Spot market risk prevention and control method and system based on real-time data analysis

Abstract

The invention discloses a power grid planning theme dynamic iteration method based on a fuzzy evolutionary algorithm. Comprising. According to the method, regional spot market operation data is acquired in real time; establishing a regional spot market operation risk evaluation index system, and constructing a regional spot market operation risk comprehensive evaluation model based on an entropy weight TOPSIS method; solving the regional spot market operation risk comprehensive evaluation model, and performing data analysis and processing; judging a current risk level according to a data analysis and processing result; early warning and response mechanisms of different levels are formulated. According to the invention, targeted real-time monitoring, evaluation, prevention and control are carried out on the operation risk of the regional spot market, the market response speed is improved, the operation risk is reduced, the market stability is enhanced, and the stable and efficient operation of the market is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of power market risk prevention and control, and in particular to a spot market risk prevention and control method and system based on real-time data analysis. Background Art

[0002] With the large-scale access of new energy, complex changes in market supply and demand, and the influence of factors such as extreme weather, the risks in the electricity spot market have increased significantly. These risks not only affect the stable operation of the market, but may also pose a threat to the security of electricity supply.

[0003] Risk prevention and control research helps to improve the stability and predictability of the power market and reduce the uncertainty and negative impact caused by price fluctuations. Through effective risk prevention and control, fair competition in the power market can be guaranteed, the healthy development of power generation enterprises and power users can be promoted, decision-making support can be provided to the government and relevant regulatory agencies, and the improvement and optimization of the power market mechanism can be promoted. Real-time monitoring of spot market operation risks has become the key to ensuring the stable operation of the market. On the one hand, it can help market participants identify, evaluate and manage potential risks, and formulate reasonable trading strategies and risk management plans. On the other hand, risk prevention and control can also promote the transparency of market information and improve the decision-making efficiency and accuracy of market participants. In addition, by building a risk early warning system and emergency response mechanism, risk prevention and control can also respond quickly when risk events occur, reduce losses and restore the normal operation of the market.

[0004] However, the trial operation of regional and provincial spot markets may produce an unbalanced effect, which will have a great impact on market operations and all parties and be difficult to handle. In addition, existing risk monitoring methods may have data processing delays, inaccurate or lack of real-time identification of regional and provincial risks, and other problems.

[0005] In summary, targeted real-time monitoring and accurate assessment of regional spot market operational risks are needed to improve market response speed, reduce operational risks, enhance market stability, and provide scientific decision-making support for market operators. Summary of the invention

[0006] In view of the problems existing in the prior art, the present invention is proposed.

[0007] Therefore, the problem to be solved by the present invention is how to solve the problem that the unbalanced effects may be produced in the trial operation of the regional and provincial spot markets, which has a great impact on the market operation and all parties and is difficult to handle. In addition, the existing risk monitoring methods may have data processing delays, inaccurate or lack of real-time identification of regional and provincial risks, and other problems.

[0008] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0009] In a first aspect, an embodiment of the present invention provides a spot market risk prevention and control method based on real-time data analysis, which includes obtaining regional power spot market operation data in real time;

[0010] Establish a regional spot market operation risk evaluation index system and construct a regional spot market operation risk comprehensive evaluation model based on the entropy weight TOPSIS method;

[0011] Solve the comprehensive evaluation model of regional spot market operation risks and conduct data analysis and processing;

[0012] Determine the current risk level based on the results of data analysis and processing;

[0013] Develop early warning and response mechanisms at different levels.

[0014] As a preferred solution of the spot market risk prevention and control method based on real-time data analysis described in the present invention, the real-time acquisition of regional power spot market operation data includes historical power load and real-time power load; power generation capacity; spare capacity; clearing price; long-term average price; price limit; number of power outages and power outage duration; number of failures; number of contracts; number of contract breaches; number of contract disputes;

[0015] The power generation capacity data represents the installed capacity and actual output information of various types of power generation units;

[0016] The reserve capacity data represents the reserve capacity available in the system, including both spinning reserve and non-spinning reserve capacity;

[0017] Long-term average price data represents the long-term average or expected price of historical electricity market prices and is used to calculate price deviation;

[0018] Price limits data characterizes upper and lower price thresholds set by the market and is used to identify extreme price events.

[0019] As a preferred solution of the spot market risk prevention and control method based on real-time data analysis described in the present invention, the establishment of a regional spot market operation risk evaluation index system includes four primary indicators, namely, power supply and demand risk, market price abnormality risk, power system safe operation risk, and contract performance risk;

[0020] The secondary indicators corresponding to the first-level indicator of power supply and demand risk are supply and demand balance deviation rate, peak-valley load difference rate, and spare capacity adequacy rate;

[0021] The secondary indicators corresponding to the first-level indicator of abnormal market price risk are price volatility, price deviation, and extreme price frequency;

[0022] The first-level indicator of power system safe operation risk corresponds to the second-level indicators of system reliability index (SAIFI, SAIDI), power grid accident rate, and equipment failure rate;

[0023] The first-level indicator of performance risk corresponds to the second-level indicators of contract default rate, credit loss rate, and dispute resolution time.

[0024] As a preferred solution of the spot market risk prevention and control method based on real-time data analysis described in the present invention, the construction of a comprehensive evaluation model for regional spot market operation risks based on the entropy weight TOPSIS method includes the following steps:

[0025] Design and distribute questionnaires;

[0026] The entropy weight method is used to obtain the weights of the risk probability and risk loss consequences;

[0027] Substitute the weight results into the TOPSIS model;

[0028] Get the comprehensive evaluation index and sort the indicators from large to small.

