Multi-energy access electric power spot market risk prevention and control method and system

Abstract

The invention discloses a multi-energy access electric power spot market risk prevention and control method and system, and relates to the technical field of risk prevention and control, and the method comprises the following steps: obtaining multi-dimensional data of a to-be-detected region, obtaining the risk conduction intensity between nodes based on a power grid topological structure, and constructing a dynamic risk topological network model; according to the risk topology network model, node operation characteristics are extracted, evaluation data of the classification model are synchronously obtained, model switching regulation and control are carried out according to the evaluation data, and the model switching regulation and control further comprises the step of outputting a risk overflow probability; through the method based on the dynamic risk topology network and multi-model adaptive switching, the problems of inaccurate risk prediction and untimely response in multi-energy access to the electric power spot market are solved.

Description

Technical Field

[0001] The present invention relates to the field of risk prevention and control technology, and more specifically, to a method and system for risk prevention and control in a multi-energy access electricity spot market. Background Art

[0002] With the transformation of the global energy structure and the advancement of technology, the power system is developing in a diversified and intelligent direction. The traditional power system mainly relies on non-renewable energy, including coal, oil and natural gas, but in recent years, the rapid rise of renewable energy such as solar and wind power has brought new vitality and challenges to the power system. Especially when these renewable energy sources are connected to the power grid in the form of distributed generation, the complexity and uncertainty of the power system increase significantly.

[0003] The power system with multi-energy access has become an inevitable trend of future development. The so-called multi-energy access refers to the power system's ability to simultaneously accept different types of energy, including renewable energy, energy storage equipment, traditional fossil energy, etc., and achieve the complementarity and optimal configuration of these energy sources. This multi-energy access power system can not only improve energy utilization efficiency, but also enhance the flexibility and reliability of the power system. However, multi-energy access also brings about the problem of risk prevention and control in the electricity spot market.

[0004] For example, the invention patent with announcement number: CN114967647A discloses a method and device for assessing the attack risk of a load frequency control system, which belongs to the field of electric power information-physical systems. The method includes: establishing a discrete simulation model of a load frequency control system including multiple regions; starting from any initial moment, continuously superimposing private excitations that satisfy the set distribution in the control instruction input of the load frequency control system; in a set detection window, by obtaining the detection index of the load frequency control system after superimposing the private excitation, updating the confidence of the corresponding sensors in each region of the system; and obtaining the risk assessment results of the load frequency control system according to the confidence ranking results. The present invention can be applied to abnormal sensor data warning and risk assessment of multi-region load frequency control systems, and can quickly locate abnormal sensor data without increasing costs, realize attack identification and positioning, and improve the safety and reliability of the load frequency control system.

[0005] For example, the invention patent with announcement number: CN116360388B discloses a reasoning method and device for a performance-fault relationship map based on a graph neural network, which includes: constructing a performance-fault relationship map of a spacecraft control system based on FMEA; calculating the intimacy of each relationship in the performance-fault relationship map based on historical failure cases; using a graph neural network to infer the fault cause corresponding to each fault symptom based on the intimacy of each relationship in the performance-fault relationship map; for each fault symptom, using a Bayesian network to fuse the fault cause corresponding to the current fault symptom to obtain the final reasoning result of the current fault symptom. The present invention can improve the accuracy of fault reasoning of spacecraft control systems.

[0006] The above disclosed technical solutions have at least the following technical problems: Traditional methods focus on monitoring the overall risk level, but lack an intuitive display of the risk transmission path, resulting in inaccurate risk prediction and untimely response. In view of the above problems, the present invention proposes a solution. Summary of the invention

[0007] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a risk prevention and control method and system for a multi-energy access electricity spot market. Through a method based on a dynamic risk topology network and multi-model adaptive switching, it solves the problems of inaccurate risk prediction and untimely response in a multi-energy access electricity spot market, and achieves an innovative breakthrough in real-time dynamic regulation and multi-level risk visualization.

