A method and system for monitoring power market performance behavior

By applying a language item description method based on intuitive fuzzy sets in the power market, real-time monitoring and evaluation of the performance risks of the power market, the problem of not being able to effectively identify false data and performance risks in the existing technology is solved, and more efficient and flexible Risk monitoring and management.

CN117455232BActive Publication Date: 2024-11-29GUANGDONG ELECTRIC POWER TRADING CENT CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202311406692.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-10-26
Publication Date
2024-11-29
Estimated Expiration
2043-10-26

AI Technical Summary

Technical Problem

The existing power market risk monitoring system is unable to effectively monitor and identify false data in market transactions, resulting in poor prevention of performance risks, lack of flexibility and adaptability, and it is difficult to cope with complex market transactions.

Method used

Using a language term description method based on intuitive fuzzy sets, the power market operation data is obtained in real time, and the indicators to be evaluated for the types of risk to be evaluated are obtained through feature extraction, the decision-maker evaluation opinions are integrated, and the performance risk score is calculated to achieve real-time monitoring of performance behavior. and effective control.

Benefits of technology

It improves the accuracy and adaptability of monitoring of performance behavior in the power market, can more effectively identify and evaluate market risks, and helps regulators and corporate management take risk response measures in a timely manner.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117455232B_ABST
    Figure CN117455232B_ABST
Patent Text Reader

Abstract

The present invention discloses a method and system for monitoring the performance of an electricity market, including: obtaining the operation data of an electricity market in real time; extracting features from the operation data to obtain a plurality of indicators to be evaluated corresponding to the risk type to be evaluated; using a language item description based on an intuitionistic fuzzy set to describe the performance and importance of each indicator to be evaluated, and screening out a plurality of decision makers associated with each indicator to be evaluated from an expert group; obtaining the indicator evaluation opinions obtained by each decision maker evaluating the importance and performance of each indicator to be evaluated using the language item description, and then integrating the indicator evaluation opinions of all decision makers to obtain the performance risk score of the risk type to be evaluated as the performance monitoring result of the electricity market. The present invention uses a language item description based on an intuitionistic fuzzy set to enable the indicator evaluation opinions to be closer to the subjective perception of the decision maker, thereby optimizing the overall monitoring effect of the performance of the electricity market.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of power market contract performance monitoring, and in particular to a power market contract performance monitoring method and system. Background Art

[0002] The existing power market risk monitoring system generally adopts a rule-based analysis method, which identifies or evaluates possible abnormal market behaviors and risks through pre-set rules, such as a pre-established medium- and long-term power market risk indicator system and evaluation standards.

[0003] However, driven by interests, there may be unreasonable and false data performance behaviors in the power market transaction data, which affects the performance risk prevention effect of the power market. The above method is not based on the real-time monitoring data and data verification technology of the power market entities, and cannot establish effective supervision of the authenticity of transaction data and performance behaviors, and cannot effectively identify false data in market transactions. It lacks flexibility and adaptability, and it is difficult to cope with the increasingly complex market transaction situation, thus affecting the effect of market risk analysis, and it is difficult to achieve real-time monitoring and effective control of various existing transactions and abnormal performance behaviors in the power market. Summary of the invention

[0004] The embodiment of the present invention provides a method and system for monitoring contract performance in the electricity market, which uses language item description based on intuitionistic fuzzy sets to express subjective evaluation, so that the indicator evaluation opinions can be closer to the subjective perception of decision makers, thereby improving the accuracy of monitoring contract performance in the electricity market.

[0005] In order to solve the above technical problems, an embodiment of the present invention provides a method for monitoring performance in an electricity market, comprising:

[0006] Acquire the operational data of the power market in real time; wherein the power market includes a number of market players, a market trading platform and a power dispatching platform;

[0007] Extracting features from the operational data to obtain a plurality of indicators to be evaluated corresponding to the risk type to be evaluated;

[0008] Using language item descriptions based on intuitionistic fuzzy sets, the performance and importance of each of the indicators to be evaluated are described, and multiple decision makers associated with each of the indicators to be evaluated are screened out from the expert group;

[0009] Obtain the indicator evaluation opinions of each decision maker, and then integrate the indicator evaluation opinions of all the decision makers to obtain the performance risk score of the risk type to be evaluated as the performance behavior monitoring result of the power market; wherein the indicator evaluation opinions are obtained by the decision makers using language items to describe the importance and performance of each indicator to be evaluated.

[0010] By implementing the embodiment of the present invention, by acquiring the operational data of the power market in real time, relevant information of market entities, market trading platforms and power dispatching platforms can be timely understood, including market participants, trading behaviors, market prices and power supply and demand, etc., to provide real-time and effective data support for the risk assessment of the power market operation monitoring and performance behavior, and then feature extraction is performed on the operational data to obtain multiple indicators to be evaluated corresponding to the risk type to be evaluated, and then the performance and importance of each indicator to be evaluated are described using the language item description based on the intuitive fuzzy set, so that the subjective evaluation can be expressed in fuzzy language, so that the indicator evaluation opinions can be closer to the subjective perception of the decision maker, which is convenient for subsequent decision analysis and evaluation. In addition, the evaluation opinions of each decision maker on the evaluation indicators are collected and integrated, and the views of different decision makers are comprehensively considered to cope with the increasingly complex market trading situation, and the performance risk score of the risk type to be evaluated is obtained as an indicator for measuring the performance risk of the power market, which can help market regulatory agencies or enterprise management to monitor the overall risk status of the market so as to take corresponding management measures and risk response strategies in a timely manner.

