Green electricity market risk assessment and prediction method

Abstract

The invention provides a green electricity market risk assessment and prediction method, and relates to the technical field of risk assessment, and the method comprises the steps: carrying out the real-time monitoring of a green electricity market, obtaining electricity supply and demand, transaction and environmental protection monitoring data, building a risk assessment and prediction model, carrying out the risk assessment and prediction of the green electricity market, and obtaining an electricity market risk coefficient; and judging whether the risk coefficient is greater than or equal to a power market risk threshold, and if the risk coefficient is greater than or equal to the power market risk threshold, generating a risk early warning signal, thereby solving the problem that a green power market risk assessment method cannot reflect the latest change of market data in real time, resulting in incomplete risk assessment results and low timeliness. The effects of effectively improving the accuracy and response speed of green electricity market risk assessment and providing timely risk early warning and decision support for market participants are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of risk assessment, and in particular to a green electricity market risk assessment and prediction method. Background Art

[0002] Green power markets refer to the trading and supply of electricity generated from renewable energy sources such as wind and solar energy. Such markets have developed rapidly around the world and are supported by policies and consumer preferences. The expansion of green power markets not only helps reduce carbon emissions, but also promotes the diversification of energy supply. At present, risk assessment of green power markets usually relies on statistical methods and empirical judgments, which often cannot fully capture market dynamics and environmental changes. In addition, existing assessment tools often lack real-time performance and cannot respond to factors such as changes in market supply and demand, price fluctuations, and policy adjustments in a timely manner.

[0003] In summary, existing green electricity market risk assessment methods often fail to reflect the latest changes in market data in real time, and lack the ability to comprehensively consider electricity supply and demand, transactions, and environmental protection data, resulting in technical problems such as incomplete and low timeliness of risk assessment results. Summary of the invention

[0004] The present application provides a green electricity market risk assessment and prediction method, which is used to solve the technical problems that the existing green electricity market risk assessment methods cannot reflect the latest changes in market data in real time, and lack the ability to comprehensively consider electricity supply and demand, transactions and environmental protection data, resulting in incomplete and low timeliness of risk assessment results.

[0005] The present application provides a green power market risk assessment and prediction method, the method comprising:

[0006] Conduct real-time monitoring on the green electricity market to obtain electricity supply and demand monitoring data, electricity transaction monitoring data and electricity environmental monitoring data; build an electricity market risk assessment and prediction model; based on the electricity supply and demand monitoring data, the electricity transaction monitoring data and the electricity environmental monitoring data, conduct risk assessment and prediction on the green electricity market according to the electricity market risk assessment and prediction model to obtain an electricity market risk coefficient; determine whether the electricity market risk coefficient is greater than or equal to an electricity market risk threshold; if the electricity market risk coefficient is greater than or equal to the electricity market risk threshold, generate an electricity market risk warning signal.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The present application provides a green electricity market risk assessment and prediction method, which performs real-time monitoring on the green electricity market to obtain electricity supply and demand monitoring data, electricity transaction monitoring data and electricity environmental protection monitoring data; builds an electricity market risk assessment and prediction model; based on the electricity supply and demand monitoring data, the electricity transaction monitoring data and the electricity environmental protection monitoring data, performs risk assessment and prediction on the green electricity market according to the electricity market risk assessment and prediction model to obtain an electricity market risk coefficient; determines whether the electricity market risk coefficient is greater than or equal to the electricity market risk threshold; if the electricity market risk coefficient is greater than or equal to the electricity market risk threshold, generates an electricity market risk warning signal, which solves the technical problems that the existing green electricity market risk assessment methods cannot reflect the latest changes in market data in real time, and lack the ability to comprehensively consider electricity supply and demand, transaction and environmental protection data, resulting in incomplete and low timeliness of risk assessment results. By integrating electricity supply and demand, transaction and environmental protection data in real time and using deep learning models, the accuracy and response speed of green electricity market risk assessment are effectively improved, providing timely risk warnings and decision-making support for market participants. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A flow chart of a green electricity market risk assessment and prediction method is provided for this application.