[0029] As a preferred solution of the spot market risk prevention and control method based on real-time data analysis described in the present invention, wherein: the questionnaire adopts a 7-level scale scoring method, and the risk occurrence probability P1 is divided from low to high into "very low (1 point); low (2 points); relatively low (3 points); average (4 points); relatively high (5 points); high (6 points); very high (7 points)"; the risk loss consequence P2 is divided into very low loss (1 point); low loss (2 points); relatively low loss (3 points); average loss (4 points); large loss (5 points); large loss (6 points); very large loss (7 points);

[0030] Then, the average values ​​of the scoring results collected from the questionnaire are calculated according to the two dimensions of risk probability P1 and risk loss consequence P2, and then the entropy weight method is used to calculate the weights of risk probability and risk loss consequence.

[0031] As a preferred solution of the spot market risk prevention and control method based on real-time data analysis described in the present invention, the step of using the entropy weight method to obtain the weight of the risk probability and the risk loss consequence includes three steps: constructing an initialization decision matrix, standardizing indicators, and calculating the entropy value and weight of the risk probability and the risk loss consequence. The three steps are specifically as follows:

[0032] The construction of the initialization decision matrix can be expressed as follows:

[0033]

[0034] The initialization decision matrix is ​​constructed using the average value, where n represents the number of risk factor indicators being evaluated, m represents the risk importance of each indicator, and B i (i=1,2,……,n) represents the nth risk factor index B, C evaluated for the event j (j=1,2,……,m) represents the risk importance of the jth indicator of the event, x ij It indicates the importance of the i-th indicator, and all indicators are represented as x ij The form of is the decision matrix;

[0035] The questionnaire can be randomly distributed to multiple experts, enterprises, and governments to score the probability and the degree of loss. A 7-level scale model is used. The lower the score, the smaller the probability of the risk or the smaller the degree of loss. When the credibility requirement is met, the questionnaire can be used, and the entropy weight method can be used to obtain the weights of the risk probability and the risk loss consequences.

[0036] The index standardization process can be expressed as follows:

[0037]

[0038] Among them, x ij + The normalized formula for positive indices, x ij - The normalized formula for negative indices, x ij represents the importance of the i-th indicator, represents the minimum risk importance of the jth indicator, Indicates the highest risk importance of the jth indicator;

[0039] The selection of the positive and negative index normalization formula is based on the positive and negative relationship between the result and the index;

[0040] If the change of the result indicator is positively correlated, the normalization formula of the positive indicator is selected; if the change of the result indicator is negatively correlated, the normalization formula of the negative indicator is selected;

[0041] If negative values ​​appear after substituting numerical values ​​for index standardization, further data translation processing is required to eliminate negative values: ij '=h+x ij + or x ij '=h+x ij - , where h represents the amplitude of the index translation. Generally, h = 1. After dimensionless processing, the following expression is used:

[0042]

[0043] Among them, y ij represents the value after dimensionless processing, x ij ' represents the value after translation processing;

[0044] The entropy value and weight for calculating the risk occurrence probability and risk loss consequence can be expressed as follows:

[0045]

[0046] Among them, ω i represents the weight of the i-th indicator, G i It represents the coefficient of difference of the ith indicator, expressed as G i =1-E i ; E i Represents the entropy value of the i-th indicator, expressed as

[0047] Substituting the weight results into the TOPSIS model includes three steps: calculating the weighted normalization matrix, calculating the positive ideal solution and the negative ideal solution, and calculating the Euclidean distance;

[0048] The calculation weighted normalization matrix can be expressed as follows:

[0049] d ij =ω i f ij

[0050] Among them, d ij represents the weighted normalization matrix, ω i represents the weight of the i-th indicator, f ij Represents the data after vector normalization, expressed as or

[0051] The calculation of the positive ideal solution and the negative ideal solution may be performed using the following expressions:

[0052]

[0053] in, and denote the positive ideal solution and the negative ideal solution respectively, d ij represents the weighted normalization matrix;

[0054] The calculation of the Euclidean distance may be performed using the following expression:

[0055]

[0056] in, and Respectively represent the distance of each scheme i from the ideal point and the negative ideal point;

[0057] The obtaining of the comprehensive evaluation index and sorting the indicators from large to small includes two steps: calculating the comprehensive evaluation index and sorting the evaluation index, which are specifically expressed as follows:

[0058] The calculation of the comprehensive evaluation index can be performed using the following expression:

[0059]

[0060] Among them, L i Represents the comprehensive evaluation index value, and Respectively represent the distance of each scheme i from the ideal point and the negative ideal point;

[0061] The ranking of the evaluation indexes includes sorting the evaluation indexes according to the comprehensive evaluation index L i Sort the indicators from large to small and filter out key risk indicators.

[0062] As a preferred solution of the spot market risk prevention and control method based on real-time data analysis described in the present invention, the solving of the regional spot market operation risk comprehensive evaluation model, data analysis and processing include the following steps:

[0063] Cluster analysis was performed using the K-means clustering method, and the selected key indicators were further divided into three categories using the SPSS tool, ultimately obtaining three cluster centers of risk values;

[0064] The following method is used to solve the comprehensive evaluation model of regional spot market operation risk to obtain the level of risk events:

[0065] Solving the model: inputting data and indicators through spass to obtain results;

[0066] Cluster analysis: Cluster analysis classifies data based on its own information, and divides data into several categories based on the distance, so that the similarity between data belonging to the same category is as large as possible, and the similarity between different types is as small as possible, so as to achieve classification;

[0067] Based on the data and data comparison distribution, the key risk indicators are clustered using SPSS tools;

[0068] Risk level classification: Based on the clustering results, risk events are classified into different risk levels. Each cluster represents a risk level, and events with similar risk characteristics are classified into the same level.