[0011] As a preferred solution, the indicator evaluation opinions of each decision maker are obtained, and then the indicator evaluation opinions of all decision makers are integrated to obtain the performance risk score of the risk type to be evaluated as the performance behavior monitoring result of the power market, specifically:

[0012] Traversing each indicator to be evaluated, obtaining the indicator level evaluation opinions of each decision maker on the currently traversed indicator to be evaluated, and fusing the indicator level evaluation opinions corresponding to each decision maker according to the decision weight of each decision maker, and constructing a first intuitive fuzzy decision vector based on the opinion fusion results of all the indicators to be evaluated; wherein the indicator level evaluation opinions are obtained by the decision maker using language items to describe the evaluation of the performance of each indicator to be evaluated;

[0013] Traverse each indicator to be evaluated, obtain the indicator importance evaluation opinions of each decision maker on the currently traversed indicator to be evaluated, and fuse the indicator importance evaluation opinions corresponding to each decision maker according to the decision weight of each decision maker, and construct a weight vector based on the opinion fusion results of all the indicators to be evaluated; wherein the indicator importance evaluation opinions are obtained by the decision maker using language items to describe the evaluation of the performance of each indicator to be evaluated;

[0014] Using the weight vector, weighting the first intuitive fuzzy decision vector to obtain a second intuitive fuzzy decision vector;

[0015] According to the preset comprehensive scoring algorithm, based on the second intuitive fuzzy decision vector, the performance risk score of the risk type to be evaluated is calculated as the performance behavior monitoring result of the electricity market.

[0016] The preferred scheme of the embodiment of the present invention is implemented by traversing each indicator to be evaluated, obtaining the evaluation opinions of each decision maker on the indicator, and fusing the evaluation opinions corresponding to different decision makers according to the decision weights of the decision makers. In this way, the views and weights of different decision makers can be comprehensively considered to make the evaluation results more comprehensive and objective. Then, according to the fusion results of the opinions of all the indicators to be evaluated, a first intuitive fuzzy decision vector is constructed. The first intuitive fuzzy decision vector is used to reflect the relative importance and performance of different indicators to be evaluated in the eyes of decision makers. Moreover, the language item description of the intuitive fuzzy set is closer to the subjective perception of the decision maker, reducing the ambiguity of the evaluation.

[0017] As a preferred solution, the decision weight of the decision maker is obtained as follows:

[0018] Using the language term description based on the intuitionistic fuzzy set, the importance of the evaluation opinions of each decision maker is described, and the importance level and intuitionistic fuzzy number corresponding to each decision maker are obtained;

[0019] According to the preset decision weight algorithm, the decision weight of each decision maker is calculated according to the importance level and intuitive fuzzy number corresponding to each decision maker;

[0020] The intuitive fuzzy number corresponding to the decision maker includes the membership and non-membership corresponding to the decision maker. The preferred solution of the embodiment of the present invention uses a linguistic term description based on an intuitive fuzzy set to describe the importance of the evaluation opinions of each decision maker, and obtains the importance level and intuitive fuzzy number corresponding to each decision maker, wherein the importance level reflects the relative importance of the decision maker to the evaluation opinion, and the intuitive fuzzy number includes membership and non-membership, which is used to describe the uncertainty and ambiguity of the decision maker to the importance level. By adopting the linguistic term description based on an intuitive fuzzy set and the decision weight calculation method, the decision-making process is made more transparent and explainable.

[0021] As a preferred solution, the feature extraction of the operation data is performed to obtain multiple indicators to be evaluated for the risk type to be evaluated, specifically:

[0022] Analyze and process the operation data to obtain corresponding multiple first indicators;

[0023] Through the pre-constructed risk classification model, the first indicator associated with the risk type to be assessed is selected as the indicator to be evaluated corresponding to the risk type to be assessed.

[0024] By implementing the preferred scheme of the embodiment of the present invention and analyzing and processing the operating data, multiple indicators related to the performance risks in the electricity market can be extracted, and through a pre-constructed risk classification model, indicators associated with the risk type to be assessed are selected from multiple first indicators as indicators to be assessed, so that the indicators to be assessed can be subsequently used to monitor the performance behavior risks related to the risk type to be assessed.

[0025] As a preferred solution, the construction of the risk classification model is specifically as follows:

[0026] Acquire historical operation data of the power market and a plurality of second indicators corresponding to the historical operation data;

[0027] According to the data type of the historical operation data, a corresponding mining model is selected, and then the historical operation data is mined using the mining model to obtain several risk types corresponding to the historical operation data, and a correlation coefficient between each of the second indicators and a risk feature library corresponding to each of the risk types, and the risk feature library corresponding to the risk type is used as a risk feature library associated with each of the second indicators;

[0028] A training data set is formed by using a risk feature library associated with each of the second indicators and a correlation coefficient between each of the second indicators and each of the risk feature libraries;

[0029] The obtained training data set is used to train and optimize the parameters of the pre-built initial classification model to obtain the risk classification model.

[0030] By implementing the preferred scheme of the embodiment of the present invention, through the use of data mining technology, risk-related information and patterns can be mined from the historical operating data of the power market, the risk characteristics and trends of the power market can be analyzed, and based on the mining results, a training data set can be constructed for training and parameter optimization of a pre-built initial classification model to improve the classification accuracy of the risk classification model, thereby further improving the accuracy and comprehensiveness of the monitoring of power market performance behavior.

[0031] As a preferred solution, the real-time acquisition of the operation data of the power market is specifically as follows:

[0032] By installing a data processing device on any of the market entities, data collection and verification of the power market are performed in real time to obtain relevant data of the power market;

[0033] Preprocessing the relevant data of the power market to obtain the operation data of the power market;

[0034] Among them, preprocessing includes outlier removal and noise elimination.

[0035] By implementing the preferred scheme of the embodiment of the present invention, by installing a data processing device in a market entity, data collection and verification of the power market can be carried out in real time, and the collected data can be pre-processed by outlier deletion and noise elimination to prevent the presence of outliers from having a negative impact on subsequent analysis and model building, thereby reducing random errors in power market data and improving data accuracy and credibility.