[0010] Figure 2 A schematic diagram of a process for generating a power market risk warning signal in a green power market risk assessment and prediction method is provided for this application. DETAILED DESCRIPTION

[0011] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0012] Examples, such as Figure 1 As shown, the present application provides a green electricity market risk assessment and prediction method, the method comprising:

[0013] Carry out real-time monitoring based on the green electricity market to obtain electricity supply and demand monitoring data, electricity transaction monitoring data and electricity environmental protection monitoring data.

[0014] Optionally, the real-time monitoring system has high-frequency data collection capabilities and can continuously collect and update dynamic data in the market. Specifically, power supply and demand monitoring data include but are not limited to power generation, power consumption, and peak and valley electricity price information, which are essential for evaluating the supply and demand balance of the market. Power trading monitoring data covers trading prices, trading volumes, and market participant behaviors in the power market, which help analyze the health and degree of competition of market economic activities. In addition, power environmental monitoring data focuses on environmental impact indicators in the process of power production and consumption, such as carbon dioxide emissions and emission levels of other pollutants, which are of great significance for evaluating the environmental performance of the green power market. During this monitoring process, real-time data is continuously collected through advanced sensors and monitoring equipment, which are deployed at key power supply and consumption nodes to ensure the comprehensiveness and accuracy of the data. For example, in power supply and demand monitoring, smart meters can be used to record the power consumption of each consumption point in real time, while in power trading monitoring, trading data can be tracked in real time through the trading system of the power market. Environmental monitoring may need to cooperate with environmental monitoring sites to track various environmental indicators in real time. Through these real-time monitoring data, a dynamic, multi-dimensional electricity market data set can be constructed, which not only supports daily market operation analysis, but also provides basic data for risk assessment. This continuous data flow ensures that the monitoring system can capture instant changes in the market, providing a real-time and accurate information basis for subsequent risk assessment and decision support.

[0015] Build a risk assessment and prediction model for the electricity market.

[0016] Specifically, the power market risk assessment and prediction model is built using neural network technology, with the aim of integrating and analyzing power supply and demand data, power trading data, and power environmental protection data obtained through the real-time monitoring system to predict market risks. First, three independent sub-models need to be constructed, corresponding to the assessment of power supply and demand risk, power trading risk, and power environmental protection risk. These models are based on the neural network deep learning framework and can automatically learn and extract complex features and patterns in the data, thereby improving the accuracy and efficiency of predictions. In the process of model construction, data preprocessing is first performed, including data standardization and normalization processing to meet the input requirements of the neural network. For example, when processing power supply and demand monitoring data, the data is converted into a standard format to ensure that data from different data sources can be effectively processed by the model. Subsequently, appropriate network structures and parameters are selected for each sub-model, such as the number of layers, the number of neurons, the activation function, etc. The selection of these parameters is based on the characteristics of the data and the model verification experiments conducted in advance. Next, each sub-model is trained using historical data. Through supervised learning methods, the model can learn the mapping relationship between input data and risk prediction results. At this stage, the back propagation algorithm is used to optimize the network weights to ensure that the model can accurately predict the risk level. During the training process, cross-validation of the model is also required to test the generalization ability and predictive stability of the model. After completing the training of the individual models, the next step is model integration. Through an integrated learning technique, such as weighted averaging or model fusion technology, the outputs of the three independent models are combined to form a comprehensive power market risk assessment model. The key to this step is to determine the weights of the outputs of each sub-model, which reflect the importance of different risk factors in the overall market risk assessment. Ultimately, the completed comprehensive assessment model can use real-time updated data inputs to quickly generate risk assessment results for the power market, providing a scientific basis for market operations and policy making. Through the above steps, a dynamic, real-time and comprehensive assessment of power market risks can be achieved, significantly improving the timeliness and accuracy of decision-making.

[0017] Based on the power supply and demand monitoring data, the power transaction monitoring data and the power environmental monitoring data, the green power market is subjected to risk assessment and prediction according to the power market risk assessment and prediction model to obtain a power market risk coefficient.