[0069] Determining the risk level based on the results of data analysis and processing includes the following steps:

[0070] Based on the above clustering results, the risk warning can be divided into three levels according to the risk value according to the three cluster centers: low risk, medium risk and high risk, and a radar chart of the indicator differences at different risk levels can be drawn and displayed in the decision-making module;

[0071] The formulation of different levels of early warning and response mechanisms includes the following steps:

[0072] Corresponding warnings are issued according to different levels of risks, and alarms are issued when specific thresholds are reached; specific response measures are then formulated based on the warnings, including preventive measures, mitigation measures and emergency measures, so that quick action can be taken when risks occur.

[0073] In a second aspect, an embodiment of the present invention provides a spot market risk prevention and control system based on real-time data analysis, which includes an acquisition module for acquiring regional power spot market operation data in real time;

[0074] Construct modules, establish a regional spot market operation risk evaluation index system, and construct a regional spot market operation risk comprehensive evaluation model based on the entropy weight TOPSIS method;

[0075] The solution module solves the comprehensive evaluation model of regional spot market operation risks and performs data analysis and processing;

[0076] The judgment module determines the current risk level based on the results of data analysis and processing;

[0077] Early warning module, formulate early warning and response mechanisms at different levels.

[0078] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the spot market risk prevention and control method based on real-time data analysis as described in the first aspect of the present invention are implemented.

[0079] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the spot market risk prevention and control method based on real-time data analysis as described in the first aspect of the present invention are implemented.

[0080] The beneficial effects of the present invention are as follows: the present invention provides a spot market risk prevention and control method and system based on real-time data analysis, which obtains spot market operation data in real time; establishes a regional spot market operation risk evaluation index system, and constructs a regional spot market operation risk comprehensive evaluation model based on the entropy weight TOPSIS method; solves the regional spot market operation risk comprehensive evaluation model, and performs data analysis and processing; judges the current risk level according to the results of data analysis and processing; and formulates early warning and response mechanisms of different levels. The present invention conducts targeted real-time monitoring, evaluation, and prevention and control of regional spot market operation risks, improves market response speed, reduces operational risks, enhances market stability, and ensures stable and efficient operation of the market. The regional electricity spot market risk prevention and control device, electronic device, and computer-readable storage medium based on real-time data analysis provided by the present invention also solve the problems raised in the background technology section. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0082] Figure 1 It is a flowchart of the spot market risk prevention and control method based on real-time data analysis;

[0083] Figure 2 A diagram of a computer device for a spot market risk prevention and control method based on real-time data analysis;

[0084] Figure 3 The framework diagram of the regional spot market operation risk evaluation index system based on the spot market risk prevention and control method based on real-time data analysis. DETAILED DESCRIPTION

[0085] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0086] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0087] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0088] Example 1

[0089] Reference Figure 1-2 , which is the first embodiment of the present invention, and provides a spot market risk prevention and control method based on real-time data analysis, comprising:

[0090] S100: Real-time acquisition of regional power spot market operation data;

[0091] S101: The real-time acquisition of regional power spot market operation data includes historical power load and real-time power load; power generation capacity; spare capacity; clearing price; long-term average price; price limit; number of power outages and power outage duration; number of failures; number of contracts; number of contract breaches; number of contract disputes;

[0092] The power generation capacity data represents the installed capacity and actual output information of various types of power generation units;

[0093] The reserve capacity data represents the reserve capacity available in the system, including both spinning reserve and non-spinning reserve capacity;

[0094] Long-term average price data represents the long-term average or expected price of historical electricity market prices and is used to calculate price deviation;

[0095] Price limits data characterizes upper and lower price thresholds set by the market and is used to identify extreme price events.

[0096] S200: Regional spot market operation risk evaluation index system, which builds a comprehensive evaluation model for regional spot market operation risk based on the entropy weight TOPSIS method;

[0097] S201: The establishment of a regional spot market operation risk assessment indicator system includes four primary indicators, namely, power supply and demand risk, market price abnormality risk, power system safe operation risk, and contract performance risk;

[0098] The secondary indicators corresponding to the first-level indicator of power supply and demand risk are supply and demand balance deviation rate, peak-valley load difference rate, and spare capacity adequacy rate;

[0099] The secondary indicators corresponding to the first-level indicator of abnormal market price risk are price volatility, price deviation, and extreme price frequency;

[0100] The first-level indicator of power system safe operation risk corresponds to the second-level indicators of system reliability index (SAIFI, SAIDI), power grid accident rate, and equipment failure rate;

[0101] The first-level indicator of performance risk corresponds to the second-level indicators of contract default rate, credit loss rate, and dispute resolution time.

[0102] S202: The construction of a comprehensive evaluation model for regional spot market operation risk based on the entropy weight TOPSIS method comprises the following steps:

[0103] Design and distribute questionnaires;

[0104] The entropy weight method is used to obtain the weights of the risk probability and risk loss consequences;

[0105] Substitute the weight results into the TOPSIS model;

[0106] Get the comprehensive evaluation index and sort the indicators from large to small.

[0107] S203: The questionnaire adopts a 7-level rating scale, and the risk probability P1 is divided from low to high into "very low (1 point); low (2 points); relatively low (3 points); average (4 points); relatively high (5 points); high (6 points); very high (7 points)"; the risk loss consequence P2 is divided into very low loss (1 point); low loss (2 points); relatively low loss (3 points); average loss (4 points); large loss (5 points); large loss (6 points); very large loss (7 points);

[0108] Then, the average values ​​of the scoring results collected from the questionnaire are calculated according to the two dimensions of risk probability P1 and risk loss consequence P2, and then the entropy weight method is used to calculate the weights of risk probability and risk loss consequence.