[0036] In order to solve the same technical problem, an embodiment of the present invention further provides a power market performance behavior monitoring system, including:

[0037] A data acquisition module, used to acquire the operational data of the power market in real time; wherein the power market includes a number of market entities, a market trading platform and a power dispatching platform;

[0038] A feature extraction module is used to extract features from the operation data to obtain a plurality of indicators to be evaluated corresponding to the risk type to be evaluated;

[0039] A description screening module is used to describe the performance and importance of each of the indicators to be evaluated using a linguistic term description based on an intuitionistic fuzzy set, and screen out a plurality of decision makers associated with each of the indicators to be evaluated from the expert group;

[0040] The behavior monitoring module is used to obtain the indicator evaluation opinions of each decision maker, and then integrate the indicator evaluation opinions of all the decision makers to obtain the performance risk score of the risk type to be evaluated as the performance behavior monitoring result of the power market; wherein the indicator evaluation opinion is obtained by the decision maker using language items to describe the importance and performance of each indicator to be evaluated.

[0041] As a preferred solution, the behavior monitoring module specifically includes:

[0042] The first fusion unit is used to traverse each indicator to be evaluated, obtain the indicator level evaluation opinions of each decision maker on the currently traversed indicator to be evaluated, and fuse the indicator level evaluation opinions corresponding to each decision maker according to the decision weight of each decision maker, and construct a first intuitive fuzzy decision vector based on the opinion fusion results of all the indicators to be evaluated; wherein the indicator level evaluation opinions are obtained by the decision maker using language items to describe the evaluation of the performance of each indicator to be evaluated;

[0043] The second fusion unit is used to traverse each indicator to be evaluated, obtain the indicator importance evaluation opinions of each decision maker on the currently traversed indicator to be evaluated, and fuse the indicator importance evaluation opinions corresponding to each decision maker according to the decision weight of each decision maker, and construct a weight vector according to the opinion fusion results of all the indicators to be evaluated; wherein the indicator importance evaluation opinions are obtained by the decision maker using language items to describe the evaluation of the performance of each indicator to be evaluated;

[0044] A weighting unit, configured to weight the first intuitive fuzzy decision vector using the weight vector to obtain a second intuitive fuzzy decision vector;

[0045] A calculation unit is used to calculate the performance risk score of the risk type to be evaluated according to a preset comprehensive scoring algorithm and the second intuitive fuzzy decision vector as a performance behavior monitoring result of the power market.

[0046] As a preferred solution, the power market performance monitoring system further includes:

[0047] The decision weight analysis module is used to describe the importance of the evaluation opinions of each decision maker using a language term description based on an intuitionistic fuzzy set, and obtain the importance level and intuitionistic fuzzy number corresponding to each decision maker; according to a preset decision weight algorithm, based on the importance level and intuitionistic fuzzy number corresponding to each decision maker, the decision weight of each decision maker is calculated; wherein the intuitionistic fuzzy number corresponding to the decision maker includes the membership degree and non-membership degree corresponding to the decision maker.

[0048] As a preferred solution, the feature extraction module specifically includes:

[0049] An analysis unit, configured to analyze and process the operation data to obtain a corresponding plurality of first indicators;

[0050] The selection unit is used to select the first indicator associated with the risk type to be evaluated through a pre-built risk classification model as the indicator to be evaluated corresponding to the risk type to be evaluated. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 : A flow chart of a method for monitoring contract performance in a power market provided in Embodiment 1 of the present invention;

[0052] Figure 2 : A structural diagram of a power market compliance behavior monitoring system provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0053] 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.

[0054] Embodiment one:

[0055] Please refer to Figure 1 , is a method for monitoring the performance of power market provided by an embodiment of the present invention, the method comprises steps S1 to S4, each step is specifically as follows:

[0056] Step S1, obtaining the operation data of the power market in real time.

[0057] Among them, the electricity market includes several market players, market trading platforms and power dispatching platforms.

[0058] In this embodiment, the power market operation data refers to the power market operation data related to the power market performance, including but not limited to market price, power sales and purchase, power load, line operation status, power plant power generation capacity and surplus capacity, transaction execution status, etc. Specifically, the power market operation data obtained in real time includes the following contents (1) to (5):

[0059] (1) Market trading platform: market prices, wholesale market trading data, retail market trading data, and power balance data;

[0060] (2) Power dispatching platform: power dispatching operation data, power grid safety verification data;

[0061] (3) Electricity sales companies: contract execution and actual electricity consumption of power users;

[0062] (4) Power generation enterprises: actual electricity consumption, power generation capacity, contract execution, financial audit reports, etc.;

[0063] (5) Large electricity users: actual electricity consumption and contract execution.

[0064] As a preferred solution, step S1 includes step S11 to step S12, and each step is specifically as follows:

[0065] Step S11, through the data processing device installed in any market entity, data collection and verification of the power market are carried out in real time to obtain relevant data of the power market.

[0066] In this embodiment, the data processing device refers to a device or instrument with data collection and data verification functions, which is used to realize the real-time collection and online verification of power market operation data. The device is composed of hardware and software, including a data collection system, a data processing system, a data verification system and a data upload system, thereby providing basic data for power market performance risk monitoring and management. The device collects power market operation data related to various market entities and power trading platforms and power dispatching platforms in real time through a local area network. Among them, the hardware can be electric energy meters connected to different entities (used to measure the electric energy usage of users, power sales companies and power generation companies), as well as physical channels for data transmission (used to transmit various types of unprocessed and processed data). The software can be data calling software and data processing software that obtains data from market trading platforms, power dispatching platforms, and annual data platforms of various enterprises (development environment: ① Hardware environment (CPU: Pentium Dual-core 3220 or above, running memory: 2G or above, graphics card: integrated graphics card or above); ② Software environment (64-bit Windows operating system, Java environment, MySQL database environment, Tomcat server environment, Firefox browser / Google browser); ③ Development language (Java, HTML, CSSJS))

[0067] Step S12, pre-processing the relevant data of the power market to obtain the operation data of the power market.

