic power generation equipment to avoid grid reverse flow;

[0112] The dynamic adjustment submodule is responsible for obtaining the load demand and the current and future capacity of the power grid, dynamically adjusting the output of photovoltaic power generation equipment to adapt to changes in load demand; real-time monitoring of the operating status of photovoltaic power generation equipment, including the health status of the equipment and power generation efficiency; setting the priority of the equipment according to factors such as the operating status of the equipment, geographical location and power generation efficiency, and giving priority to dispatching high-priority equipment when the power grid load fluctuates;

[0113] The emergency plan submodule is responsible for automatically starting the backup power supply when the power grid fails or the equipment fails, and transferring the load from the fault area to other areas with normal operation.

[0114] The working principle and beneficial effects of the above technical scheme are as follows: the target setting submodule of this embodiment adjusts the operating state of photovoltaic power generation equipment by optimizing the scheduling strategy; dynamically adjusts the photovoltaic output to balance the grid load; dispatches photovoltaic power generation equipment to avoid grid reverse flow; the dynamic adjustment submodule obtains the load demand and the current capacity and future capacity demand of the grid, dynamically adjusts the output of photovoltaic power generation equipment to adapt to the changes in load demand; monitors the operating state of photovoltaic power generation equipment in real time, including the health status and power generation efficiency of the equipment; sets the priority of the equipment according to the operating state, geographical location and power generation efficiency of the equipment, and prioritizes the high-priority equipment when the grid load fluctuates; the emergency plan submodule automatically starts the backup power supply when the grid fails or the equipment fails; and transfers the load from the fault area to other normal operating areas. The target setting submodule of the above scheme optimizes the operating strategy of photovoltaic power generation equipment through advanced algorithms and models to ensure that the equipment operates in the best state; dynamically adjusts the output of photovoltaic power generation equipment according to the real-time changes of the grid load to maintain the balance of the grid; through precise scheduling, prevents photovoltaic power generation equipment from generating reverse flow when the grid load is low, and protects the safety of the grid. Significance achieved: By optimizing scheduling, the overall efficiency of distributed photovoltaic power generation systems can be improved and energy waste can be reduced; by dynamically adjusting photovoltaic output, the grid load can be effectively balanced and the stability and reliability of the grid can be enhanced; optimizing scheduling strategies can reduce unnecessary equipment operation and reduce operation and maintenance costs. The dynamic adjustment submodule adjusts the output of photovoltaic power generation equipment in real time according to changes in load demand to ensure the balance of grid supply and demand; monitors the operating status of photovoltaic power generation equipment in real time, including the health status and power generation efficiency of the equipment, to provide data support for scheduling decisions; sets the priority of equipment according to factors such as equipment operating status, geographical location and power generation efficiency to ensure that high-efficiency and high-priority equipment are dispatched first when the grid load fluctuates. Significance achieved: Real-time monitoring and dynamic adjustment can quickly respond to load changes and improve the response speed and flexibility of the system; optimize resource allocation through priority setting, ensure the priority use of high-efficiency equipment, and improve the overall power generation efficiency; timely adjust the operation strategy through real-time monitoring of equipment health status to extend the service life of the equipment. The emergency plan submodule automatically starts the backup power supply when the grid or equipment fails to ensure the continuity of power supply; transfers the load from the fault area to other normal operating areas to reduce the impact of the fault on the grid. Significance achieved: Improve the reliability and risk resistance of the system by automatically starting the backup power supply and load transfer; in the event of a fault, ensure the continuity of power supply, avoid large-scale power outages, and guarantee the electricity needs of users; through rapid response and load transfer, reduce the impact of faults on the power grid and users, and improve the overall stability of the system.

[0115] In summary, this embodiment realizes the optimized scheduling, dynamic adjustment and emergency management of the distributed photovoltaic power generation system, which not only improves the efficiency and stability of the distributed photovoltaic power generation system, but also enhances the reliability and risk resistance of the distributed photovoltaic power generation system, and provides a strong guarantee for the safe and stable operation of the power grid.