[0109] S204: The step of using the entropy weight method to obtain the weights of the risk occurrence probability and the risk loss consequences includes three steps: constructing an initialization decision matrix, standardizing indicators, and calculating the entropy value and weight of the risk occurrence probability and the risk loss consequences. The three steps are specifically as follows:

[0110] The construction of the initialization decision matrix can be expressed as follows:

[0111]

[0112] The initialization decision matrix is ​​constructed using the average value, where n represents the number of risk factor indicators being evaluated, m represents the risk importance of each indicator, and B i (i=1,2,……,n) represents the nth risk factor index B, C evaluated for the event j(j=1,2,……,m) represents the risk importance of the jth indicator of the event, x ij It indicates the importance of the i-th indicator, and all indicators are represented as x ij The form of is the decision matrix;

[0113] The questionnaire can be randomly distributed to multiple experts, enterprises, and governments to score the probability and the degree of loss. A 7-level scale model is used. The lower the score, the smaller the probability of the risk or the smaller the degree of loss. When the credibility requirement is met, the questionnaire can be used, and the entropy weight method can be used to obtain the weights of the risk probability and the risk loss consequences.

[0114] The index standardization process can be expressed as follows:

[0115]

[0116] Among them, x ij + The normalized formula for positive indices, x ij - The normalized formula for negative indices, x ij represents the importance of the i-th indicator, represents the minimum risk importance of the jth indicator, Indicates the highest risk importance of the jth indicator;

[0117] The selection of the positive and negative index normalization formula is based on the positive and negative relationship between the result and the index;

[0118] If the change of the result indicator is positively correlated, the normalization formula of the positive indicator is selected; if the change of the result indicator is negatively correlated, the normalization formula of the negative indicator is selected;

[0119] If negative values ​​appear after substituting numerical values ​​for index standardization, further data translation processing is required to eliminate negative values: ij '=h+x ij + or x ij '=h+x ij - , where h represents the amplitude of the index translation. Generally, h = 1. After dimensionless processing, the following expression is used:

[0120]

[0121] Among them, y ij represents the value after dimensionless processing, x ij ' represents the value after translation processing;

[0122] The entropy value and weight for calculating the risk occurrence probability and risk loss consequence can be expressed as follows:

[0123]

[0124] Among them, ω i represents the weight of the i-th indicator, G i It represents the coefficient of difference of the ith indicator, expressed as G i =1-E i ; E i Represents the entropy value of the i-th indicator, expressed as

[0125] Substituting the weight results into the TOPSIS model includes three steps: calculating the weighted normalization matrix, calculating the positive ideal solution and the negative ideal solution, and calculating the Euclidean distance;

[0126] The calculation weighted normalization matrix can be expressed as follows:

[0127] d ij =ω i f ij

[0128] Among them, d ij represents the weighted normalization matrix, ω i represents the weight of the i-th indicator, f ij Represents the data after vector normalization, expressed as or

[0129] The calculation of the positive ideal solution and the negative ideal solution may be performed using the following expressions:

[0130]

[0131] in, and denote the positive ideal solution and the negative ideal solution respectively, d ij represents the weighted normalization matrix;

[0132] The calculation of the Euclidean distance may be performed using the following expression:

[0133]

[0134] in, and Respectively represent the distance of each scheme i from the ideal point and the negative ideal point;

[0135] The obtaining of the comprehensive evaluation index and sorting the indicators from large to small includes two steps: calculating the comprehensive evaluation index and sorting the evaluation index, which are specifically expressed as follows:

[0136] The calculation of the comprehensive evaluation index can be performed using the following expression:

[0137]

[0138] Among them, L i Represents the comprehensive evaluation index value, and Respectively represent the distance of each scheme i from the ideal point and the negative ideal point;

[0139] The ranking of the evaluation indexes includes sorting the evaluation indexes according to the comprehensive evaluation index L i Sort the indicators from large to small and filter out key risk indicators.

[0140] S300: Solve the comprehensive evaluation model of regional spot market operation risk and conduct data analysis and processing;

[0141] S301: The solving of the regional spot market operation risk comprehensive evaluation model and performing data analysis and processing include the following steps:

[0142] Cluster analysis was performed using the K-means clustering method, and the selected key indicators were further divided into three categories using the SPSS tool, ultimately obtaining three cluster centers of risk values;

[0143] The following method is used to solve the comprehensive evaluation model of regional spot market operation risk to obtain the level of risk events:

[0144] Solving the model: inputting data and indicators through spass to obtain results;

[0145] Cluster analysis: Cluster analysis classifies data based on its own information, and divides data into several categories based on the distance, so that the similarity between data belonging to the same category is as large as possible, and the similarity between different types is as small as possible, so as to achieve classification;

[0146] Based on the data and data comparison distribution, SPSS tools are used to cluster key risk indicators;

[0147] Risk level classification: Based on the clustering results, risk events are classified into different risk levels. Each cluster represents a risk level, and events with similar risk characteristics are classified into the same level.

[0148] Taking the "supply and demand balance deviation rate" indicator as an example, the following calculation process is described in detail:

[0149] Collect 30 days of data:

[0150] date Actual load (MW) Forecast load (MW) Deviation rate (%) 1 10000 10200 2.00 2 10500 10300 -1.90 ... ... ... ... 30 11000 11100 0.91

[0151] Substituting the above mentioned expressions into the calculation, we can get the following data:

[0152] Calculate the average deviation rate: 1.5%;

[0153] Normalization: Divide the daily deviation rate by the maximum deviation rate;

[0154] Calculate the entropy value: The entropy value obtained is 0.85;

[0155] Calculate the weight: The weight obtained is 0.08;

[0156] Construct a weighted normalization matrix;

[0157] Determine the ideal solution (0% deviation) and the negative ideal solution (maximum deviation);

[0158] Calculate Euclidean distance;

[0159] Finally, the comprehensive evaluation index was obtained: 0.72;

[0160] A similar calculation process was performed on the other 11 indicators, and finally we obtained the 12 comprehensive evaluation index values ​​we listed.