[0068] Among them, preprocessing includes outlier removal and noise elimination.

[0069] In this embodiment, a method for monitoring electricity market performance behavior provided by an embodiment of the present invention also includes a data uploading process, which mainly includes: a process of uploading real-time acquired electricity market operation data to a blockchain network.

[0070] Among them, the blockchain network is a distributed system used to store the real-time data of the above steps and realize mutual verification of the stored data. The real-time monitoring and acquisition of the operation data of the power market are mainly carried out by the verification nodes in the blockchain. The verification nodes supervise the basic data and transaction behaviors of each node in the platform, review the transaction information of the nodes that have completed the transaction, determine whether there is a breach of contract in the transaction and report it to the central node. For data verification, it is mainly based on the data verification technology of power system analysis, analyzes the logical relationship between different variables, verifies the rationality of market operation data, and the data verification technology based on blockchain mainly uses the supervision mechanism of blockchain to verify the authenticity of market operation data.

[0071] It should be noted that the adoption of a multi-node design based on blockchain technology can improve the reliability and security of the blockchain network. These nodes can be provided by power market entities, regulators or third-party nodes. If higher node performance or security is required, dedicated hardware nodes can also be used. This step requires a blockchain network, suitable nodes, smart contracts and related data transmission and processing equipment and channels in the power market. Among them, the power market blockchain can adopt a DAG data structure to enable all information between blocks to be transmitted and maintained synchronously. DAG is a tree structure without a return pointer, which can achieve data deduplication, storage and transmission, greatly improving the storage efficiency of the blockchain; the power market blockchain can also adopt a distributed peer-to-peer network, which is developed on the basis of a P2P network. It sets up distributed ledgers on different nodes to protect data and enable users to safely maintain each block. In the process of power market performance, blockchain technology can be used to record information such as transaction history, capital flow and contract agreements of each participant. This information will be recorded in a public ledger, and all participants can verify and monitor transactions on the public ledger. This transparency can enhance trust between participants and provide regulators with more precise regulatory means.

[0072] Step S2: extract features from the operational data to obtain multiple indicators to be evaluated corresponding to the risk type to be evaluated.

[0073] As a preferred solution, step S2 includes step S21 to step S22, and each step is specifically as follows:

[0074] Step S21, analyzing and processing the operation data to obtain corresponding multiple first indicators.

[0075] Step S22: selecting a first indicator associated with the risk type to be assessed through a pre-constructed risk classification model as the indicator to be evaluated corresponding to the risk type to be assessed.

[0076] As a preferred solution, the process of obtaining the "risk classification model" mentioned in step 22 includes steps S201 to S204, and each step is specifically as follows:

[0077] Step S201, obtaining historical operation data of the power market and a plurality of second indicators corresponding to the historical operation data.

[0078] As an example, if the historical operating data of the electricity market is the annual financial audit report of each power sales company, then the corporate financial related indicators are calculated based on the annual financial audit report of each power sales company, and the calculated related indicators are used as the second indicators.

[0079] Step S202: According to the data type of the historical operation data, a corresponding mining model is selected, and then the mining model is used to perform data mining on the historical operation data to obtain several risk types corresponding to the historical operation data, as well as the correlation coefficient between each second indicator and the risk feature library corresponding to each risk type, and the risk feature library corresponding to the risk type is used as the risk feature library associated with each second indicator.

[0080] It should be noted that the core of data mining is the mining model, and different mining models are applicable to different scenarios and different data. Some data (power supply and demand ratio, market prices, wholesale and retail market transaction data, etc.) are suitable for mining models based on mathematical statistics, such as principal component analysis and factor analysis; some data (performance status of each subject, transaction status, credit status, unit status) are suitable for mining models based on knowledge discovery, such as classification, cluster analysis, and association analysis; more complex data (annual audit reports of power sales companies and power generation companies, market-based transaction details, etc.) are suitable for a combination of various models, such as a classification model based on principal component analysis and cluster analysis.

[0081] In this embodiment, risk types include, but are not limited to, supply and demand risk, market power risk, transaction risk, regulatory risk, and grid security risk.

[0082] Step S203: construct a training data set using the risk feature library associated with each second indicator and the correlation coefficient between each second indicator and each risk feature library.

[0083] Step S204, using the obtained training data set, the pre-built initial classification model is trained and its parameters are optimized to obtain a risk classification model.

[0084] Step S3, please refer to Table 1, use the language item description based on intuitionistic fuzzy sets to describe the importance of each indicator to be evaluated, please refer to Table 2, use the language item description based on intuitionistic fuzzy sets to describe the performance of each indicator to be evaluated, and screen out multiple decision makers associated with each indicator to be evaluated in the expert group.

[0085] In this embodiment, taking the supply and demand risk index library as an example, three experts are invited to form a decision-making group to assess its risks, and the members of the decision-making group are decision makers. Assume that there are only five indicators in the supply and demand risk index library, namely: A. Economic prosperity index; B. GDP growth rate; C. Power supply and demand ratio; D. Coal price index; E. Renewable energy consumption ratio. The importance level of decision makers is described in 5 levels using language items, namely: very important, important, moderately important, unimportant, and very unimportant.