[0008] To achieve the above object, the present invention provides the following technical solutions: A method and system for risk prevention and control in a multi-energy access electricity spot market comprises the following steps: obtaining multidimensional data of a test area, obtaining the risk transmission intensity between nodes based on a power grid topology, and constructing a dynamic risk topology network model; extracting node operation characteristics according to the risk topology network model, synchronously obtaining evaluation data of a classification model, and performing model switching control according to the evaluation data, wherein the model switching control also includes outputting a risk spillover probability.

[0009] In a preferred embodiment, the multidimensional data of the area to be tested is obtained, the risk conduction intensity between nodes is obtained based on the power grid topology structure, and a dynamic risk topology network model is constructed, specifically as follows: the wind farm, photovoltaic power station, load center, and energy storage station nodes are regarded as network nodes, and the risk conduction intensity between nodes is output through Granger causality analysis and impulse response function; a dynamic risk topology network is constructed according to the risk conduction intensity between nodes.

[0010] In a preferred embodiment, the dynamic risk topology network is constructed according to the risk conduction strength between nodes, specifically as follows: taking the power grid topology map as the basis, and taking the risk conduction strength as the weight of the edge, a risk-weighted network is constructed; obtaining the first data of the risk-weighted network, analyzing the key nodes and key lines according to the first data, and identifying the high-risk conduction paths; simulating the risk propagation process in the power grid through a dynamic propagation model, analyzing the risk diffusion behavior under different control strategies, and constructing a dynamic risk topology network.

[0011] In a preferred embodiment, the key nodes and key lines are analyzed according to the first data to identify high-risk transmission paths, specifically as follows: based on the first data, the second data is output, and the node risk propagation paths are screened according to the risk transmission intensity and the second data to obtain key nodes and key paths; a risk propagation chain is constructed according to the key nodes and key paths, and high-risk transmission paths are identified according to the risk propagation chain.

[0012] In a preferred embodiment, the node operation characteristics are extracted according to the risk topology network model, and the evaluation data of the classification model is obtained synchronously. The model switching regulation is performed according to the evaluation data, and the model switching regulation also includes outputting the risk spillover probability, which is as follows: extracting the node operation characteristics, and the node operation characteristics include a first data set, a second data set, and a third data set; performing reliability evaluation on the classification model through a sliding time window to obtain reliability evaluation data of the classification model, and the reliability evaluation data include prediction error and confidence; according to the reliability evaluation data of the classification model, switching the classification model, and outputting the risk spillover probability according to the node operation characteristics corresponding to the data, and the classification model includes LSTM, Bayesian network, and Markov chain; generating a risk heat map according to the risk spillover probability to carry out risk prevention and control.

[0013] In a preferred embodiment, the classification model is switched according to the reliability evaluation data of the classification model, as follows: According to the reliability evaluation data of the classification model, the classification model is switched as follows: When the prediction error is small and the confidence is high, switch to the LSTM model; When the prediction error is high and the confidence is high, switch to the Bayesian network to quickly locate abnormal nodes and fault causes; When the confidence is high and the state transition rate is higher than the threshold, switch to the Markov chain.

[0014] In a preferred embodiment, the confidence includes a Markov chain confidence, and the steps for obtaining the Markov chain confidence are as follows: obtaining historical state change data of the node to be tested, fine-grained division of the node state through a clustering algorithm, obtaining the number of node transitions between different states, and constructing an initial state transfer matrix; predicting the state at the current moment based on the state at the previous moment through a Markov chain model, and comparing the actual state with the predicted state, and outputting the state residual; based on the state transition probability, state residual and historical prediction accuracy in the initial state transfer matrix, outputting the confidence at the current moment.

[0015] In a preferred embodiment, the steps for obtaining the confidence penalty factor of the state transfer matrix are as follows: obtaining historical state residual data of the Markov chain model; selecting several groups of historical windows of different lengths based on the historical state residual data, obtaining the average value of all state residuals in the historical windows, and constructing a state residual data set; predicting the state residual through the Markov chain model, and dynamically adjusting the confidence penalty factor by selecting the optimal window length based on the state residual data set.