[0116] Example 7: Figure 7 As shown, based on Example 6, the dynamic adjustment submodule provided in this embodiment of the present invention includes:

[0117] The condition setting unit is responsible for setting the conditions for dynamically adjusting the start-up of photovoltaic power generation equipment, where the future capacity is greater than the current capacity; when the conditions are met, the real-time change in load demand is obtained;

[0118] The value acquisition unit is responsible for calling the output mapping table of the photovoltaic power generation equipment. The mapping table records the corresponding relationship between the real-time change of the load demand and the output, and calls the corresponding output value according to the real-time change; the mapping table is calculated based on the historical load demand and the output of the photovoltaic power generation equipment;

[0119] The adjustment sequence unit is responsible for forming an adjustment strategy based on the load demand and the corresponding output, and at the same time, adjusting the output sequence according to the priority of the photovoltaic power generation equipment.

[0120] The working principle and beneficial effects of the above technical solution are as follows: the condition setting unit of this embodiment sets the condition for dynamically adjusting the photovoltaic power generation equipment to start processing, and the condition is that the future capacity is greater than the current capacity; when the condition is met, the real-time change of the load demand is obtained; the value acquisition unit calls the output mapping table of the photovoltaic power generation equipment, and the mapping table records the corresponding relationship between the real-time change of the load demand and the output, and calls the corresponding output value according to the real-time change; the mapping table is calculated based on the historical load demand and the output of the photovoltaic power generation equipment; the adjustment sequence unit forms an adjustment strategy with the load demand and the corresponding output, and at the same time uses the priority of the photovoltaic power generation equipment as the order of adjusting the output. The condition setting unit of the above scheme can set the condition for starting dynamic adjustment according to the comparison between the future capacity and the current capacity; when the condition is met, the dynamic change of the load demand can be obtained in real time, ensuring that the distributed photovoltaic power generation system responds quickly to environmental changes. The significance achieved: through the preset conditions, the distributed photovoltaic power generation system can start the adjustment at the best time to reduce unnecessary energy waste; ensure that the adjustment is made when the capacity is sufficient to avoid overload or inefficiency of the distributed photovoltaic power generation system caused by insufficient capacity. The numerical acquisition unit can quickly retrieve the output mapping table of photovoltaic power generation equipment according to the change in real-time load demand and obtain the corresponding output value; the mapping table is calculated based on historical load demand and output data to ensure the accuracy and reliability of the data. Significance achieved: Through the analysis of historical data, the distributed photovoltaic power generation system can more accurately predict future output demand and improve power generation efficiency; make decisions based on actual data, reduce human errors, and improve the intelligence level of distributed photovoltaic power generation systems. The adjustment sequence unit can generate specific adjustment strategies based on load demand and corresponding output values; determine the order of adjusting output according to the priority of photovoltaic power generation equipment to ensure that key equipment is adjusted first. Significance achieved: Through priority management, the distributed photovoltaic power generation system can allocate resources more reasonably and ensure the stable operation of key equipment; orderly adjustment strategies can reduce the risk of failure of distributed photovoltaic power generation systems and improve the reliability of overall operation.

[0121] In summary, the dynamic adjustment submodule of this embodiment can realize intelligent management and optimized operation of photovoltaic power generation equipment, improve energy utilization efficiency, reduce energy waste, and enhance system reliability and stability. It not only helps to reduce operating costs, but also contributes to environmental protection and sustainable development.

[0122] Example 8: Figure 8 As shown, based on Example 7, the numerical value acquisition unit provided in this embodiment of the present invention includes:

[0123] The numerical calculation subunit is responsible for calculating the load demand change at each time point based on the collected load demand data. The load demand change is calculated by the load demand difference between the previous and next time points. The output value corresponding to each load demand change is determined based on the collected output data. The output value is determined by the corresponding relationship in the historical data.

[0124] The table creation subunit is responsible for creating a table whose columns include load demand change and corresponding output value; the calculated load demand change and corresponding output value are filled into the table;

[0125] The data sorting subunit is responsible for sorting the data in the table from small to large according to the size of the load demand change; for the missing load demand change and output value, the interpolation method is used to supplement the data in the table in a personalized way.

