New energy power transaction method and system based on block chain and artificial intelligence

Abstract

The invention provides a new energy power transaction method based on a block chain and artificial intelligence, and the method comprises the steps: constructing an intelligent contract analysis model, carrying out the connection with an external data source, and carrying out the real-time adjustment of an existing intelligent contract term; predicting the real-time new energy power generation power, and analyzing historical transaction data between participants to obtain a real-time prediction result and personalized transaction characteristics of each participant; intelligent contract terms are screened according to personalized transaction features, exclusive contract terms meeting the demands of participants are generated, and supply and demand parties are dynamically matched through an artificial intelligence algorithm; and constructing a risk assessment model of the new energy power transaction, assessing real-time transaction data in the transaction process of the supplier and the demander, and automatically generating a corresponding coping strategy. According to the invention, the block chain and the artificial intelligence technology are combined, so that the new energy power transaction efficiency and the risk control capability are improved.

Description

Technical Field

[0001] The present invention relates to the field of new energy power trading technology, and in particular to a new energy power trading method and system based on blockchain and artificial intelligence. Background Art

[0002] Energy is the foundation of today's economy and society. New energy, in its clean, renewable form, has become a key focus in the global energy sector. New energy power trading specifically refers to medium- and long-term electricity transactions involving green power products, designed to meet the needs of power users for purchasing and consuming green electricity. However, traditional power market trading mechanisms have demonstrated numerous shortcomings in addressing the challenges posed by the rapid growth of the new energy industry, including information asymmetry, high transaction costs, and low efficiency. Information asymmetry makes it difficult for market participants to obtain accurate market information, increasing transaction uncertainty and risk. Transactions involve multiple intermediaries and complex settlement processes, resulting in high transaction costs and limiting market activity and development potential. Furthermore, the centralized data structure of traditional trading mechanisms is vulnerable to single points of failure and data tampering, lacking transparency and trust, and hindering fair competition and healthy market development. Summary of the Invention

[0003] The purpose of this invention is to provide a new energy power trading method and system based on blockchain and artificial intelligence, by combining blockchain with artificial intelligence technology, thereby improving the efficiency and risk control capabilities of new energy power trading.

[0004] To achieve the above objectives, the present invention provides a new energy power trading method based on blockchain and artificial intelligence in a first aspect, the method comprising:

[0005] Analyze historical smart contracts through artificial intelligence algorithms, build smart contract analysis models, connect smart contract analysis models with external data sources, and make real-time adjustments to existing smart contract terms;

[0006] By combining meteorological data and historical power generation data with artificial intelligence algorithms, real-time renewable energy power generation is predicted. Historical transaction data between participants is analyzed to obtain real-time prediction results and personalized transaction characteristics for each participant.

[0007] Smart contract terms are screened, combined, and optimized based on the participants’ personalized transaction characteristics to generate exclusive contract terms that meet the participants’ needs;

[0008] Based on real-time prediction results and exclusive contract terms for each participant's needs, the supply and demand sides are dynamically matched through artificial intelligence algorithms, a risk assessment model for new energy power transactions is constructed, the real-time transaction data of the supply and demand sides during the transaction process is evaluated, and corresponding response strategies are automatically generated based on the evaluation results.

[0009] Furthermore, the existing smart contract terms will be adjusted in real time, including:

[0010] Collect historical smart contract data, extract key information from the historical smart contract data, and build a smart contract knowledge graph based on the extracted key information;

[0011] Based on the smart contract knowledge graph, we can reason and learn the implicit relationships between clauses, potential logical loopholes, and possible improvement directions, thereby building a smart contract analysis model.

[0012] Connect the smart contract analysis model with external data sources, analyze the real-time updated data from external data sources, and make real-time adjustments to existing smart contract terms based on the analysis results.

[0013] Furthermore, exclusive contract terms are generated to meet the needs of participants, including:

[0014] Based on the smart contract knowledge graph, candidate contract terms that meet the needs of participants are screened according to their personalized transaction characteristics;

[0015] Combine candidate contract clauses to obtain a draft contract;

[0016] The draft contract is input into the smart contract analysis model for evaluation, and the draft contract is optimized to generate exclusive contract terms that meet the needs of participants.