[0086] Table 1 Linguistic description of the importance of decision makers and indicators

[0087] Language Item Description Corresponding intuitionistic fuzzy number Very important (0.9,0.10) important (0.70,0.25) General Important (0.50,0.45) unimportant (0.30,0.65) Very unimportant (0.10,0.90)

[0088] Table 2 Description of language items for decision makers' indicator level classification

[0089] Language Item Description Corresponding intuitionistic fuzzy number Very good (I): Very low risk (1.00,0.00) Good (II): Low risk (0.75,0.15) Medium (III): medium risk (0.50,0.40) Bad (IV): Higher risk (0.25,0.65) Very bad (V): High risk (0.10,0.90)

[0090] Step S4, obtaining the indicator evaluation opinions of each decision maker, and then integrating the indicator evaluation opinions of all decision makers to obtain the performance risk score of the risk type to be evaluated as the performance behavior monitoring result of the power market.

[0091] Among them, the indicator evaluation opinions are obtained by decision makers using language items to describe the importance and performance of each indicator to be evaluated.

[0092] As a preferred solution, step S4 includes step S41 to step S44, and each step is specifically as follows:

[0093] Step S41, traverse each indicator to be evaluated, obtain the indicator level evaluation opinions of each decision maker on the current traversed indicator to be evaluated, and fuse the indicator level evaluation opinions corresponding to each decision maker according to the decision weight of each decision maker, and construct the first intuitive fuzzy decision vector R based on the opinion fusion results of all indicators to be evaluated.

[0094] Among them, the indicator level evaluation opinion is obtained by decision makers using language items to describe the performance of each indicator to be evaluated.

[0095] In this embodiment, taking the supply and demand risk index library as an example, the decision makers are experts 1, expert 2 and expert 3. For the evaluation opinions of each decision maker on the indicator level of each indicator to be evaluated, please refer to Table 3. Taking the indicator A to be evaluated as an example, the language item description of expert 1 (i.e., the indicator level evaluation opinion) is "IV", the language item description of expert 2 is "IV", and the language item description of expert 3 is "V". According to the language item descriptions of each expert, it can be seen from Table 2 that the intuitive fuzzy number corresponding to expert 1 is (0.25, 0.65), the intuitive fuzzy number corresponding to expert 2 is (0.25, 0.65), and the intuitive fuzzy number corresponding to expert 3 is (0.10, 0.90). Next, please refer to formula (1). According to the indicator level evaluation opinions and intuitive fuzzy numbers corresponding to each expert, and the decision weights of each expert, the indicator level evaluation opinions of each decision maker on the indicator A to be evaluated are fused to obtain the opinion fusion result of the indicator A to be evaluated. Then, please refer to formula (2). According to the opinion fusion results r of the indicators A, B, C, D, and E to be evaluated, j , forming the first intuitive fuzzy decision vector R.

[0096]

[0097] Where IFMA represents the IFMA operator, λ l represents the decision weight of the lth member in the decision-making group, and the ordered pair The intuitive fuzzy number corresponding to the evaluation opinion of the kth member of the decision-making group on the indicator level of the jth indicator to be evaluated, r j =(μ(x j ),v(x j ),π(x j ))(j=1,2,…,n).

[0098]

[0099] Table 3 Evaluation opinions of decision makers on the indicator level of each indicator to be evaluated

[0100]

[0101] Step S42, traverse each indicator to be evaluated, obtain the indicator importance evaluation opinions of each decision maker on the current traversed indicator to be evaluated, and fuse the indicator importance evaluation opinions corresponding to each decision maker according to the decision weight of each decision maker, and construct a weight vector W based on the opinion fusion results of all indicators to be evaluated.

[0102] Among them, the indicator importance evaluation opinion is obtained by decision makers using language items to describe the performance of each indicator to be evaluated.

[0103] In this embodiment, taking the supply and demand risk index library as an example, the decision makers are experts 1, 2 and 3. For the evaluation opinions of each decision maker on the importance of each indicator to be evaluated, please refer to Table 4. Taking the indicator A to be evaluated as an example, the language item description of expert 1 (i.e., the evaluation opinion on the importance of the indicator) is "generally important", the language item description of expert 2 is "important", and the language item description of expert 3 is "important". According to the language item descriptions of each expert, it can be seen from Table 1 that the intuitive fuzzy number corresponding to expert 1 is (0.50, 0.45), the intuitive fuzzy number corresponding to expert 2 is (0.70, 0.25), and the intuitive fuzzy number corresponding to expert 3 is (0.70, 0.25). Next, please refer to formula (3). According to the corresponding indicator importance evaluation opinions and intuitive fuzzy numbers of each expert and the decision weights of each expert, the evaluation opinions of each decision maker on the importance of the indicator A to be evaluated are fused to obtain the opinion fusion result of the indicator A to be evaluated. Next, please refer to formula (4). According to the opinion fusion results of the indicators A, B, C, D, and E to be evaluated, a weight vector W is formed.

[0104]

[0105] Where IFMA represents the IFMA operator, λ l represents the decision weight of the lth member in the decision-making group, and the ordered pair represents the intuitive fuzzy number corresponding to the k-th member of the decision-making group's evaluation opinion on the j-th indicator to be evaluated, w j =(μ wj ,v wj ,π wj )(j=1,2,…,n).

[0106]

[0107] Table 4 Evaluation opinions of decision makers on the importance of each indicator to be evaluated

[0108]

[0109] As a preferred solution, the process of obtaining the “decision weight of the decision maker” mentioned in step S41 and step S42 includes step S401 to step S402, and each step is specifically as follows:

[0110] Step S401, using the language term description based on the intuitionistic fuzzy set to describe the importance of the evaluation opinions of each decision maker, and obtaining the importance level and intuitionistic fuzzy number corresponding to each decision maker.

[0111] The intuitionistic fuzzy number corresponding to the decision maker includes the membership degree and non-membership degree corresponding to the decision maker.

[0112] Step S402: Calculate the decision weight λ of each decision maker according to the importance level and intuitive fuzzy number corresponding to each decision maker according to the preset decision weight algorithm. k .