[0016] In a preferred embodiment, the risk heat map is generated according to the risk spillover probability, and risk prevention and control is performed as follows: scenario data of each node is obtained, the nodes are divided into scenarios, dynamic model switching is performed according to the scenario division, and the risk spillover probability is predicted; a probability matrix is ​​constructed according to the risk spillover probability prediction data; the probability matrix is ​​mapped to a power grid geographic map, and discrete node probabilities are spatially interpolated to generate a risk heat map; and risk prevention and control is performed according to the risk heat map.

[0017] A system for a risk prevention and control method of a multi-energy access electricity spot market includes a risk topology network module and a risk spillover probability output module, and there is a connection between the modules; the risk topology network module is used to obtain multidimensional data of a test area, obtain the risk conduction intensity between nodes based on the power grid topology structure, and build a dynamic risk topology network model; the risk spillover probability output module is used to extract node operation characteristics according to the risk topology network model, synchronously obtain evaluation data of a classification model, and perform model switching regulation according to the evaluation data, wherein the model switching regulation includes outputting the risk spillover probability.

[0018] Technical effects and advantages of a multi-energy access power spot market risk prevention and control method and system of the present invention: 1. The present invention constructs a dynamic risk topology network model based on the risk transmission strength of the power grid topology structure, which can accurately simulate the risk propagation path between different nodes in the power grid. Through Granger causality analysis and impulse response function, the risk transmission relationship between nodes can be quantitatively analyzed, providing accurate basic data for subsequent risk prediction.

[0019] 2. The present invention can extract the operating characteristics of the power grid nodes by analyzing the risk topology network model, and then evaluate the reliability of the classification model through a sliding time window to obtain the prediction error and confidence. The node operation characteristics include indicators such as voltage offset, load rate, and short-term fluctuation amplitude. These characteristics can help determine the stability or abnormality of the current state of the power grid, and on this basis, divide the nodes into different risk scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 The present invention is a flowchart of a method for risk prevention and control in a spot electricity market with multi-energy access.

[0021] Figure 2 This is a schematic diagram of the system structure of a method for risk prevention and control in a power spot market with multi-energy access according to the present invention.

[0022] Figure 3 This is the risk spillover probability curve. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] Embodiment 1, Figure 1 The present invention provides a method for risk prevention and control of the electric power spot market with multi-energy access, and the specific steps are as follows: S1, obtain multi-dimensional data of the area to be tested, obtain the risk transmission intensity between nodes based on the power grid topology structure, and build a dynamic risk topology network model.

[0025] In this embodiment, multi-dimensional data of the area to be tested is obtained, the risk transmission strength between nodes is obtained based on the power grid topology structure, and a dynamic risk topology network model is constructed according to the risk transmission strength, as follows: Through smart meters and energy storage management system equipment, we can obtain data such as power generation output, load curve, voltage frequency, energy storage charging and discharging status, and electricity price. We also use the KNN interpolation method to repair missing data and the 3σ principle to remove outliers. Use the Z-score standardization method to uniformly process data of different dimensions, and use a unified timestamp for the processed data to ensure the timeliness and consistency of the data; The wind farm, photovoltaic power station, load center, and energy storage station nodes are regarded as network nodes, and the risk transmission intensity between nodes is output through Granger causality analysis and impulse response function; A risk topology network is constructed based on the risk conduction intensity between nodes. The edge weights between nodes change dynamically, reflecting the risk conduction relationship in real time.

[0026] In this embodiment, a dynamic risk topology network is constructed according to the risk transmission strength between nodes, as follows: Taking the power grid topology as the basis and the risk transmission intensity as the edge weight, a risk-weighted network is constructed; Acquire first data of the risk-weighted network, wherein the first data includes a clustering coefficient and an average path length of the network; Analyze key nodes and key lines based on the first data to identify high-risk transmission paths; The dynamic propagation model is used to simulate the risk propagation process in the power grid, analyze the risk diffusion behavior under different control strategies, and construct a dynamic risk topology network.

[0027] The calculation formula of the risk dynamic propagation model is as follows:

[0028] Where: is the risk state of node i at the next time t+1, is the risk value of node i at time t, is the risk value of node j at time t, is the set of adjacent nodes of node i, representing all nodes directly connected to node i. is the risk transmission intensity from node j to node i, is the sum of risks received by node i from its neighbor node j.