[0126] The working principle and beneficial effects of the above technical solution are as follows: the numerical calculation subunit of this embodiment calculates the load demand change at each time point according to the collected load demand data, and the load demand change is calculated by the load demand difference between the previous and next time points; the output value corresponding to each load demand change is determined according to the collected output data, and the output value is determined by the corresponding relationship in the historical data; the table establishment subunit establishes a table, and the columns of the table include the load demand change and the corresponding output value; the calculated load demand change and the corresponding output value are filled into the table; the data sorting subunit sorts the data in the table from small to large according to the size of the load demand change, and the data in the table; for the missing load demand change and output value, the interpolation method is used to perform personalized supplement. The numerical calculation subunit of the above scheme calculates the load demand change by the load demand difference between the previous and next time points to ensure the real-time and accuracy of the data; the output value corresponding to each load demand change is determined according to the corresponding relationship in the historical data to ensure the rationality and reliability of the output value. The significance achieved: ensure the real-time calculation of load demand change and output value, and provide timely support for subsequent data processing and decision-making; through the combination of historical data and real-time data, ensure the accuracy of output value, and avoid decision-making deviation caused by data errors. The table establishment subunit establishes a structured table, including load demand change and corresponding output value, which is convenient for data storage and subsequent processing; the calculated load demand change and corresponding output value are filled into the table to ensure the integrity and consistency of the data. The significance achieved: storing data in the form of a table is convenient for long-term storage and management of data; the table-structured data is convenient for subsequent data analysis and visualization, and improves the readability and operability of data. The data sorting subunit sorts the data in the table from small to large according to the size of the load demand change, ensuring the order and logic of the data; for the missing load demand change and output value, the interpolation method is used to supplement the personalized data to ensure the continuity and integrity of the data. The significance achieved: ensure the order of data through sorting, which is convenient for subsequent data analysis and processing; supplement the missing data through the interpolation method to ensure the integrity of the data and avoid analysis errors caused by missing data.

[0127] In summary, the various sub-units of this embodiment work together to ensure the real-time, accuracy, integrity and order of data, providing a solid foundation for subsequent data analysis, decision support and system optimization. Specifically, the high quality of data is ensured through real-time calculation, historical data combination and interpolation supplementation; accurate data support helps optimize the decision-making process and improve the operating efficiency and reliability of distributed photovoltaic power generation systems; complete and accurate data support helps enhance the stability of the system and reduce system failures caused by data problems.

[0128] Example 9: Fig. 9 As shown, based on Example 8, the data sorting subunit provided in this embodiment of the present invention includes:

[0129] The grid division component is responsible for setting a coordinate system and dividing the load demand change area into multiple grids. Each grid contains a certain range of load demand change, which is used to preliminarily locate the position of the missing output value; obtain the coordinate longitude and latitude information of each grid, and determine the load demand change and output value within the grid;

[0130] The data correction component is responsible for obtaining information related to the output value through historical data records, which is used to preliminarily determine the candidate value of the missing output value; divide the second type of grid, and correct the load demand change and output value;

[0131] The interpolation calculation component is responsible for solving the optimal missing data value based on the corrected load demand change and output value using an interpolation algorithm (such as the Dijkstra algorithm); and adjusting the grid size and related parameters based on the interpolation results.