[0113] In this embodiment, taking the supply and demand risk index library as an example, the decision makers are experts 1, 2 and 3. The language item description (i.e., importance level) of expert 1 is "generally important", the language item description of expert 2 is "very important", and the language item description of expert 3 is "important". According to the language item descriptions of each expert, it can be seen from Table 1 that the intuitive fuzzy number corresponding to expert 1 is (0.50, 0.45), the intuitive fuzzy number corresponding to expert 2 is (0.9, 0.10), and the intuitive fuzzy number corresponding to expert 3 is (0.70, 0.25). Next, please refer to formula (5), according to the importance level and intuitive fuzzy number corresponding to each expert, the decision weight λ of each expert is calculated. k , the decision weight λ1 of expert 1 is 0.243, the decision weight λ2 of expert 2 is 0.416, and the decision weight λ3 of expert 3 is 0.341.

[0114]

[0115] In the formula, l represents the number of members of the decision-making group, and the ordered pair (μ k ,v k ) represents the intuitionistic fuzzy number of the importance of the kth member in the decision-making group, μ k Indicates the membership degree of element x in non-empty set X to X, v k Indicates the non-membership degree of element x in a non-empty set X, π k represents the hesitation of an element x in a non-empty set X to belong to X. Among them, π k =1-μ k -v k ,

[0116] Step S43, please refer to equations (6), (7), (8), and (9), use the weight vector W to weight the first intuitive fuzzy decision vector R to obtain the second intuitive fuzzy decision vector R'.

[0117]

[0118] μ w (x j )=μ(x j )·μ wj (7)

[0119] v w (x j )=v(x j )+vwj -v(x j )·v wj (8)

[0120] π w (x j )=1-v(x j )-v wj -μ(x j )·μ wj +v(x j )·v wj (9)

[0121] In the formula, r′ j =(μ′ j ,v′ j ,π′ j )=(μ w (x j ),v w (x j ),π w (x j )).

[0122] Step S44, according to the preset comprehensive scoring algorithm, based on the second intuitive fuzzy decision vector R', calculate the performance risk score of the risk type to be evaluated as the performance behavior monitoring result of the power market.

[0123] In this embodiment, step S44 includes step S441 to step S442, and each step is specifically as follows:

[0124] Step S441, according to r′ in the second intuitive fuzzy decision vector R′ j =(μ w (x j ),v w (x j ),π w (x j )), determine the intuitive fuzzy number α of the jth indicator to be evaluated in the second intuitive fuzzy decision vector R' j =(μ w (x j ),v w (x j )), then refer to formula (10) to calculate the score function S(α) of each indicator to be evaluated in the second intuitive fuzzy decision vector R'.

[0125] S(α)=μ A (x)-v A (x) (10)

[0126] In the formula, for the intuitionistic fuzzy number α=(μ A (x),vA (x)), whose accuracy function is S(α).

[0127] Step S442, performing weighted summation on the score function of each indicator to be evaluated and the membership function in the weight vector to obtain a comprehensive score as the performance risk score of the risk type to be evaluated.

[0128] In this embodiment, taking the supply and demand risk index library as an example, the comprehensive score calculated according to the above steps is 0.252, indicating that its risk level is between medium and high, and risk prevention needs to be strengthened.

[0129] It should be noted that, in addition to analyzing the performance risk score of the risk type to be evaluated, please refer to formula (11) to calculate the accuracy function of each indicator to be evaluated to form an accuracy function vector P. Among them, the accuracy function can be used to analyze the acceptability of the indicator to be evaluated.

[0130]

[0131] In the formula, for the intuitionistic fuzzy number α=(μ A (x),v A (x)), whose accuracy function is E(α), π A (x) = 1-μ A (x)-v A (x).

[0132] In this embodiment, taking the supply and demand risk index library as an example, the precision function vector P is calculated according to the above steps. It can be seen that the precision functions of various indicators to be evaluated are very small, that is, the accuracy of the final calculation result is credible and the result is acceptable.

[0133] Please refer to Figure 2 , is a schematic diagram of the structure of a power market performance monitoring system provided by an embodiment of the present invention, the system includes a data acquisition module M1, a feature extraction module M2, a description screening module M3 and a behavior monitoring module M4, and each module is specifically as follows:

[0134] The data acquisition module M1 is used to obtain the operation data of the power market in real time; wherein the power market includes several market entities, market trading platforms and power dispatching platforms;

[0135] The feature extraction module M2 is used to extract features from the operation data to obtain multiple indicators to be evaluated corresponding to the risk type to be evaluated;

[0136] The description and screening module M3 is used to describe the performance and importance of each indicator to be evaluated using a linguistic term description based on an intuitionistic fuzzy set, and screen out multiple decision makers associated with each indicator to be evaluated from the expert group;

[0137] The behavior monitoring module M4 is used to obtain the indicator evaluation opinions of each decision maker, and then integrate the indicator evaluation opinions of all decision makers to obtain the performance risk score of the risk type to be evaluated as the performance behavior monitoring result of the power market; among which, the indicator evaluation opinion is obtained by the decision maker using language items to describe the importance and performance of each indicator to be evaluated.