[0029] It should be noted that the commonly used represents the normalized risk value, 0 represents no risk, 1 represents complete failure, Based on the current risk status of the node And the risk transmission amount of adjacent nodes to it Decide, Depending on the topology of the grid, represents the ability of risk to spread from node j to node i, It reflects the process of fault or risk gradually spreading through the topological structure in the power grid. If the risk value of node j is high and is also high, the risk increase of node i will be more significant.

[0030] In this embodiment, the key nodes and key lines are analyzed according to the first data to identify high-risk transmission paths, as follows: Outputting second data according to the first data, wherein the second data includes betweenness centrality and degree centrality; According to the risk transmission intensity and the second data, the node risk propagation path is screened to obtain the key nodes and key paths; Construct a risk transmission chain based on key nodes and key paths, and identify high-risk transmission paths based on the risk transmission chain.

[0031] The calculation formula of betweenness centrality is as follows:

[0032] The calculation formula of the clustering coefficient is as follows:

[0033] The calculation formula of degree centrality is as follows:

[0034] Where: is the betweenness centrality, is the number of shortest paths passing through node i, is the number of shortest paths from node s to node t, is the degree centrality, is the adjacency matrix, is the degree of node i, is the clustering coefficient, is the actual number of links between adjacent nodes of node i.

[0035] It should be noted that nodes with high betweenness centrality are usually "hub" nodes in the power grid. They connect multiple subnets. Once the risk spreads at these nodes, it will aggravate the risk diffusion of the entire power grid. Nodes with high degree centrality are connected to more power grid equipment, which directly affects the risk propagation of multiple nodes. Once the risk of these nodes increases, it may affect a larger range of power grid areas. The clustering coefficient reflects the risk propagation characteristics within a local area. If the clustering coefficient of a certain area is high, the risk may spread rapidly in the local area without affecting remote areas.

[0036] S2, extracting node operation characteristics according to the risk topology network model, synchronously acquiring evaluation data of the classification model, and performing model switching regulation according to the evaluation data, wherein the model switching regulation also includes outputting the risk spillover probability.

[0037] In this embodiment, the node operation characteristics are extracted according to the risk topology network model, and the evaluation data of the classification model is obtained synchronously. The model switching regulation is performed according to the evaluation data. The model switching regulation also includes outputting the risk spillover probability, which is as follows: Extracting node operation features, wherein the extracted node operation features include a first data set, a second data set, and a third data set; The first data set includes voltage offset, load factor, and short-term fluctuation amplitude, which are used to input into LSTM (Long Short-Term Memory Network); The second data set includes neighborhood aggregation features, risk transmission strength, and probability distribution of the current state of the node, which is used to input the Bayesian network; The third data set includes a state transition rate, a state transition probability matrix, and a trend change rate, and is used to input into a Markov chain; Performing reliability evaluation on the classification model through a sliding time window to obtain reliability evaluation data of the classification model, wherein the reliability evaluation data includes prediction error and confidence level; According to the reliability assessment data of the classification model, the classification model is switched, and the node operation characteristics corresponding to the data are used to output the risk spillover probability. The classification model includes LSTM (Long Short-Term Memory Network), Bayesian Network, and Markov Chain; Based on the risk spillover probability, a risk heat map is generated to carry out risk prevention and control.

[0038] The calculation formula of the prediction error is as follows:

[0039] The calculation formula of the Bayesian network confidence is as follows:

[0040] Where: is the prediction error, is the number of prediction moments in the current window, is the true value at the ith moment, is the model prediction value at the i-th moment, is the confidence of the Bayesian network, is the system status, is the observed input data, is the posterior probability of the state.

[0041] In this embodiment, the classification model is switched according to the reliability evaluation data of the classification model, as follows: When the prediction error is small and the confidence is high, switch to the LSTM model; When the prediction error is high and the confidence is high, the network switches to the Bayesian network to quickly locate abnormal nodes and fault causes; When the confidence is high and the state transition rate is higher than the threshold, it switches to the Markov chain.