[0132] The working principle and beneficial effects of the above technical solution are as follows: the grid division component of this embodiment is responsible for setting a coordinate system and dividing the load demand change area into multiple grids, each grid contains a certain range of load demand change, which is used to preliminarily locate the position of the missing output value; obtain the coordinate longitude and latitude information of each grid, and determine the load demand change and output value within the grid; the data correction component is responsible for obtaining information related to the output value through historical data records, which is used to preliminarily determine the candidate value of the missing output value; divide the second type of grid, and correct the load demand change and output value; the interpolation calculation component is responsible for using the interpolation algorithm (such as the Dijkstra algorithm) to solve the optimal missing data value on the corrected load demand change and output value; adjust the grid size and related parameters according to the interpolation result. The grid division component of the above scheme sets a unified coordinate system to ensure that all data are processed under the same reference frame; the load demand change area is divided into multiple grids, each grid contains a certain range of load demand change; through grid division, the location of the missing output value is preliminarily located to provide a basis for subsequent processing; the coordinate longitude and latitude information of each grid is obtained to facilitate the integration and analysis of the geographic information system; the load demand change and output value within the grid are determined to provide a data basis for subsequent correction and interpolation. The significance achieved: Through grid division, the location of missing data can be more accurately located and errors can be reduced; the unified coordinate system and grid division make data processing more standardized and consistent; the longitude and latitude information is obtained to facilitate integration with the geographic information system and enhance the geographic analysis capability of the data. The data correction component obtains information related to the output value through historical data records and provides candidate values ​​for the missing output value; on the basis of the preliminary grid division, the second type of grid is further divided to refine the correction of the load demand change and output value; the load demand change and output value are corrected to improve the accuracy and reliability of the data. Significance achieved: Through historical data and the second type of grid division, the load demand change and output value can be corrected more accurately to reduce data errors; the corrected data is more reliable, providing a high-quality data basis for subsequent interpolation calculations; the data correction component optimizes the data processing process, making data processing more efficient and accurate. The interpolation calculation component uses an interpolation algorithm (such as the Dijkstra algorithm) to solve the optimal missing data value and fill in the data gaps; adjust the grid size and related parameters according to the interpolation results to optimize the data processing effect. Significance achieved: Through interpolation calculations, missing data can be effectively filled and data integrity can be improved; adjust the grid size and related parameters to optimize data distribution, making data processing more reasonable and efficient; the interpolation calculation component improves the quality of the data, making the data processing results more accurate and reliable.

[0133] In summary, this embodiment achieves refinement, standardization and optimization of data processing. The grid division component provides preliminary positioning and standardization processing, the data correction component improves the accuracy and reliability of the data, and the interpolation calculation component fills the data gaps and optimizes the data distribution. Together, the overall quality and efficiency of data processing are improved, providing a solid foundation for analysis and decision-making.

[0134] Example 10: Fig.10 As shown, based on Embodiment 1 to Embodiment 9, the distributed photovoltaic power generation grid-connected anti-backflow control method provided by the embodiment of the present invention comprises the following steps:

[0135] S100: Receives input from multiple data sources such as grid dispatch data, user electricity usage habits, and weather forecasts, and predicts load demand and grid capacity in the future in real time, providing decision-making basis for control and dispatch components;

[0136] S200: receiving the load demand and grid capacity predicted by the load prediction and optimization component, as well as real-time grid status data; when detecting that reverse flow is about to occur in the grid, sending an instruction to adjust the photovoltaic output to the control and optimization scheduling component;

[0137] S300: receiving predicted load demand and grid capacity, and instructions for adjusting photovoltaic output; optimizing the scheduling strategy of photovoltaic power generation, dynamically adjusting the output of photovoltaic power generation, and sending control signals to photovoltaic power generation equipment.