[0018] Exemplarily, the established power market risk assessment and prediction model is used to conduct risk assessment and prediction of the green power market based on the collected power supply and demand monitoring data, power transaction monitoring data and power environmental monitoring data, so as to obtain the power market risk coefficient. This process first involves the integration and input of data, among which the power supply and demand monitoring data provides a real-time view of the market supply and demand balance, the power transaction monitoring data reflects the market transaction dynamics and price fluctuations, and the power environmental monitoring data provides important information about the impact of market operations on the environment. Specifically, the real-time monitoring data is fed into a pre-trained risk assessment model, which has learned how to extract risk signals from these three types of data based on historical data, and then identify potential risk factors and quantify them. For example, if the power supply and demand data show signs of insufficient supply, the model will use this as a risk indicator of market instability; if the transaction data shows abnormal price fluctuations, the model will interpret this as a risk of market manipulation or increased volatility; if the environmental data indicates a sudden increase in emission levels, the model may predict that this will lead to an increased risk of policy intervention. In the risk assessment model, after these data are processed by the neural network, each type of data will generate a predicted risk value. These risk values ​​are then taken into account, and a total power market risk coefficient is calculated by weighted average based on the relevance and importance of various risks. This risk coefficient is a quantitative value that reflects the current risk level of the entire market and can be used to guide market participants and policymakers in making decisions. For example, if the power market risk coefficient exceeds the preset threshold, this indicates that the market is in a high-risk state and market operators or regulators need to take measures, such as increasing power supply, adjusting power prices, or strengthening environmental protection controls, to reduce market risks. The real-time calculation and update of this risk coefficient ensures that market regulators and participants can respond quickly to market changes, effectively manage risks, and maintain market stability and sustainable development. In this way, a dynamic and real-time assessment of green power market risks is achieved, which improves the efficiency and accuracy of risk management.

[0019] Determine whether the electricity market risk coefficient is greater than or equal to the electricity market risk threshold.

[0020] If the electricity market risk coefficient is greater than or equal to the electricity market risk threshold, an electricity market risk warning signal is generated.

[0021] Specifically, it is determined whether the power market risk coefficient is greater than or equal to the preset power market risk threshold, and based on this judgment, a corresponding power market risk warning signal is generated. The main purpose of this stage is to use the risk coefficient obtained previously to determine whether the market is currently in a high-risk state, and to activate the early warning mechanism accordingly to attract the attention of market participants and regulators and take necessary preventive or mitigation measures. First, the power market risk threshold is set through detailed market analysis and historical data evaluation, which represents the highest risk level that the market can bear. This threshold is determined based on historical risk events, market volatility, and other key factors that may have a significant impact on market stability. In specific implementation, the set risk threshold will be encoded into the risk assessment system for comparison with the market risk coefficient calculated in real time. When the power market risk assessment model outputs a new risk coefficient, the system automatically compares this coefficient with the preset risk threshold. If the risk coefficient is greater than or equal to the risk threshold, this indicates that the current risk state of the market has reached or exceeded the level that the market can safely manage. In this case, the system will automatically trigger the risk warning mechanism. The generated power market risk warning signal is an automated response, which includes sending alarms to relevant regulators, market operators, and other relevant parties who may need to take action. The content of the early warning signal may include the specific value of the risk factor, recommended response measures, emergency contact information, and any immediate procedures that need to be executed. For example, if the risk factor indicates that the power supply is tight, the early warning signal may include suggestions for increasing power generation, starting backup energy supply, or implementing power demand side management measures. In this way, ensuring that market risks are identified and responded to in a timely manner greatly improves the operational safety and responsiveness of the power market, thereby maintaining the stability of the market and the continuous power supply capacity. In addition, real-time risk monitoring and early warning mechanisms also help market participants better understand market dynamics and enhance their adaptability to market changes and crisis management capabilities.