[0042] Further, the reasons for selecting models based on scene division are explained: Select the LSTM model: LSTM performs well in processing long-term sequence data and can effectively capture the time series characteristics of node operation status. In a stable scenario, the changes in parameters such as voltage and load usually have strong regularity. LSTM is good at extracting trends and periodic features from historical data. Since the system state is stable and there is less noise, the LSTM model can provide lower prediction errors and higher confidence levels, and can achieve long-term window predictions and identify potential risk spillover tendencies in advance. Select Bayesian network: Bayesian network uses a probabilistic graph model to represent the causal relationship between nodes. When an anomaly is detected, it can quickly trace the cause of the anomaly. Using Bayesian reasoning, it can quantify the probability of abnormal events when the data is incomplete or noisy. It has the ability to quickly respond to sudden abnormal events such as load surges and equipment failures, and can provide risk spillover probability and abnormal development trends. A high confidence level indicates that the anomaly detection result is reliable and the cause of the anomaly can be further inferred. Select the Markov chain model: The Markov chain model is based on the state transfer matrix and can effectively capture the transition rules of nodes between different states. In the state transition scenario, the operating state of the node may change rapidly. The Markov chain is good at predicting the state evolution in a short period of time. In the scenario where the node state switches frequently, the model has a small amount of calculation and can quickly update the state transfer matrix to achieve real-time prediction. The Markov chain is good at modeling the state transfer process and is suitable for analyzing the trend of node state changes. If the confidence is high and the state transfer rate is fast, the power grid state may be in a transition stage.

[0043] In this embodiment, the steps for obtaining the LSTM confidence are as follows: Obtain the historical time series data of the node to be tested, detect and mark abnormal data, use anomaly detection algorithms based on IQR (interquartile range) or LOF (local outlier factor), remove outliers or repair data with interpolation methods, and remove noise through smoothing filtering or wavelet transform to ensure the quality of input data; The time series features of historical time series data are captured by combining LSTM and multi-scale convolutional layers, and the node states at different time granularities are predicted using a multi-step prediction method to further enhance the model's time dependency capture capability, obtain the residual between the predicted value and the actual value, and construct a residual sequence. Decompose the residual into trend term, seasonal term and random term through decomposition technology (such as STL decomposition), analyze the errors from different sources respectively, and judge whether the abnormal residual is caused by data anomaly by combining expert rules or Bayesian reasoning to reduce the misjudgment rate of errors; Perform statistical analysis on the residual sequence through the sliding window method and output the mean and standard deviation of the residual; Output the confidence level of the current prediction based on the mean and standard deviation of the residuals and the normal distribution assumption; According to the confidence change trend of historical predictions, abnormal jumps in confidence are identified. If an abnormality occurs, the confidence correction mechanism is triggered to re-evaluate the model performance; If the residuals are close to the mean and within the threshold range, the confidence level is high.

[0044] The calculation formula of LSTM confidence is as follows:

[0045] Where: is the confidence of LSTM, ranging from 0 to 1, is the residual at the current moment, that is, the difference between the true value and the predicted value, is the residual mean within the sliding window, which is used to measure the normal error level of the system. It is the residual standard deviation within the sliding window, reflecting the fluctuation range of the error.

[0046] It should be noted that 0 and 1 in the calculation formula of LSTM confidence are the lower and upper limits of confidence respectively. Greater than When , it means that the current prediction error has seriously deviated from the historical error distribution, and the confidence is judged to be 0; when the prediction error When it is close to 0, that is, the current predicted value is very close to the actual value, the confidence is judged to be 1; if A value less than 0 means that the prediction error is too large and the confidence level is 0.

[0047] In this embodiment, the steps for obtaining the Markov chain confidence are as follows: Obtain the historical state change data of the node to be tested, perform fine-grained division of the node state through clustering algorithm, divide the continuous state into a finer discrete state space, count the number of node transitions between different states, and construct the initial state transition matrix; The Markov chain model is used to predict the current state based on the state at the previous moment, and the actual state is compared with the predicted state, and the state residual is output. The abnormal state is identified using anomaly detection algorithms (such as Kalman filtering or Bayesian anomaly detection) to prevent the abnormal state from misleading the confidence calculation; Based on the state transition probability, state residual and historical prediction accuracy in the initial state transfer matrix, output the confidence level at the current moment; If the state prediction is correct and the transition probability is high, the confidence level is high; If the state residual is large or the prediction is inaccurate, the confidence level will be reduced.