[0138] The working principle and beneficial effects of the above technical solution are as follows: the present embodiment first receives inputs from multiple data sources such as grid dispatching data, user electricity usage habits and weather forecasts, and predicts load demand and grid capacity in the future in real time, providing a decision basis for the control and dispatching components; secondly, receives load demand and grid capacity predicted by the load forecasting and optimization components, as well as real-time grid status data; when it is detected that reverse flow is about to occur in the grid, sends instructions to the control and optimization dispatching components to adjust photovoltaic output; finally, receives predicted load demand and grid capacity, and instructions to adjust photovoltaic output; optimizes the dispatching strategy of photovoltaic power generation, dynamically adjusts the output of photovoltaic power generation, and sends control signals to photovoltaic power generation equipment. Step S100 of the above scheme, data input and load forecasting, can comprehensively and accurately predict load demand and grid capacity in the future by receiving multiple data sources such as grid dispatching data, user electricity usage habits and weather forecasts; the real-time prediction function ensures the timeliness and accuracy of data, and provides a reliable decision basis for subsequent control and dispatching. Significance: Through the fusion and real-time prediction of multi-source data, it is possible to quickly respond to changes in the power grid and improve the efficiency and accuracy of decision-making; accurate load prediction helps to optimize the configuration of power resources, reduce unnecessary energy waste, and improve the overall operation efficiency of the power grid. Step S200 Load prediction and optimization scheduling, by receiving the prediction data and real-time power grid status data of the load prediction and optimization components, it can monitor the power grid status in real time and detect the risk of reverse flow in time; when it is detected that reverse flow is about to occur, it can quickly send instructions to the control and optimization scheduling components to adjust the photovoltaic output to avoid the occurrence of reverse flow. Significance: Through real-time monitoring and dynamic adjustment, it can effectively prevent the occurrence of reverse flow and ensure the safe and stable operation of the power grid; timely adjustment instructions help improve the reliability of the power grid system and reduce power outages or equipment damage caused by reverse flow. Step S300 Photovoltaic power generation scheduling optimization, according to the predicted load demand and power grid capacity, as well as the instructions for adjusting photovoltaic output, optimize the scheduling strategy of photovoltaic power generation to ensure that the output of photovoltaic power generation matches the power grid demand; dynamically adjust the output of photovoltaic power generation, and send control signals to photovoltaic power generation equipment to ensure that the photovoltaic power generation system can flexibly respond to changes in the power grid. Significance: By optimizing scheduling strategies and dynamic control, the energy utilization rate of photovoltaic power generation can be improved and energy waste can be reduced; effective scheduling strategies and control methods can help promote the widespread application of renewable energy and promote the green transformation of the energy structure.

[0139] To sum up, the distributed photovoltaic power generation grid-connected anti-backflow control method of this embodiment not only improves the safety and reliability of the power grid, but also optimizes the allocation and utilization of energy resources, promotes the development of renewable energy, and has important technical significance and application value through technical means such as multi-source data fusion, real-time prediction, dynamic adjustment and optimized scheduling.

[0140] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of equivalent technologies of the present invention, the present invention is also intended to include these modifications and variations.