[0138] As a preferred solution, the behavior monitoring module M4 specifically includes a data acquisition unit 11 and a preprocessing unit 12, and the specific details of each unit are as follows:

[0139] The first fusion unit 41 is used to traverse each indicator to be evaluated, obtain the indicator level evaluation opinions of each decision maker on the currently traversed indicator to be evaluated, and fuse the indicator level evaluation opinions corresponding to each decision maker according to the decision weight of each decision maker, and construct a first intuitive fuzzy decision vector based on the opinion fusion results of all indicators to be evaluated; wherein the indicator level evaluation opinions are obtained by the decision maker using language items to describe the evaluation of the performance of each indicator to be evaluated;

[0140] The second fusion unit 42 is used to traverse each indicator to be evaluated, obtain the indicator importance evaluation opinions of each decision maker on the currently traversed indicator to be evaluated, and fuse the indicator importance evaluation opinions corresponding to each decision maker according to the decision weight of each decision maker, and construct a weight vector according to the fusion results of the opinions of all indicators to be evaluated; wherein the indicator importance evaluation opinions are obtained by the decision maker using language items to describe the evaluation of the performance of each indicator to be evaluated;

[0141] A weighting unit 43, configured to weight the first intuitive fuzzy decision vector using a weight vector to obtain a second intuitive fuzzy decision vector;

[0142] The calculation unit 44 is used to calculate the performance risk score of the risk type to be evaluated according to the preset comprehensive scoring algorithm and the second intuitive fuzzy decision vector as the performance behavior monitoring result of the power market.

[0143] As a preferred option, please refer to Figure 2 , a power market performance monitoring system provided by an embodiment of the present invention further includes a decision weight analysis module M5, which is specifically as follows:

[0144] The decision weight analysis module M5 is used to describe the importance of the evaluation opinions of each decision maker using a linguistic term description based on an intuitionistic fuzzy set, and obtain the importance level and intuitionistic fuzzy number corresponding to each decision maker; according to a preset decision weight algorithm, based on the importance level and intuitionistic fuzzy number corresponding to each decision maker, the decision weight of each decision maker is calculated; wherein the intuitionistic fuzzy number corresponding to the decision maker includes the membership and non-membership corresponding to the decision maker.

[0145] As a preferred solution, the feature extraction module M2 specifically includes an analysis unit 21 and a selection unit 22, and the details of each unit are as follows:

[0146] An analysis unit 21 is used to analyze and process the operation data to obtain corresponding multiple first indicators;

[0147] The selection unit 22 is used to select a first indicator associated with the risk type to be evaluated through a pre-built risk classification model as the indicator to be evaluated corresponding to the risk type to be evaluated.

[0148] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0149] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0150] The present invention provides a method and system for monitoring the performance of power market. By acquiring the operation data of the power market in real time, the relevant information of market entities, market trading platforms and power dispatching platforms can be timely understood, including market participants, trading behaviors, market prices and power supply and demand, etc., to provide real-time and effective data support for the operation monitoring of the power market and the risk assessment of performance behavior. Then, feature extraction is performed on the operation data to obtain multiple indicators to be evaluated corresponding to the risk type to be evaluated. Then, the performance and importance of each indicator to be evaluated are described using the language item description based on the intuitionistic fuzzy set, so that the subjective evaluation can be expressed in fuzzy language, so that the indicator evaluation opinions can be closer to the subjective perception of the decision maker, which is convenient for subsequent decision analysis and evaluation. In addition, the evaluation opinions of each decision maker on the evaluation indicators are collected and integrated, and the views of different decision makers are comprehensively considered to cope with the increasingly complex market transaction situation, and the performance risk score of the risk type to be evaluated is obtained. As an indicator for measuring the performance risk of the power market, it can help market regulatory agencies or enterprise management to monitor the overall risk status of the market so as to take corresponding management measures and risk response strategies in a timely manner.

[0151] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for monitoring contract performance in the power market, characterized in that: include: Acquire the operational data of the power market in real time; wherein the power market includes a number of market players, a market trading platform and a power dispatching platform; Extracting features from the operational data to obtain a plurality of indicators to be evaluated corresponding to the risk type to be evaluated; Using language item descriptions based on intuitionistic fuzzy sets, the performance and importance of each of the indicators to be evaluated are described, and multiple decision makers associated with each of the indicators to be evaluated are screened out from the expert group; Obtain the indicator evaluation opinions of each decision maker, and then integrate the indicator evaluation opinions of all the decision makers to obtain the performance risk score of the risk type to be evaluated as the performance behavior monitoring result of the power market; wherein the indicator evaluation opinions are obtained by the decision makers using language items to describe the importance and performance of each indicator to be evaluated; The feature extraction of the operational data to obtain multiple indicators to be evaluated for the risk type to be evaluated is specifically: Analyze and process the operation data to obtain corresponding multiple first indicators; By using a pre-constructed risk classification model, the first indicator associated with the risk type to be assessed is selected as the indicator to be evaluated corresponding to the risk type to be assessed; wherein the construction of the risk classification model is specifically as follows: Acquire historical operation data of the power market and a plurality of second indicators corresponding to the historical operation data; According to the data type of the historical operation data, a corresponding mining model is selected, and then the historical operation data is mined using the mining model to obtain several risk types corresponding to the historical operation data, and a correlation coefficient between each of the second indicators and a risk feature library corresponding to each of the risk types, and the risk feature library corresponding to the risk type is used as a risk feature library associated with each of the second indicators; A training data set is formed by using a risk feature library associated with each of the second indicators and a correlation coefficient between each of the second indicators and each of the risk feature libraries; The obtained training data set is used to train and optimize the parameters of the pre-built initial classification model to obtain the risk classification model.