[0048] The calculation formula of Markov chain confidence is as follows:

[0049]

[0050] Where: is the Markov chain confidence, is the state transition probability, that is, the probability of transitioning from the state at the previous moment to the state at the current moment, is the state residual, which is 1 if the state prediction is wrong, otherwise it is 0. is the confidence penalty factor of the state transfer matrix, is the state residual at the i-th moment, is the history window length.

[0051] In this embodiment, the steps for obtaining the confidence penalty factor of the state transfer matrix are specifically as follows: Obtain the historical state residual data of the Markov chain model; According to the historical state residual data, several groups of historical windows of different lengths are selected, and the average value of all state residuals in the historical window is obtained to indicate the accuracy of the historical state prediction, and the state residual data set is constructed; If most of the predictions within the window are correct, the average value is close to 0, indicating good prediction performance; If most of the predictions within the window are wrong, the average value is close to 1, indicating poor prediction performance; The state residual is predicted through the Markov chain model, and the confidence penalty factor is dynamically adjusted by selecting the optimal window length according to the state residual data set.

[0052] S3, based on the risk spillover probability, generates a risk heat map and conducts risk prevention and control.

[0053] In this embodiment, a risk heat map is generated according to the risk spillover probability to carry out risk prevention and control, as follows: Obtain scenario data of each node, divide the nodes into scenarios, perform dynamic model switching based on scenario divisions, and predict the probability of risk spillover; Construct a probability matrix based on the risk spillover probability prediction data; The probability matrix is ​​mapped to the power grid geographic map, and the kernel density estimation (KDE) algorithm is used to spatially interpolate the discrete node probabilities to generate a risk heat map; Carry out risk prevention and control based on the risk heat map.

[0054] Embodiment 2, Figure 2The present invention provides a system for a multi-energy access power spot market risk prevention and control method, including a risk topology network module and a risk spillover probability output module, and there is a connection between the modules; The risk topology network module is used to obtain multi-dimensional data of the area to be tested, obtain the risk transmission intensity between nodes based on the power grid topology structure, and build a dynamic risk topology network model; The risk spillover probability output module is used to extract node operation characteristics according to the risk topology network model, synchronously obtain the evaluation data of the classification model, and perform model switching regulation according to the evaluation data, wherein the model switching regulation includes outputting the risk spillover probability.

[0055] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0056] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0057] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0058] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0059] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0060] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for risk prevention and control of a multi-energy access electricity spot market, characterized in that: The steps include: Obtain multi-dimensional data of the area to be tested, obtain the risk transmission intensity between nodes based on the power grid topology, and build a dynamic risk topology network model; Node operation characteristics are extracted according to the risk topology network model, and evaluation data of the classification model is obtained synchronously. Model switching regulation is performed according to the evaluation data, and the model switching regulation also includes outputting the risk spillover probability.

2. The method for risk prevention and control of the electric power spot market with multi-energy access according to claim 1 is characterized in that: The multi-dimensional data of the area to be tested is obtained, the risk transmission intensity between nodes is obtained based on the power grid topology structure, and a dynamic risk topology network model is constructed, as follows: The wind farm, photovoltaic power station, load center, and energy storage station nodes are regarded as network nodes, and the risk transmission intensity between nodes is output through Granger causality analysis and impulse response function; Construct a dynamic risk topology network based on the risk transmission intensity between nodes.

3. The method for risk prevention and control of the electric power spot market with multi-energy access according to claim 2 is characterized in that: The dynamic risk topology network is constructed according to the risk transmission intensity between nodes, as follows: Taking the power grid topology as the basis and the risk transmission intensity as the edge weight, a risk-weighted network is constructed; Obtaining first data of the risk-weighted network, analyzing key nodes and key lines based on the first data, and identifying high-risk transmission paths; The dynamic propagation model is used to simulate the risk propagation process in the power grid, analyze the risk diffusion behavior under different control strategies, and construct a dynamic risk topology network.