Claims

1. A distributed photovoltaic power generation grid-connected anti-backflow control system, characterized in that: Include: The load forecasting and optimization component is responsible for receiving inputs from multiple data sources such as power grid dispatching data, user electricity consumption habits, and weather forecasts, and predicting load demand and power grid capacity in the future in real time, providing decision-making basis for the control and dispatching components; The reverse flow detection and response component is responsible for receiving the load demand and grid capacity predicted by the load prediction and optimization component, as well as real-time grid status data; when it detects that a reverse flow is about to occur in the grid, it sends instructions to the control and optimization scheduling component to adjust the photovoltaic output; The control and optimization dispatching component is responsible for receiving the predicted load demand and grid capacity, as well as the instructions for adjusting the photovoltaic output; optimizing the dispatching strategy of photovoltaic power generation, dynamically adjusting the photovoltaic power generation output, and sending control signals to photovoltaic power generation equipment; Load forecasting and optimization components, including: The data preprocessing module is responsible for acquiring data from multiple data sources, identifying and removing outliers and noise data, unifying data of different dimensions to the same scale, and obtaining preprocessed data; The feature extraction module is responsible for extracting the load change curve over the past period of time to reflect the time series characteristics of the load; Analyze users' electricity consumption habits and extract electricity consumption characteristics during peak and off-peak hours; extract temperature, humidity, wind speed and sunshine intensity meteorological parameters as important features affecting photovoltaic power generation; select the most influential features for load forecasting and grid capacity forecasting from time series features, electricity consumption features and important features to obtain the model training feature set; The model building module is responsible for dividing the model training feature set into training set, validation set and test set, using the training set data to train the support vector machine, and adjusting the support vector machine parameters through gradient descent to minimize the prediction error; Use validation set data to evaluate the performance of the support vector machine, and use the mean square error indicator to measure the prediction accuracy of the support vector machine; adjust the support vector machine parameters and structure based on the evaluation results; Among them, the equation for adjusting the support vector machine parameters is: Using a kernel function Calculate the similarity between training set data; The original problem is transformed into a dual problem. The original problem refers to directly using the linear inner product to calculate the similarity between the training set data before introducing the kernel technique and the loss function, and trying to find an optimal hyperplane to separate the training set data of different categories; the goal of the dual problem is to minimize the following function: In the formula, is the dual variable vector, is the label of the i-th training set data, is the kernel function; The label of the jth training set data; is the total number of training set data; is a dual variable; Adjusting the dual variable via gradient descent , it is necessary to calculate the gradient of the dual problem and update the dual variables. The specific steps are as follows: For the dual variable , the gradient is: In the formula, is the loss function; Dual Variable The update formula is: In the formula, is the learning rate, which controls the step size of each parameter update; After updating the dual variables, the weight vector is calculated by the following formula And the bias term b: In the formula, is the number of support vectors, is the set of support vectors; Backflow detection and response components, including: The reverse flow detection module is responsible for continuously monitoring the key parameters of the power grid, such as voltage, current and frequency, and identifying the typical characteristics of abnormal changes in the current direction of reverse flow in the power grid; combining load demand and power grid capacity forecast, it evaluates the probability and impact of reverse flow; The reverse flow decision module is responsible for detecting that a reverse flow is about to occur. The reverse flow detection and response component will immediately issue a warning signal and start the decision-making mechanism; generate a reverse flow warning signal to notify the control and optimization scheduling component of the imminent reverse flow; combine the load demand and grid capacity forecast to evaluate the feasibility of adjusting the photovoltaic output and obtain the reverse flow decision result; The instruction generation module is responsible for generating specific instructions for adjusting the photovoltaic output according to the results of the reverse flow risk assessment, and sending the instructions to the control and optimization scheduling component in real time; Backflow detection module, including: The probability evaluation submodule is responsible for constructing a dynamic Bayesian network model using load demand, grid capacity and reverse flow occurrence variables as nodes. Each node represents a random variable, and the connection between nodes represents the dependency relationship between variables. The parameters of the dynamic Bayesian network model are learned using historical variable data to determine the conditional probability distribution between nodes and calculate the probability distribution of reverse flow occurrence under the current load demand and grid capacity conditions. The impact assessment submodule is responsible for identifying key nodes and vulnerable links in the power grid, assessing the impact of reverse flow on the stability of the power grid structure; simulating the operating status of the equipment when reverse flow occurs, assessing the thermal stress and mechanical stress parameters of the equipment, and predicting the life and failure risk of the equipment; constructing a causal relationship map between each device and node in the power grid, and through the causal map, assessing the chain reaction that may be caused by the reverse flow, and predicting the potential fault diffusion path and impact range; The comprehensive assessment submodule is responsible for dynamically adjusting the weights of probability assessment and impact assessment based on historical assessment results and actual operation feedback; and weightedly integrating the results of probability assessment and impact assessment to generate a backflow risk index.

2. The distributed photovoltaic power generation grid-connected anti-backflow control system according to claim 1, characterized in that: Control and optimization scheduling components, including: The result evaluation module is responsible for analyzing the predicted load demand and obtaining the peak and valley periods of electricity consumption of users in the future; analyzing the grid capacity and evaluating the grid's carrying capacity and potential risks; The strategy formulation module is responsible for setting the maximum utilization rate of photovoltaic power generation, minimizing grid load fluctuations and reducing the risk of reverse flow; based on the settings and analysis results, the photovoltaic power generation scheduling strategy is formulated, including dynamic adjustment of photovoltaic output, priority setting and emergency plans; The strategy execution module is responsible for sending control signals to photovoltaic power generation equipment and adjusting output according to the formulated scheduling strategy; real-time monitoring of the operating status of photovoltaic power generation equipment and the status of the power grid, and timely adjustment of the scheduling strategy.