2. A method for monitoring electricity market performance according to claim 1, characterized in that: The index evaluation opinions of each decision maker are obtained, and then the index evaluation opinions of all decision makers are integrated to obtain the performance risk score of the risk type to be evaluated as the performance behavior monitoring result of the power market, specifically: Traversing each indicator to be evaluated, obtaining the indicator level evaluation opinions of each decision maker on the currently traversed indicator to be evaluated, and fusing the indicator level evaluation opinions corresponding to each decision maker according to the decision weight of each decision maker, and constructing a first intuitive fuzzy decision vector based on the opinion fusion results of all the indicators to be evaluated; wherein the indicator level evaluation opinions are obtained by the decision maker using language items to describe the evaluation of the performance of each indicator to be evaluated; Traverse each indicator to be evaluated, obtain the indicator importance evaluation opinions of each decision maker on the currently traversed indicator to be evaluated, and fuse the indicator importance evaluation opinions corresponding to each decision maker according to the decision weight of each decision maker, and construct a weight vector based on the opinion fusion results of all the indicators to be evaluated; wherein the indicator importance evaluation opinions are obtained by the decision maker using language items to describe the evaluation of the performance of each indicator to be evaluated; Using the weight vector, weighting the first intuitive fuzzy decision vector to obtain a second intuitive fuzzy decision vector; According to the preset comprehensive scoring algorithm, based on the second intuitive fuzzy decision vector, the performance risk score of the risk type to be evaluated is calculated as the performance behavior monitoring result of the electricity market.

3. A method for monitoring electricity market performance according to claim 2, characterized in that: The acquisition of the decision weight of the decision maker is specifically as follows: Using the language term description based on the intuitionistic fuzzy set, the importance of the evaluation opinions of each decision maker is described, and the importance level and intuitionistic fuzzy number corresponding to each decision maker are obtained; According to the preset decision weight algorithm, the decision weight of each decision maker is calculated according to the importance level and intuitive fuzzy number corresponding to each decision maker; The intuitionistic fuzzy number corresponding to the decision maker includes the membership degree and non-membership degree corresponding to the decision maker.

4. A method for monitoring electricity market performance according to claim 1, characterized in that: The real-time acquisition of the operation data of the power market is specifically: By installing a data processing device on any of the market entities, data collection and verification of the power market are performed in real time to obtain relevant data of the power market; Preprocessing the relevant data of the power market to obtain the operation data of the power market; Among them, preprocessing includes outlier removal and noise elimination.

5. A power market performance monitoring system, characterized in that: include: A data acquisition module is used to acquire the operation data of the power market in real time; wherein the power market includes a number of market entities, a market trading platform and a power dispatching platform; A feature extraction module is used to extract features from the operation data to obtain a plurality of indicators to be evaluated corresponding to the risk type to be evaluated; A description screening module is used to describe the performance and importance of each of the indicators to be evaluated using a linguistic term description based on an intuitionistic fuzzy set, and screen out a plurality of decision makers associated with each of the indicators to be evaluated from the expert group; A behavior monitoring module is used to obtain the indicator evaluation opinions of each decision maker, and then integrate the indicator evaluation opinions of all the decision makers to obtain the performance risk score of the risk type to be evaluated as the performance behavior monitoring result of the power market; wherein the indicator evaluation opinion is obtained by the decision maker using language items to describe the importance and performance of each indicator to be evaluated; Wherein, the feature extraction module specifically includes: An analysis unit, configured to analyze and process the operation data to obtain a corresponding plurality of first indicators; A selection unit is used to select the first indicator associated with the risk type to be evaluated as the indicator to be evaluated corresponding to the risk type to be evaluated through a pre-constructed risk classification model; wherein the construction of the risk classification model is specifically as follows: obtaining the historical operation data of the power market and a plurality of second indicators corresponding to the historical operation data; selecting a corresponding mining model according to the data type of the historical operation data, and then using the mining model to perform data mining on the historical operation data to obtain several risk types corresponding to the historical operation data and a correlation coefficient between each of the second indicators and a risk feature library corresponding to each of the risk types, and using the risk feature library corresponding to the risk type as the risk feature library associated with each of the second indicators; using the risk feature library associated with each of the second indicators and the correlation coefficient between each of the second indicators and each of the risk feature libraries to form a training data set; using the obtained training data set to train and optimize the parameters of the pre-constructed initial classification model to obtain the risk classification model.

6. A power market contract performance monitoring system according to claim 5, characterized in that: The behavior monitoring module specifically includes: The first fusion unit is used to traverse each indicator to be evaluated, obtain the indicator level evaluation opinions of each decision maker on the currently traversed indicator to be evaluated, and fuse the indicator level evaluation opinions corresponding to each decision maker according to the decision weight of each decision maker, and construct a first intuitive fuzzy decision vector based on the opinion fusion results of all the indicators to be evaluated; wherein the indicator level evaluation opinions are obtained by the decision maker using language items to describe the evaluation of the performance of each indicator to be evaluated; The second fusion unit is used to traverse each indicator to be evaluated, obtain the indicator importance evaluation opinions of each decision maker on the currently traversed indicator to be evaluated, and fuse the indicator importance evaluation opinions corresponding to each decision maker according to the decision weight of each decision maker, and construct a weight vector according to the opinion fusion results of all the indicators to be evaluated; wherein the indicator importance evaluation opinions are obtained by the decision maker using language items to describe the evaluation of the performance of each indicator to be evaluated; A weighting unit, configured to weight the first intuitive fuzzy decision vector using the weight vector to obtain a second intuitive fuzzy decision vector; A calculation unit is used to calculate the performance risk score of the risk type to be evaluated according to a preset comprehensive scoring algorithm and the second intuitive fuzzy decision vector as a performance behavior monitoring result of the power market.

7. A power market performance monitoring system as claimed in claim 6, characterized in that: Also includes: The decision weight analysis module is used to describe the importance of the evaluation opinions of each decision maker using a language term description based on an intuitionistic fuzzy set, and obtain the importance level and intuitionistic fuzzy number corresponding to each decision maker; according to a preset decision weight algorithm, based on the importance level and intuitionistic fuzzy number corresponding to each decision maker, the decision weight of each decision maker is calculated; wherein the intuitionistic fuzzy number corresponding to the decision maker includes the membership degree and non-membership degree corresponding to the decision maker.

Citation Information

Patent Citations

  • Market risk assessment method based on power transaction data

    CN111950918A

  • Quantitative evaluation method and device for expressway operation safety index and server

    CN115796606A