Power market risk early warning method and system based on multi-modal fusion
By performing curve fitting and stability coefficient calculation on the historical electricity prices and power generation data of power generation enterprises and integrating them into power supply characteristic data, the problem of inaccurate power market risk warning in existing technologies is solved, and dynamic risk identification and graded warning of the power market are achieved.
Patent Information
- Application Number
- CN202510762687.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-09
- Publication Date
- 2025-09-16
AI Technical Summary
Existing technologies make it difficult to conduct in-depth analysis of key data such as electricity prices and power generation of power generation companies, resulting in low accuracy in early warning of power market risks.
By obtaining the historical electricity prices and power generation data of multiple power generation companies, curve fitting is performed to calculate the stability coefficient, which is integrated into power supply characteristic data. The transaction risk coefficient is calculated based on the power supply characteristic data, and the companies are ranked and risk warnings are issued.
It has achieved dynamic identification and graded early warning of power market risks, improved the accuracy and foresight of risk warnings, and can accurately locate high-risk enterprises, assist regulatory authorities in timely intervention, and reduce the negative impact of market fluctuations on grid stability and economy.
Smart Images
Figure CN120655098A_ABST
Abstract
Claims
1. A power market risk warning method based on multimodal fusion, characterized in that: Obtain the historical electricity price data and historical power generation data of multiple power generation enterprises in the target area; Analyze the historical electricity price data and historical power generation data of each power generation enterprise to obtain the corresponding power supply characteristic data; Calculate the transaction risk coefficient of each power generation enterprise based on the power supply characteristic data; Rank multiple power generation enterprises in the target area based on the transaction risk coefficient to obtain a power supply risk warning sequence; Conduct risk warning on the power market based on the power supply risk warning sequence.
2. The power market risk early warning method based on multimodal fusion according to claim 1 is characterized in that: The analysis of the historical electricity price data and historical power generation data of each power generation enterprise includes: Extract the historical electricity price data and historical power generation data of several power generation enterprises in the target area; among them, the historical electricity price data and historical power generation data refer to the electricity price data and power generation data of power generation enterprise j in several consecutive periods, t is time, and t1 < t < t2, t1 is the start time of several consecutive periods, t2 is the end time of several consecutive periods; i = 1, 2,..., n, n is the total number of power generation enterprises in the target area; Perform curve fitting on the historical electricity price data of each power generation enterprise respectively to obtain the corresponding electricity price change curve; perform curve fitting on the historical power generation data of each power generation enterprise respectively to obtain the corresponding power generation change curve; Calculate the root mean square error between the electricity price change curve and the electricity price standard curve, and mark it as the electricity price stability coefficient; among them, the electricity price standard curve is drawn from the average electricity price of several power generation enterprises in the target area; Calculate the root mean square error between the power generation change curve and the power generation standard curve, and mark it as the power generation stability coefficient; among them, the power generation standard curve is drawn from the average power generation of several power generation enterprises in the target area; Integrate the electricity price stability coefficient and the power generation stability coefficient into the power supply characteristic data.
3. The power market risk early warning method based on multimodal fusion according to claim 2 is characterized in that: The calculation of the root mean square error between the electricity price change curve and the electricity price standard curve includes: Obtain the electricity price change curve and electricity price standard curve of each power generation enterprise; through the root mean square error formula Calculate the electricity price stability coefficient RDi of power generation enterprise i; where Di(t) is the electricity price change curve of power generation enterprise i, and P(t) is the electricity price standard curve in the target area.
4. The power market risk early warning method based on multimodal fusion according to claim 2 is characterized in that: The calculation of the root mean square error between the power generation change curve and the power generation standard curve includes: Obtain the power generation change curve and power generation standard curve of each power generation enterprise; through the root mean square error formula Calculate the power generation stability coefficient RFi of power generation company j; where Ui(t) is the power generation change curve of power generation company i, and Q(t) is the power generation standard curve in the target area.
5. The power market risk early warning method based on multimodal fusion according to claim 2 is characterized in that: The calculation of the transaction risk coefficient of each power generation enterprise based on the power supply characteristic data includes: Extract the power supply characteristic data of each power generation enterprise; calculate through the linear mapping relationship between the influence factor, electricity price stability coefficient and power generation stability coefficient of power generation enterprise i to obtain the corresponding transaction risk coefficient JFXi; among them, the influence factor is obtained through the power generation composition information.
6. The power market risk early warning method based on multimodal fusion according to claim 5 is characterized in that: The influence factor is obtained through the power generation composition information, including: Obtain the power generation composition information of several power generation enterprises in the target area; among them, the power generation composition information is the proportion of several power generation types in the power generation enterprise; the power generation types include thermal power, hydropower, wind power and photovoltaic power generation; Calculate the sum of the products of the proportions of several power generation types of each power generation enterprise and the corresponding weight coefficients respectively to obtain the corresponding influence factor.
7. The power market risk early warning method based on multimodal fusion according to claim 5 is characterized in that: The influence factor is obtained through the power generation composition information, including: The power generation structure information of several power generation enterprises in the target area is obtained and integrated into several groups of training data and test data; the artificial intelligence model is trained using the training data; the trained artificial intelligence model is tested using the test data, and the artificial intelligence model is adjusted according to the test results; and finally a power generation stability analysis model is obtained with the power generation structure information of the power generation enterprises as input and the influencing factors of the corresponding power generation enterprises as output; wherein the artificial intelligence model includes a BP neural network model or an RBF neural network model.
8. The power market risk early warning method based on multimodal fusion according to claim 1 is characterized in that: The ranking of multiple power generation enterprises in the target area based on transaction risk coefficients includes: Extract transaction risk coefficients of multiple power generation enterprises in the target area; Determine whether the transaction risk coefficient is greater than the preset risk threshold; if yes, add the corresponding power generation enterprise number to the first warning sequence; if no, add the corresponding power generation enterprise number to the second warning sequence; The first warning sequence is sorted in descending order according to the transaction risk coefficient to obtain a power supply risk warning sequence; wherein the power supply risk warning sequence includes the numbers of multiple power generation enterprises.
9. The power market risk early warning method based on multimodal fusion according to claim 1 is characterized in that: The risk warning for the power market based on the power supply risk warning sequence includes: Extract the power supply risk warning sequence; extract the numbers of the power generation enterprises with the largest transaction risk coefficients from the power supply risk warning sequence in turn, and generate warning information for the corresponding power generation enterprises.
10. A power market risk early warning system based on multimodal fusion, used to implement the power market risk early warning method based on multimodal fusion according to any one of claims 1 to 9, characterized in that: include: Data processing module, and the connected data acquisition module and risk warning module; The data acquisition module is used to obtain historical electricity price data and historical power generation data of multiple power generation enterprises in the target area; The data processing module is used to analyze the historical electricity price data and historical power generation data of each power generation enterprise to obtain corresponding power supply characteristic data; Calculate the transaction risk coefficient of each power generation enterprise based on power supply characteristic data; The risk warning module is used to sort multiple power generation enterprises in the target area based on the transaction risk coefficient to obtain a power supply risk warning sequence; Provide risk warning for the power market based on the power supply risk warning sequence.