[0022] Furthermore, according to the real-time monitoring of the green power market, power supply and demand monitoring data, power transaction monitoring data and power environmental monitoring data are obtained, including:

[0023] Real-time monitoring is performed on the green electricity market to obtain an electricity market monitoring data set; data cleaning is performed on the electricity market monitoring data set according to predetermined data processing factors to obtain an electricity market monitoring result; feature recognition is performed based on the electricity market monitoring result to generate the electricity supply and demand monitoring data, the electricity transaction monitoring data and the electricity environmental protection monitoring data.

[0024] Furthermore, the predetermined data processing factors include duplicate value deletion, missing value supplementation, error value correction and data standardization.

[0025] Furthermore, the real-time monitoring system is deployed at key nodes of the power market, such as power stations, distribution networks and their trading markets. The system regularly collects raw data on power supply, demand, trading conditions and environmental protection indicators to form a power market monitoring data set containing extensive market information. Next, in order to ensure the quality of the data and the accuracy of subsequent processing, data cleaning is carried out. Data cleaning is mainly performed through predetermined data processing factors, including: duplicate value deletion, missing value supplementation, error value correction and data standardization. Specifically, duplicate value deletion is to eliminate redundant information in the data set and ensure the uniqueness of the data; missing value supplementation is to fill missing data points through historical data, average values ​​or other statistical methods to maintain the integrity of the data set; error value correction involves identifying and correcting abnormal or illogical data in the data set to ensure the authenticity and reliability of the data; data standardization is to convert data from different sources and formats into a unified format to make it suitable for subsequent analysis and processing. The power market monitoring result obtained after data cleaning is a high-quality, pre-processed data set, which lays the foundation for the next step of feature identification. The process of feature identification is to use data analysis technology, such as machine learning algorithms or statistical analysis methods, to extract key features related to power supply and demand balance, market trading activities and environmental protection standards from the processed data set. These features are then used to form more refined and specialized power supply and demand monitoring data, power trading monitoring data and power environmental protection monitoring data. For example, peak consumption periods and supply shortages can be identified from supply and demand data, price fluctuations and trading patterns can be analyzed from trading data, and pollution emission levels and compliance with environmental protection regulations can be monitored from environmental protection data. Through the above data collection, cleaning and feature extraction steps, a set of comprehensive, accurate and real-time updated data support can be provided for power market risk assessment, greatly enhancing the ability of green power market monitoring and risk management, which not only improves the transparency of market operation, but also enhances the ability of market participants to respond to upcoming market changes, thereby helping to achieve a more sustainable and environmentally friendly power supply.

[0026] Furthermore, the construction of a power market risk assessment and prediction model includes:

[0027] Conduct deep learning on the neural network to build a power supply and demand risk assessment and prediction model; conduct supervised learning on the neural network to build a power trading risk assessment and prediction model; conduct supervised training on the neural network to establish a power environmental risk assessment and prediction model; connect the power supply and demand risk assessment and prediction model, the power trading risk assessment and prediction model and the power environmental risk assessment and prediction model to generate the power market risk assessment and prediction model.

[0028] Specifically, by using deep learning neural network technology, the power supply and demand risk assessment prediction model, the power trading risk assessment prediction model and the power environmental risk assessment prediction model are constructed respectively, and these models are integrated to provide a comprehensive market risk assessment tool. First, a suitable neural network architecture is selected, such as a convolutional neural network (CNN) or a recurrent neural network (RNN), depending on the time series characteristics of the data and its complexity. This model automatically learns the potential patterns and trends in the power supply and demand data through deep learning technology, so that it can predict the risk of market supply and demand imbalance. For example, the model may learn the increasing trend of power demand under specific holidays or climatic conditions. Next, the construction of the power trading risk assessment prediction model involves the use of supervised learning methods, in which the network is trained to identify and predict risk factors that lead to market price fluctuations or abnormal transactions. This model predicts the risk of possible market manipulation or abnormal fluctuations by analyzing factors such as trading volume, price changes and market participant behavior. Further, the establishment of the power environmental risk assessment prediction model also adopts a supervised training method, which focuses on extracting key indicators such as emission levels and renewable energy utilization from environmental monitoring data to assess the risk of possible violations of environmental regulations or increased environmental impacts. This model helps predict risks caused by changes in environmental policies or unpredictable environmental disasters. Finally, the three models are connected through an integration method to form a comprehensive power market risk assessment and prediction model. During the integration process, the output of each individual model is assigned a certain weight, which is determined according to the importance of each risk type in the overall market risk assessment. In this way, the final integrated model can output a comprehensive market risk score that comprehensively considers the impact of supply and demand balance, market trading behavior and environmental protection standards on market stability. Through the above method, a highly accurate and real-time power market risk assessment can be provided to help market operators, policymakers and relevant regulatory agencies better understand market dynamics and respond to potential risks in a timely manner, thereby maintaining the health and sustainable development of the power market.