4. The method for risk prevention and control of the electric power spot market with multi-energy access according to claim 3 is characterized in that: The analysis of key nodes and key lines based on the first data to identify high-risk transmission paths is as follows: According to the first data, the second data is output, and according to the risk transmission intensity and the second data, the node risk propagation path is screened to obtain the key nodes and key paths; Construct a risk transmission chain based on key nodes and key paths, and identify high-risk transmission paths based on the risk transmission chain.

5. The method for risk prevention and control of the electric power spot market with multi-energy access according to claim 4 is characterized in that: The node operation characteristics are extracted according to the risk topology network model, and the evaluation data of the classification model is obtained synchronously. The model switching regulation is performed according to the evaluation data. The model switching regulation also includes outputting the risk spillover probability, which is specifically as follows: Extracting node operation features, wherein the node operation features include a first data set, a second data set, and a third data set; Performing reliability evaluation on the classification model through a sliding time window to obtain reliability evaluation data of the classification model, wherein the reliability evaluation data includes prediction error and confidence level; According to the reliability assessment data of the classification model, the classification model is switched, and the node operation characteristics corresponding to the data are output to output the risk spillover probability. The classification model includes LSTM, Bayesian network, and Markov chain; Based on the risk spillover probability, a risk heat map is generated to carry out risk prevention and control.

6. The method for risk prevention and control of the electric power spot market with multi-energy access according to claim 5 is characterized in that: The classification model is switched according to the reliability evaluation data of the classification model, as follows: When the prediction error is small and the confidence is high, switch to the LSTM model; When the prediction error is high and the confidence is high, the network switches to the Bayesian network to quickly locate abnormal nodes and fault causes; When the confidence is high and the state transition rate is higher than the threshold, it switches to the Markov chain.

7. The method for risk prevention and control of the electric power spot market with multi-energy access according to claim 6 is characterized in that: The confidence includes the Markov chain confidence, and the steps for obtaining the Markov chain confidence are as follows: Obtain the historical state change data of the node to be tested, perform fine-grained division of the node state through clustering algorithm, obtain the number of node transitions between different states, and construct the initial state transition matrix; The Markov chain model is used to predict the current state based on the state at the previous moment, and the actual state is compared with the predicted state, and the state residual is output; Based on the state transition probability, state residual and historical prediction accuracy in the initial state transfer matrix, the confidence level at the current moment is output.

8. The method for risk prevention and control of the electric power spot market with multi-energy access according to claim 7 is characterized in that: The steps for obtaining the confidence penalty factor of the state transfer matrix are specifically as follows: Obtain the historical state residual data of the Markov chain model; According to the historical state residual data, several groups of historical windows of different lengths are selected, the average value of all state residuals in the historical window is obtained, and the state residual data set is constructed; The state residual is predicted through the Markov chain model, and the confidence penalty factor is dynamically adjusted by selecting the optimal window length according to the state residual data set.

9. The method for risk prevention and control of the electric power spot market with multi-energy access according to claim 8 is characterized in that: According to the risk spillover probability, a risk heat map is generated to carry out risk prevention and control, as follows: Obtain scenario data of each node, divide the nodes into scenarios, perform dynamic model switching based on scenario divisions, and predict the probability of risk spillover; Construct a probability matrix based on the risk spillover probability prediction data; Map the probability matrix to the grid geographic map and perform spatial interpolation on discrete node probabilities to generate a risk heat map; Carry out risk prevention and control based on the risk heat map.

10. A system using the multi-energy access power spot market risk prevention and control method as described in any one of claims 1 to 9, characterized in that: It includes risk topology network module and risk spillover probability output module, and there are connections between modules; The risk topology network module is used to obtain multi-dimensional data of the area to be tested, obtain the risk transmission intensity between nodes based on the power grid topology structure, and build a dynamic risk topology network model; The risk spillover probability output module is used to extract node operation characteristics according to the risk topology network model, synchronously obtain the evaluation data of the classification model, and perform model switching regulation according to the evaluation data, wherein the model switching regulation includes outputting the risk spillover probability.

Images