[0161] The other 11 indicators are as follows:

[0162] Supply and demand balance deviation rate: 0.72; peak-valley load difference rate: 0.65; spare capacity adequacy rate: 0.58; price volatility: 0.81; price deviation: 0.76; extreme price frequency: 0.69; SAIFI: 0.54; SAIDI: 0.52; power grid accident rate: 0.48; equipment failure rate: 0.51; contract default rate: 0.63; credit loss rate: 0.59;

[0163] After using SPSS for K-means cluster analysis, the following results can be obtained:

[0164] Cluster Center:

[0165] Cluster 1 (low risk): 0.53;

[0166] Cluster 2 (medium risk): 0.64;

[0167] Cluster 3 (high risk): 0.78;

[0168] Clustering results:

[0169] Low-risk group: SAIFI, SAIDI, power grid accident rate, equipment failure rate;

[0170] Medium risk group: peak-valley load difference rate, reserve capacity adequacy rate, extreme price frequency, contract default rate, credit loss rate;

[0171] High-risk group: supply-demand balance deviation rate, price volatility, and price deviation.

[0172] S400: Determine the current risk level based on the results of data analysis and processing;

[0173] S401: Determining the risk level based on the results of data analysis and processing includes the following steps:

[0174] Based on the above clustering results, the risk warning can be divided into three levels according to the risk value according to the three cluster centers: low risk, medium risk and high risk, and a radar chart of the indicator differences at different risk levels can be drawn and displayed in the decision-making module;

[0175] Based on the above clustering results, the current risk level can be determined:

[0176] The risk of safe operation of the power system is at a low risk level, with index values ​​ranging from 0.48 to 0.54, lower than the low-risk cluster center of 0.53;

[0177] The power supply and demand risk and contract performance risk are at a medium risk level, with index values ​​between 0.58 and 0.65, close to the medium risk cluster center of 0.64;

[0178] The risk of abnormal market prices is at a high risk level, with the three indicator values ​​ranging from 0.69 to 0.81, which is higher than the high-risk cluster center of 0.78.

[0179] Comprehensively judging, the overall risk level of the current regional electricity spot market is medium to high, among which the risk of abnormal market prices is the most prominent and requires special attention and control.

[0180] S500: Develop early warning and response mechanisms at different levels.

[0181] S501: The formulation of different levels of early warning and response mechanisms includes the following steps:

[0182] Corresponding warnings are issued according to different levels of risks, and alarms are issued when specific thresholds are reached; specific response measures are then formulated based on the warnings, including preventive measures, mitigation measures and emergency measures, so that quick action can be taken when risks occur.

[0183] Based on the above risk assessment results, the following early warning and response mechanism can be formulated:

[0184] Low risk warning (power system safe operation risk):

[0185] Warning: When any of the indicators of SAIFI, SAIDI, power grid accident rate or equipment failure rate exceeds 0.60, a yellow warning is issued;

[0186] Response measures:

[0187] a) Strengthen the frequency of daily equipment inspection and maintenance; b) Optimize the scheduling plan to ensure stable operation of the system; c) Conduct emergency drills regularly to improve fault handling capabilities.

[0188] Medium risk warning (power supply and demand risk and contract performance risk):

[0189] Warning: When any of the peak-valley load difference rate, spare capacity adequacy rate, extreme price frequency, contract default rate or credit loss rate exceeds 0.70, an orange warning will be issued;

[0190] Response measures:

[0191] a) Implement demand-side response to smooth out peak-to-valley differences; b) Appropriately increase spare capacity to improve system flexibility; c) Strengthen credit management of market entities and implement dynamic credit ratings; d) Improve contract performance guarantee mechanisms and increase penalties for breach of contract.

[0192] High-risk warning (abnormal market price risk):

[0193] Warning: When any indicator of supply-demand balance deviation rate, price volatility or price deviation exceeds 0.85, a red warning is issued.

[0194] Response measures:

[0195] a) Initiate price intervention mechanism and implement temporary price caps and lower limits; b) Increase market supervision and crack down on price manipulation; c) Optimize clearing algorithms and improve market clearing efficiency; d) Improve market information disclosure mechanism and increase market transparency; e) Initiate emergency power generation and cross-regional power dispatch when necessary.

[0196] Furthermore, this embodiment also provides a spot market risk prevention and control system based on real-time data analysis, including:

[0197] Acquisition module, real-time acquisition of regional power spot market operation data;

[0198] Construct modules, establish a regional spot market operation risk evaluation index system, and construct a regional spot market operation risk comprehensive evaluation model based on the entropy weight TOPSIS method;

[0199] The solution module solves the comprehensive evaluation model of regional spot market operation risks and performs data analysis and processing;

[0200] The judgment module determines the current risk level based on the results of data analysis and processing;

[0201] Early warning module, formulate early warning and response mechanisms at different levels.

[0202] This embodiment also provides a computer device, which is suitable for the case of a spot market risk prevention and control method based on real-time data analysis, including a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the spot market risk prevention and control method based on real-time data analysis proposed in the above embodiment.

[0203] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0204] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for risk prevention and control in the spot market based on real-time data analysis as proposed in the above embodiment is implemented.