[0029] Furthermore, deep learning based on neural networks is used to build a power supply and demand risk assessment and prediction model, including:

[0030] Retrieve a power supply and demand monitoring sample set and a power supply and demand risk assessment sample set; perform data cleaning based on the power supply and demand monitoring sample set and the power supply and demand risk assessment sample set to obtain a power supply and demand risk assessment record set; perform data division based on the power supply and demand risk assessment record set to obtain a supply and demand risk assessment training set and a supply and demand risk assessment test set; train and test a neural network based on the supply and demand risk assessment training set and the supply and demand risk assessment test set to generate the power supply and demand risk assessment prediction model.

[0031] Optionally, retrieve the power supply and demand monitoring sample set and the power supply and demand risk assessment sample set. This step involves collecting data samples about power supply and demand from the real-time monitoring system and the historical database. The power supply and demand monitoring sample set may include historical power supply, demand forecast, actual consumption data, etc., while the power supply and demand risk assessment sample set includes risk events and consequences recorded in specific supply and demand scenarios, such as supply shortages and excess demand event records. Then, data cleaning is performed based on the power supply and demand monitoring sample set and the power supply and demand risk assessment sample set to obtain a power supply and demand risk assessment record set. The data cleaning process includes steps such as removing erroneous data, filling missing values, and standardizing data formats to ensure the quality and consistency of the data so that it is suitable for subsequent analysis and model training. Subsequently, data partitioning is performed based on the obtained power supply and demand risk assessment record set to obtain a supply and demand risk assessment training set and a supply and demand risk assessment test set. This step usually uses random sampling or a time series-based partitioning method to ensure that the training set and the test set can represent the characteristics of the overall data while avoiding overfitting. The training set is used to build and adjust the model parameters, while the test set is used to evaluate the generalization ability and prediction accuracy of the model. Finally, the neural network is trained and tested based on the supply and demand risk assessment training set and the supply and demand risk assessment test set to generate the power supply and demand risk assessment prediction model. In this stage, the neural network adjusts its weights and biases by learning the data patterns in the training set, and optimizes the structure of the model to adapt to the characteristics of the supply and demand data. During the training process, the back propagation algorithm is used to minimize the prediction error and enhance the model's prediction ability for unknown data. The testing phase verifies the performance of the model on new data to ensure its accuracy and reliability. Through the above steps, a neural network model that can accurately predict the supply and demand risks of the electricity market was successfully constructed. This model can help market operators identify potential supply and demand imbalances in a timely manner and take appropriate regulatory measures to maintain the stability and sustainable development of the electricity market.

[0032] Furthermore, based on the power supply and demand monitoring data, the power transaction monitoring data and the power environmental monitoring data, the green power market is subjected to risk assessment and prediction according to the power market risk assessment and prediction model to obtain a power market risk coefficient, including:

[0033] Input the electricity supply and demand monitoring data into the electricity supply and demand risk assessment prediction model to obtain the electricity supply and demand risk assessment prediction coefficient; input the electricity trading monitoring data into the electricity trading risk assessment prediction model to obtain the electricity trading risk assessment prediction coefficient; input the electricity environmental monitoring data into the electricity environmental risk assessment prediction model to obtain the electricity environmental risk assessment prediction coefficient; calculate the electricity market risk coefficient based on the electricity supply and demand risk assessment prediction coefficient, the electricity trading risk assessment prediction coefficient and the electricity environmental risk assessment prediction coefficient.