3. The distributed photovoltaic power generation grid-connected anti-backflow control system according to claim 2, characterized in that: Strategy formulation module, including: The target setting submodule is responsible for adjusting the operating status of photovoltaic power generation equipment by optimizing the dispatching strategy; dynamically adjusting photovoltaic output to balance the grid load; dispatching photovoltaic power generation equipment to avoid grid reverse flow; The dynamic adjustment submodule is responsible for obtaining the load demand and the current and future capacity of the power grid, dynamically adjusting the output of photovoltaic power generation equipment to adapt to changes in load demand; real-time monitoring of the operating status of photovoltaic power generation equipment, including the health status of the equipment and power generation efficiency; setting the priority of the equipment according to the operating status, geographical location and power generation efficiency factors of the equipment, and giving priority to dispatching high-priority equipment when the power grid load fluctuates; The emergency plan submodule is responsible for automatically starting the backup power supply when the power grid fails or the equipment fails, and transferring the load from the faulty area to other areas with normal operation.

4. The distributed photovoltaic power generation grid-connected anti-backflow control system according to claim 3, characterized in that: Dynamic adjustment submodule, including: The condition setting unit is responsible for setting the conditions for dynamically adjusting the start-up of photovoltaic power generation equipment, where the future capacity is greater than the current capacity; when the conditions are met, the real-time change in load demand is obtained; The value acquisition unit is responsible for calling the output mapping table of the photovoltaic power generation equipment. The mapping table records the corresponding relationship between the real-time change of load demand and the output, and calls the corresponding output value according to the real-time change; the mapping table is calculated based on the historical load demand and the output of the photovoltaic power generation equipment; The adjustment sequence unit is responsible for forming an adjustment strategy based on the load demand and the corresponding output, and at the same time, adjusting the output sequence according to the priority of the photovoltaic power generation equipment.

5. The distributed photovoltaic power generation grid-connected anti-backflow control system according to claim 4, characterized in that: The value acquisition unit includes: The numerical calculation subunit is responsible for calculating the load demand change at each time point based on the collected load demand data. The load demand change is calculated by the load demand difference between the previous and next time points. The output value corresponding to each load demand change is determined based on the collected output data. The output value is determined by the corresponding relationship in the historical data. The table creation subunit is responsible for creating a table whose columns include load demand change and corresponding output value; the calculated load demand change and corresponding output value are filled into the table; The data sorting subunit is responsible for sorting the data in the table from small to large according to the size of the load demand change; for the missing load demand change and output value, the interpolation method is used to supplement the data in the table in a personalized way.

6. The distributed photovoltaic power generation grid-connected anti-backflow control system according to claim 5, characterized in that: Data sorting subunit, including: The grid division component is responsible for setting a coordinate system and dividing the load demand change area into multiple grids. Each grid contains a certain range of load demand change, which is used to preliminarily locate the position of the missing output value; obtain the coordinate longitude and latitude information of each grid, and determine the load demand change and output value within the grid; The data correction component is responsible for obtaining information related to the output value through historical data records, which is used to preliminarily determine the candidate value of the missing output value; Divide the second type of grid and correct the load demand change and output value; The interpolation calculation component is responsible for using the interpolation algorithm to solve the optimal missing data value based on the corrected load demand change and output value; and adjusting the grid size and related parameters according to the interpolation results.