[0205] In summary, by acquiring regional electricity spot market operation data in real time, including historical and real-time power load, power generation capacity, spare capacity, price and other multi-dimensional information, we can achieve a comprehensive and real-time grasp of the market operation status, provide a comprehensive and timely data basis for subsequent risk assessment, and effectively improve the accuracy and timeliness of risk identification.

[0206] By establishing a multi-level evaluation index system including power supply and demand risk, price anomaly risk, system safe operation risk and contract performance risk, and combining the entropy weight TOPSIS method to build a comprehensive evaluation model, a multi-dimensional and comprehensive evaluation of spot market risks is achieved. Not only single risk factors are considered, but also the interaction between various risk factors is considered, which greatly improves the comprehensiveness and accuracy of risk assessment.

[0207] The entropy weight method is used to determine the indicator weights, avoiding the bias that may be caused by subjective weighting, making the risk assessment results more objective and reliable. At the same time, the application of the TOPSIS method makes the assessment results highly comparable and helps identify key risk factors.

[0208] By using the K-means clustering method to classify risks and develop corresponding early warning and response mechanisms, accurate risk classification and differentiated management are achieved. This not only improves the pertinence and efficiency of risk management, but also provides decision makers with a clear risk situation map, which helps to quickly formulate response strategies.

[0209] The entire solution forms a closed-loop risk management process, from data acquisition, risk assessment to early warning response, with each link closely connected, realizing dynamic and continuous monitoring of spot market risks. This systematic management method significantly improves the market risk prevention and control capabilities and effectively reduces the impact of emergencies on the market.

[0210] The implementation of this plan will not only help improve the stability and predictability of the electricity spot market, but also provide more transparent and reliable market information to market participants, promote fair competition and healthy development of the market. At the same time, it will provide scientific decision-making support for regulatory agencies, which will be conducive to the continuous optimization and improvement of the electricity market mechanism.

[0211] Example 2

[0212] Reference Figure 2 - Figure 3 , which is the second embodiment of the present invention, and this embodiment provides a spot market risk prevention and control method based on real-time data analysis. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0213] In order to verify the effectiveness of the spot market risk prevention and control method based on real-time data analysis proposed in this invention, the operating data of a regional power spot market from January to December 2023 were selected for analysis. The region includes 5 provincial power grids with a total installed capacity of about 200 million kilowatts, of which new energy installed capacity accounts for 40%.

[0214] First, an evaluation system of 12 key indicators including power supply and demand risk, market price abnormality risk, power system safe operation risk and contract performance risk was established. Through the real-time data collection system, data of various indicators were collected every 15 minutes, and preliminary processing and storage were carried out.

[0215] Next, the entropy weight TOPSIS method was used to construct a comprehensive evaluation model. To ensure the objectivity of weight allocation, 50 experts from power companies, regulatory agencies and academia were invited to conduct a questionnaire survey, using a 7-level scale scoring method. After the questionnaires were collected, the entropy weight method was used to calculate the weights of each indicator. The results showed that the price volatility, supply and demand balance deviation rate and reserve capacity adequacy rate had higher weights, which were 0.15, 0.13 and 0.11 respectively.

[0216] Subsequently, the 12-month daily average data was substituted into the TOPSIS model to calculate the daily comprehensive risk index. In order to dynamically reflect the risk status, the 7-day sliding window method was used to calculate the moving average risk index. At the same time, the exponential smoothing method was introduced to make short-term predictions of the risk index, with a prediction period of 3 days.

[0217] In terms of risk classification, K-means clustering and fuzzy C-means clustering algorithms are innovatively combined. First, K-means clustering is used to preliminarily divide risks into three levels, and then fuzzy C-means clustering is applied to refine the attribution of boundary cases, which improves the accuracy and robustness of classification. Cluster analysis is implemented using MATLAB software, with the number of iterations set to 100 and the convergence threshold set to 0.001.

[0218] In order to improve the timeliness and accuracy of risk warning, this method also integrates an anomaly detection module based on the long short-term memory network (LSTM). By learning the temporal characteristics of historical data, this module can quickly identify abnormal values ​​that deviate from the normal pattern and provide additional judgment basis for risk warning. The LSTM network adopts a 2-layer structure, each layer contains 64 neurons, and is trained using the Adam optimizer with a learning rate set to 0.001.

[0219] Finally, based on the risk assessment results, a three-level early warning mechanism and corresponding response strategy were designed. In a low-risk state, the system updates the risk report every 6 hours; in a medium-risk state, it updates every 2 hours and initiates daily prevention and control measures; in a high-risk state, the system updates every 30 minutes and activates the emergency response plan. In addition, a visual dashboard was developed to display the status of various indicators and risk trends in real time, supporting decision makers to conduct intuitive analysis and make quick decisions.

[0220] The experimental results are shown in the following table:

[0221]

[0222]

[0223] By comparing the experimental results, it can be clearly seen that our solution has achieved significant improvements in multiple key indicators.

[0224] First, in terms of risk identification accuracy, our solution reached 94.3%, an increase of 14.3 percentage points compared with traditional methods. This significant improvement is mainly due to the establishment of a multi-dimensional evaluation index system and the application of the entropy weight TOPSIS method. The multi-dimensional index system comprehensively covers key risk factors such as power supply and demand, price anomalies, system safety and contract performance, while the introduction of the entropy weight method effectively avoids the bias that may be caused by subjective weighting, making the risk assessment results more objective and credible.

[0225] In terms of early warning timeliness, our solution has shortened the time from identifying risks to issuing early warnings from 45 minutes to 18 minutes, an improvement of 60%. This improvement is mainly due to the application of real-time data collection system and the integration of LSTM anomaly detection module. High-frequency data collection every 15 minutes ensures that the system can capture market changes in a timely manner, while the LSTM module can quickly identify abnormal patterns by learning the time series characteristics of historical data, greatly improving the timeliness of early warning.