[0034] Further, the power supply and demand monitoring data is input into the power supply and demand risk assessment prediction model. This step involves collecting real-time data on the supply and demand of the power market, such as power generation, power consumption, and demand peak information. These data are preprocessed by advanced data processing technology and then fed into the neural network model. The model analyzes these data, predicts possible supply and demand imbalances, and outputs a power supply and demand risk assessment prediction coefficient. This coefficient reflects the degree of risk that the market may face from the perspective of supply and demand. Next, the power transaction monitoring data is input into the power transaction risk assessment prediction model. This step includes collecting data on power transactions, such as transaction prices, transaction volumes, and behavior patterns of market participants. These data are analyzed by a supervised learning neural network model, which aims to identify risk factors that may lead to price fluctuations or market manipulation, and output power transaction risk assessment prediction coefficients accordingly. Next, the power environmental monitoring data is input into the power environmental risk assessment prediction model, which involves the impact of power production and consumption on the environment, including emission data, waste disposal, and resource utilization efficiency. The environmental risk model takes these data as input, analyzes possible environmental risks such as excessive emissions and resource depletion risks through a prediction model, and outputs a prediction coefficient for power environmental risk assessment. Finally, based on the obtained power supply and demand risk assessment prediction coefficient, power trading risk assessment prediction coefficient, and power environmental risk assessment prediction coefficient, a comprehensive assessment method is used to calculate the power market risk coefficient. This usually involves a weighted average of the various coefficients, where the weights reflect the importance of different risk types in the overall market risk. The final power market risk coefficient is a comprehensive value that provides market participants and regulators with a quantitative and comprehensive risk assessment result to help them formulate corresponding risk management strategies and countermeasures. Through the above steps, the multi-dimensional risks of the green power market can be accurately predicted and assessed, thereby promoting the stable operation and sustainable development of the market.

[0035] Furthermore, the power market risk coefficient is calculated according to the power supply and demand risk assessment prediction coefficient, the power transaction risk assessment prediction coefficient and the power environmental protection risk assessment prediction coefficient, including:

[0036] Set power supply and demand risk weights, power trading risk weights and power environmental risk weights; construct a power risk calculation weighted network based on the power supply and demand risk weights, the power trading risk weights and the power environmental risk weights; input the power supply and demand risk assessment prediction coefficients, the power trading risk assessment prediction coefficients and the power environmental risk assessment prediction coefficients into the power risk calculation weighted network, and output the power market risk coefficients.

[0037] In a specific embodiment, the power supply and demand risk weight, power trading risk weight and power environmental protection risk weight are set. The core of this step is to determine the weight according to the degree of influence of various risks on market stability. The power supply and demand risk weight reflects the potential threat of supply and demand imbalance to market stability; the power trading risk weight takes into account the risks of price fluctuations and market manipulation; the power environmental protection risk weight reflects the importance of compliance with environmental protection laws and regulations and their environmental impact. These weights are determined through market historical data analysis, expert consultation and risk management strategies to ensure that each type of risk is properly considered. Next, according to the various risk weights set, a power risk calculation weighted network is constructed. This network uses mathematical models and algorithms to combine the weights of different risks with the corresponding risk assessment prediction coefficients to calculate the overall market risk value. The construction of the weighted network is based on linear or nonlinear models to ensure that various risk factors can be correctly integrated into the overall risk assessment according to their relative importance. Finally, the power supply and demand risk assessment prediction coefficient, the power trading risk assessment prediction coefficient and the power environmental protection risk assessment prediction coefficient are input into the power risk calculation weighted network. These coefficients are generated by their respective risk assessment models, and each coefficient represents the current risk level of its corresponding field. These input data are processed through a weighted network, and a comprehensive power market risk coefficient is finally output. This coefficient is a quantitative value that reflects the overall risk status of the market after considering all important risk factors. This method ensures the comprehensiveness and accuracy of risk assessment through scientific weight configuration and efficient network algorithm, and provides a reliable risk management tool for market managers and decision makers. Through this systematic risk assessment process, it can help relevant parties to understand the market situation in a timely manner and take corresponding measures to mitigate potential negative impacts and promote the stable development of the power market.