7. A distributed photovoltaic power generation grid-connected anti-backflow control method, characterized in that: The following steps are involved: Receive input from multiple data sources such as grid dispatch data, user electricity usage habits, and weather forecasts, and make real-time predictions of load demand and grid capacity in the future, providing decision-making basis for control and dispatch components; Receive the load demand and grid capacity predicted by the load forecasting and optimization component, as well as real-time grid status data; when detecting that reverse flow is about to occur in the grid, send instructions to the control and optimization scheduling component to adjust the photovoltaic output; Receive predicted load demand and grid capacity, and instructions for adjusting photovoltaic output; optimize photovoltaic power generation scheduling strategy, dynamically adjust photovoltaic power generation output, and send control signals to photovoltaic power generation equipment; The load forecasting and optimization components include: The data preprocessing module obtains data from multiple data sources, identifies and removes outliers and noise data, and unifies data of different dimensions to the same scale to obtain preprocessed data; The feature extraction module extracts the load variation curve in the past period of time to reflect the time series characteristics of the load; Analyze users' electricity consumption habits and extract electricity consumption characteristics during peak and off-peak hours; extract temperature, humidity, wind speed and sunshine intensity meteorological parameters as important features affecting photovoltaic power generation; select the most influential features for load forecasting and grid capacity forecasting from time series features, electricity consumption features and important features to obtain the model training feature set; The model building module divides the model training feature set into training set, validation set and test set, uses the training set data to train the support vector machine, and adjusts the support vector machine parameters through gradient descent to minimize the prediction error; Use validation set data to evaluate the performance of the support vector machine, and use the mean square error indicator to measure the prediction accuracy of the support vector machine; adjust the support vector machine parameters and structure based on the evaluation results; Among them, the equation for adjusting the support vector machine parameters is: Using a kernel function Calculate the similarity between training set data; The original problem is transformed into a dual problem. The original problem refers to directly using the linear inner product to calculate the similarity between the training set data before introducing the kernel technique and the loss function, and trying to find an optimal hyperplane to separate the training set data of different categories; the goal of the dual problem is to minimize the following function: In the formula, is the dual variable vector, is the label of the i-th training set data, is the kernel function; The label of the jth training set data; is the total number of training set data; is a dual variable; Adjusting the dual variable via gradient descent , it is necessary to calculate the gradient of the dual problem and update the dual variables. The specific steps are as follows: For the dual variable , the gradient is: In the formula, is the loss function; Dual Variable The update formula is: In the formula, is the learning rate, which controls the step size of each parameter update; After updating the dual variables, the weight vector is calculated by the following formula And the bias term b: In the formula, is the number of support vectors, is the set of support vectors; The process of sending instructions to the control and optimization scheduling component to adjust the photovoltaic output includes: The reverse flow detection module continuously monitors the key parameters of the grid voltage, current and frequency, identifies the typical characteristics of abnormal changes in the current direction of reverse flow in the grid; combines load demand and grid capacity forecasts to assess the probability and impact of reverse flow; When the reverse flow decision module detects that a reverse flow is about to occur, the reverse flow detection and response component will immediately issue a warning signal and start the decision-making mechanism; generate a reverse flow warning signal to notify the control and optimization scheduling component of the imminent reverse flow; combine the load demand and grid capacity forecast to evaluate the feasibility of adjusting the photovoltaic output and obtain the reverse flow decision result; The instruction generation module generates specific instructions for adjusting the PV output based on the backflow risk assessment results, and sends the instructions to the control and optimization scheduling component in real time; Backflow detection module, including: The probability assessment submodule uses load demand, grid capacity and reverse flow occurrence variables as nodes to construct a dynamic Bayesian network model. Each node represents a random variable, and the connection between nodes represents the dependency relationship between variables. The parameters of the dynamic Bayesian network model are learned using historical variable data to determine the conditional probability distribution between nodes and calculate the probability distribution of reverse flow occurrence under the current load demand and grid capacity conditions. The impact assessment submodule identifies key nodes and vulnerable links in the power grid, and assesses the impact of reverse flow on the stability of the power grid structure; simulates the operating status of the equipment when reverse flow occurs, assesses the thermal stress and mechanical stress parameters of the equipment, and predicts the life and failure risk of the equipment; constructs a causal relationship map between each device and node in the power grid, and through the causal map, assesses the chain reaction that may be caused by the reverse flow, and predicts the potential fault diffusion path and impact range; The comprehensive assessment submodule dynamically adjusts the weights of probability assessment and impact assessment based on historical assessment results and actual operation feedback; and weightedly integrates the results of probability assessment and impact assessment to generate a backflow risk index.

Images