[0226] The significant reduction in false positive and false negative rates (down 63.2% and 67.7% respectively) fully demonstrates the advantage of our solution in risk identification accuracy. This is mainly due to the innovative combination of K-means clustering and fuzzy C-means clustering algorithms. By refining the boundary cases, the accuracy and robustness of risk classification are effectively improved, thereby reducing the occurrence of false positives and false negatives.

[0227] The response time was shortened from 35 minutes to 12 minutes, an increase of 65.7%, which reflects the significant advantage of our solution in risk response efficiency. The three-level early warning mechanism and corresponding response strategy design, combined with real-time updated risk reports and visual dashboards, enable decision makers to quickly and accurately grasp the risk situation and respond, greatly improving the efficiency of risk management.

[0228] In addition, the two indicators of a 28.6% reduction in risk prevention and control costs and a 22.4% improvement in market stability further demonstrate the comprehensive benefits of our solution in practical applications. The cost reduction mainly comes from more accurate risk identification and more efficient response measures, avoiding unnecessary excessive intervention. The improvement in market stability reflects the positive role of our solution in maintaining market order and enhancing the confidence of market participants.

[0229] It is worth noting that our solution not only has obvious improvements in individual indicators, but more importantly, it has achieved systematization and intelligence of the entire risk management process. From data collection, risk assessment, early warning release to response measures formulation, a closed-loop intelligent risk management system has been formed. This systematic approach has greatly enhanced the risk prevention and control capabilities of the power spot market and provided a strong guarantee for the healthy and stable operation of the market.

[0230] In summary, our solution shows obvious advantages and innovation in terms of risk identification accuracy, early warning timeliness, false alarm and missed alarm control, response efficiency and comprehensive benefits. It not only solves the shortcomings of traditional methods in dealing with complex and changeable power spot market risks, but also provides new technical support for the stable operation and sustainable development of the power market through intelligent and systematic management methods. These improvements will undoubtedly have a far-reaching impact on improving the overall operation efficiency and risk prevention and control level of the power market.

[0231] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A spot market risk prevention and control method based on real-time data analysis, characterized in that: include, Obtain regional electricity spot market operation data in real time; Establish a regional spot market operation risk evaluation index system and construct a regional spot market operation risk comprehensive evaluation model based on the entropy weight TOPSIS method; Solve the comprehensive evaluation model of regional spot market operation risks and conduct data analysis and processing; Determine the current risk level based on the results of data analysis and processing; Develop early warning and response mechanisms at different levels.

2. The spot market risk prevention and control method based on real-time data analysis according to claim 1, characterized in that: The real-time acquisition of regional power spot market operation data includes historical power load and real-time power load; power generation capacity; spare capacity; clearing price; Long-term average price; price limits; number and duration of power outages; Number of failures; Number of contracts; Number of contract breaches; Number of contract disputes; The power generation capacity data represents the installed capacity and actual output information of various types of power generation units; The reserve capacity data represents the reserve capacity available in the system, including both spinning reserve and non-spinning reserve capacity; Long-term average price data represents the long-term average or expected price of historical electricity market prices and is used to calculate price deviation; Price limits data characterizes upper and lower price thresholds set by the market and is used to identify extreme price events.

3. The spot market risk prevention and control method based on real-time data analysis according to claim 2, characterized in that: The establishment of a regional spot market operation risk assessment indicator system includes four primary indicators: power supply and demand risk, market price abnormality risk, power system safe operation risk, and contract performance risk; The secondary indicators corresponding to the first-level indicator of power supply and demand risk are supply and demand balance deviation rate, peak-valley load difference rate, and spare capacity adequacy rate; The secondary indicators corresponding to the first-level indicator of abnormal market price risk are price volatility, price deviation, and extreme price frequency; The first-level indicator of power system safe operation risk corresponds to the second-level indicators of system reliability index (SAIFI, SAIDI), power grid accident rate, and equipment failure rate; The first-level indicator of performance risk corresponds to the second-level indicators of contract default rate, credit loss rate, and dispute resolution time.

4. The spot market risk prevention and control method based on real-time data analysis according to claim 3, characterized in that: The construction of a comprehensive evaluation model for regional spot market operation risk based on the entropy weight TOPSIS method includes the following steps: Design and distribute questionnaires; The entropy weight method is used to obtain the weights of the risk probability and risk loss consequences; Substitute the weight results into the TOPSIS model; Get the comprehensive evaluation index and sort the indicators from large to small.

5. The spot market risk prevention and control method based on real-time data analysis according to claim 4, characterized in that: The questionnaire adopts a 7-level rating scale, and the probability of risk occurrence P1 is divided from low to high into "very low (1 point); low (2 points); relatively low (3 points); average (4 points); relatively high (5 points); high (6 points); very high (7 points)"; the risk loss consequence P2 is divided into very low loss (1 point); low loss (2 points); relatively low loss (3 points); average loss (4 points); large loss (5 points); large loss (6 points); very large loss (7 points); Then, the average values ​​of the scoring results collected from the questionnaire are calculated according to the two dimensions of risk probability P1 and risk loss consequence P2, and then the entropy weight method is used to calculate the weights of risk probability and risk loss consequence.