[0038] Furthermore, if Figure 2 As shown, if the power market risk coefficient is greater than or equal to the power market risk threshold, a power market risk warning signal is generated, including:

[0039] If the electricity market risk coefficient is greater than or equal to the electricity market risk threshold, the risk warning grading device is activated; the electricity market risk coefficient is input into the risk warning grading device to obtain the electricity market risk warning signal.

[0040] Specifically, it is determined whether the power market risk coefficient is greater than or equal to the set power market risk threshold. The power market risk threshold is a standard set based on historical data analysis, market operation mode and potential risk factors to quantify the maximum risk level that the market can bear. When the power market risk coefficient calculated in real time reaches or exceeds this threshold, it indicates that the market risk has reached or exceeded the controllable range, and immediate measures need to be taken to avoid potential adverse effects. Then, once the power market risk coefficient exceeds the risk threshold, the risk warning grading device will be activated. The risk warning grading device is a highly automated system that determines the corresponding warning level based on the value of the input risk coefficient. The device is designed with multiple warning levels, each level corresponding to different degrees of market risk and recommended countermeasures, thereby ensuring the specificity and practicality of the warning signal. Then, the power market risk coefficient is input into the activated risk warning grading device. This device uses a pre-set algorithm and logic to evaluate the risk status of the current market according to the input risk coefficient and selects a suitable warning level. This process is dynamic and can adjust the warning level in real time according to the changes in the market risk coefficient. Finally, the power market risk warning signal is obtained from the risk warning classification device. This warning signal includes a detailed description of the warning level and its corresponding recommendations or mandatory measures, such as increasing power supply, adjusting power prices, and launching emergency response plans. The warning signal will be conveyed to market operators, regulators, and related companies through multiple channels to ensure that all relevant parties can understand the market risk situation in a timely manner and take corresponding measures. Through the above-mentioned risk warning mechanism, the power market's response speed and handling capabilities to risks can be effectively improved, and the market's stability and security can be enhanced, thereby ensuring the continuity of power supply and the normal operation of the economy and society.

[0041] Through the technical solutions of the above embodiments, the green power market risk assessment and prediction method provided by this application has the following technical effects:

[0042] 1. Use a real-time monitoring system to collect supply and demand data, transaction data, and environmental data from the electricity market, rather than relying on historical data or intermittent updates, to ensure the timeliness and accuracy of the data and improve the response speed and reliability of the risk assessment model.

[0043] 2. Construct a power supply and demand risk assessment model, a power trading risk assessment model, and a power environmental risk assessment model, and integrate these models into a comprehensive risk assessment tool that can integrate risk factors of different dimensions and provide more comprehensive risk assessment results, greatly improving the accuracy and comprehensiveness of predictions compared to traditional methods.

[0044] 3. Based on the calculated market risk coefficient and the preset risk threshold, risk warning signals are automatically generated and issued to achieve rapid response to market emergencies, provide timely risk information to market participants, make quick decisions when necessary, and reduce potential economic losses.

[0045] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0046] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0047] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A green power market risk assessment and prediction method, characterized in that: The method comprises: Conduct real-time monitoring based on the green power market to obtain power supply and demand monitoring data, power transaction monitoring data, and power environmental protection monitoring data; Build a risk assessment and prediction model for the electricity market; Based on the power supply and demand monitoring data, the power transaction monitoring data and the power environmental monitoring data, the risk assessment and prediction model of the power market is used to perform risk assessment and prediction on the green power market to obtain a power market risk coefficient; Determining whether the power market risk coefficient is greater than or equal to the power market risk threshold; If the electricity market risk coefficient is greater than or equal to the electricity market risk threshold, an electricity market risk warning signal is generated.

2. A green power market risk assessment and prediction method as claimed in claim 1, characterized in that: Real-time monitoring is carried out according to the green power market to obtain power supply and demand monitoring data, power transaction monitoring data and power environmental protection monitoring data, including: Perform real-time monitoring according to the green power market to obtain a power market monitoring data set; Performing data cleaning on the power market monitoring data set according to a predetermined data processing factor to obtain a power market monitoring result; Feature recognition is performed based on the power market monitoring results to generate the power supply and demand monitoring data, the power transaction monitoring data and the power environmental protection monitoring data.

3. A green power market risk assessment and prediction method as claimed in claim 2, characterized in that: The predetermined data processing factors include duplicate value deletion, missing value supplementation, error value correction and data standardization.

4. A green power market risk assessment and prediction method as claimed in claim 1, characterized in that: Build a power market risk assessment and prediction model, including: Conduct deep learning based on neural networks to build a prediction model for power supply and demand risk assessment; Conduct supervised learning on neural networks and build a risk assessment and prediction model for power trading; Conduct supervised training based on neural networks to establish a prediction model for power environmental risk assessment; The power supply and demand risk assessment and prediction model, the power transaction risk assessment and prediction model and the power environmental protection risk assessment and prediction model are connected to generate the power market risk assessment and prediction model.

5. A green power market risk assessment and prediction method as claimed in claim 4, characterized in that: Based on deep learning of neural networks, a power supply and demand risk assessment and prediction model is constructed, including: Retrieve the power supply and demand monitoring sample set and the power supply and demand risk assessment sample set; Perform data cleaning according to the power supply and demand monitoring sample set and the power supply and demand risk assessment sample set to obtain a power supply and demand risk assessment record set; Data is divided according to the power supply and demand risk assessment record set to obtain a supply and demand risk assessment training set and a supply and demand risk assessment test set; The neural network is trained and tested based on the supply and demand risk assessment training set and the supply and demand risk assessment test set to generate the power supply and demand risk assessment prediction model.

6. A green power market risk assessment and prediction method as claimed in claim 4, characterized in that: Based on the power supply and demand monitoring data, the power transaction monitoring data and the power environmental monitoring data, the green power market is subjected to risk assessment and prediction according to the power market risk assessment and prediction model to obtain a power market risk coefficient, including: Inputting the power supply and demand monitoring data into the power supply and demand risk assessment prediction model to obtain a power supply and demand risk assessment prediction coefficient; Inputting the power transaction monitoring data into the power transaction risk assessment prediction model to obtain a power transaction risk assessment prediction coefficient; Inputting the power environmental monitoring data into the power environmental risk assessment prediction model to obtain a power environmental risk assessment prediction coefficient; The electricity market risk coefficient is calculated based on the electricity supply and demand risk assessment prediction coefficient, the electricity trading risk assessment prediction coefficient and the electricity environmental protection risk assessment prediction coefficient.

7. A green power market risk assessment and prediction method as claimed in claim 6, characterized in that: Calculating the power market risk coefficient according to the power supply and demand risk assessment prediction coefficient, the power transaction risk assessment prediction coefficient and the power environmental protection risk assessment prediction coefficient includes: Set risk weights for power supply and demand, power trading and power environmental protection; Constructing a power risk calculation weighted network according to the power supply and demand risk weight, the power transaction risk weight and the power environmental protection risk weight; The power supply and demand risk assessment prediction coefficient, the power transaction risk assessment prediction coefficient and the power environmental protection risk assessment prediction coefficient are input into the power risk calculation weighted network, and the power market risk coefficient is output.

8. A green power market risk assessment and prediction method as claimed in claim 1, characterized in that: If the power market risk coefficient is greater than or equal to the power market risk threshold, a power market risk warning signal is generated, including: If the power market risk coefficient is greater than or equal to the power market risk threshold, activating the risk warning classification device; The power market risk coefficient is input into the risk warning grading device to obtain the power market risk warning signal.

Images