6. The spot market risk prevention and control method based on real-time data analysis according to claim 5, characterized in that: The step of using the entropy weight method to obtain the weights of the risk occurrence probability and the risk loss consequences includes three steps: constructing an initialization decision matrix, standardizing indicators, and calculating the entropy value and weight of the risk occurrence probability and the risk loss consequences. The three steps are specifically as follows: The construction of the initialization decision matrix can be expressed as follows: The initialization decision matrix is ​​constructed using the average value, where n represents the number of risk factor indicators being evaluated, m represents the risk importance of each indicator, and B i (i=1,2,……,n) represents the nth risk factor index B, C evaluated for the event j (j=1,2,……,m) represents the risk importance of the jth indicator of the event, x ij It indicates the importance of the i-th indicator, and all indicators are represented as x ij The form of is the decision matrix; The questionnaire can be randomly distributed to multiple experts, enterprises, and governments to score the probability and the degree of loss. A 7-level scale model is used. The lower the score, the smaller the probability of the risk or the smaller the degree of loss. When the credibility requirement is met, the questionnaire can be used, and the entropy weight method can be used to obtain the weights of the risk probability and the risk loss consequences. The index standardization process can be expressed as follows: or Among them, x ij + The normalized formula for positive indices, x ij - The normalized formula for negative indices, x ij represents the importance of the i-th indicator, represents the minimum risk importance of the jth indicator, Indicates the highest risk importance of the jth indicator; The selection of the positive and negative index normalization formula is based on the positive and negative relationship between the result and the index; If the change of the result indicator is positively correlated, the normalization formula of the positive indicator is selected; if the change of the result indicator is negatively correlated, the normalization formula of the negative indicator is selected; If negative values ​​appear after substituting numerical values ​​for index standardization, further data translation processing is required to eliminate negative values: ij '=h+x ij + or x ij '=h+x ij - , where h represents the amplitude of the index translation. Generally, h = 1. After dimensionless processing, the following expression is used: Among them, y ij represents the value after dimensionless processing, x ij ' represents the value after translation processing; The entropy value and weight for calculating the risk occurrence probability and risk loss consequence can be expressed as follows: Among them, ω i represents the weight of the i-th indicator, G i It represents the coefficient of difference of the ith indicator, expressed as G i =1-E i ; E i Represents the entropy value of the i-th indicator, expressed as (i=1,2,…,n; j=1,2,…,m); Substituting the weight results into the TOPSIS model includes three steps: calculating the weighted normalization matrix, calculating the positive ideal solution and the negative ideal solution, and calculating the Euclidean distance; The calculation weighted normalization matrix can be expressed as follows: d ij =ω i f ij Among them, d ij represents the weighted normalization matrix, ω i represents the weight of the i-th indicator, f ij Represents the data after vector normalization, expressed as or The calculation of the positive ideal solution and the negative ideal solution may be performed using the following expressions: in, and denote the positive ideal solution and the negative ideal solution respectively, d ij represents the weighted normalization matrix; The calculation of the Euclidean distance may be performed using the following expression: in, and Respectively represent the distance of each scheme i from the ideal point and the negative ideal point; The obtaining of the comprehensive evaluation index and sorting the indicators from large to small includes two steps: calculating the comprehensive evaluation index and sorting the evaluation index, which are specifically expressed as follows: The calculation of the comprehensive evaluation index can be performed using the following expression: Among them, L i Represents the comprehensive evaluation index value, and Respectively represent the distance of each scheme i from the ideal point and the negative ideal point; The ranking of the evaluation indexes includes sorting the evaluation indexes according to the comprehensive evaluation index L i Sort the indicators from large to small and filter out key risk indicators.

7. The spot market risk prevention and control method based on real-time data analysis according to claim 6, characterized in that: The method of solving the comprehensive evaluation model of regional spot market operation risk and performing data analysis and processing includes the following steps: Cluster analysis was performed using the K-means clustering method, and the selected key indicators were further divided into three categories using the SPSS tool, ultimately obtaining three cluster centers of risk values; The following method is used to solve the comprehensive evaluation model of regional spot market operation risk to obtain the level of risk events: Solving the model: inputting data and indicators through spass to obtain results; Cluster analysis: Cluster analysis classifies data based on its own information, and divides data into several categories based on the distance, so that the similarity between data belonging to the same category is as large as possible, and the similarity between different types is as small as possible, so as to achieve classification; Based on the data and data comparison distribution, the key risk indicators are clustered using SPSS tools; Risk level classification: Based on the clustering results, risk events are classified into different risk levels. Each cluster represents a risk level, and events with similar risk characteristics are classified into the same level. Determining the risk level based on the results of data analysis and processing includes the following steps: Based on the above clustering results, the risk warning can be divided into three levels according to the risk value according to the three cluster centers: low risk, medium risk and high risk, and a radar chart of the indicator differences at different risk levels can be drawn and displayed in the decision-making module; The formulation of different levels of early warning and response mechanisms includes the following steps: Corresponding warnings are issued according to different levels of risks, and alarms are issued when specific thresholds are reached; specific response measures are then formulated based on the warnings, including preventive measures, mitigation measures and emergency measures, so that quick action can be taken when risks occur.

8. A spot market risk prevention and control system based on real-time data analysis, based on the spot market risk prevention and control method based on real-time data analysis according to any one of claims 1 to 7, characterized in that: Also includes, Acquisition module, real-time acquisition of regional power spot market operation data; Construct modules, establish a regional spot market operation risk evaluation index system, and construct a regional spot market operation risk comprehensive evaluation model based on the entropy weight TOPSIS method; The solution module solves the comprehensive evaluation model of regional spot market operation risks and performs data analysis and processing; The judgment module determines the current risk level based on the results of data analysis and processing; Early warning module, formulate early warning and response mechanisms at different levels.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the spot market risk prevention and control method based on real-time data analysis described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the spot market risk prevention and control method based on real-time data analysis described in any one of claims 1 to 7 are implemented.

Images