[0017] Furthermore, the draft contract was optimized, including:

[0018] The combined contract terms are represented as chromosomes, with each term as a gene, and the value of the gene is whether the term is selected or the specific parameter value of the term;

[0019] A certain number of chromosomes are randomly generated as the initial population, each chromosome represents a possible combination of contract terms;

[0020] Define a fitness function to measure the quality of the contract terms combination represented by each chromosome. Based on the value of the fitness function, a selection strategy is used to select chromosomes with a fitness value greater than the preset value from the current population as parent individuals to participate in crossover and mutation operations.

[0021] Repeat the selection, crossover, and mutation operations to generate a new generation of populations, calculate the fitness value of each individual in the new generation of populations, replace the current population with the new generation of populations, and continue iterative updates until the termination conditions are met, and output exclusive contract terms that meet the needs of the participants.

[0022] Furthermore, the real-time renewable energy power generation power is predicted, including:

[0023] Acquire real-time meteorological data and historical power generation data of the corresponding new energy power generation device, the historical power generation data including historical operation data and corresponding historical meteorological data, perform feature extraction on the historical operation data and the corresponding historical meteorological data, and obtain historical first power generation feature data and historical second power generation feature data;

[0024] Extracting correlation feature data between the historical first power generation feature data and the historical second power generation feature data, inputting the historical first power generation feature data, the historical second power generation feature data and the correlation feature data into a preset power prediction model for training, and obtaining a power prediction model for the new energy power plant;

[0025] The power prediction model includes a historical power generation data analysis sub-model and a power prediction sub-model. Real-time meteorological data and the latest historical power generation data of the corresponding new energy power generation device are input into the power prediction model to predict the power generation data of the new energy power plant in the future target time period.

[0026] Furthermore, historical transaction data between participants is analyzed, including:

[0027] Obtain complete transaction records between participants from the new energy power trading platform, analyze the transaction records, and construct a transaction network diagram. Use network analysis to identify participants' trading strategies, trading preferences, risk appetite, and credit status;

[0028] Clustering and grouping participants based on their trading strategies, trading preferences, risk appetite, and credit status;

[0029] Perform correlation analysis on the transaction characteristics of participants in the completed cluster grouping to obtain the personalized transaction characteristics of each participant.

[0030] Furthermore, artificial intelligence algorithms are used to dynamically match supply and demand, specifically including:

[0031] Build a transaction matching model, input the exclusive contract terms required by both the supply and demand sides into the transaction matching model, and output the initial matching plan;

[0032] Monitor market dynamics and changes in the situations of both parties to the transaction in real time, dynamically adjust the matching strategy based on the initial matching plan, and search for the optimal matching plan according to preset rules.

[0033] Furthermore, corresponding response strategies are automatically generated, including:

[0034] Establish a transaction risk assessment model, using real-time transaction data and risk indicators between participants based on the optimal matching plan as model input, conduct real-time risk assessment of each transaction, and output risk assessment results;

[0035] Analyze and mine historical transaction cases and response strategies to build a risk response strategy library;

[0036] Based on the risk assessment results, combined with the risk response strategy library and real-time market conditions, the corresponding response strategy is automatically generated and automatically executed through the smart contract.

[0037] A second aspect of the present invention provides a new energy power trading system based on blockchain and artificial intelligence, the system comprising:

[0038] Smart contract optimization module, which is used to analyze historical smart contracts through artificial intelligence algorithms, build smart contract analysis models, connect smart contract analysis models with external data sources, and make real-time adjustments to existing smart contract terms;

[0039] The real-time market monitoring module uses artificial intelligence algorithms to combine meteorological data and historical power generation data to predict real-time renewable energy power generation. It also analyzes historical transaction data between participants to obtain real-time prediction results and personalized transaction characteristics for each participant.

[0040] The smart contract matching module is used to screen, combine, and optimize smart contract terms based on the personalized transaction characteristics of participants to generate exclusive contract terms that meet the needs of participants. Based on real-time prediction results and the exclusive contract terms required by each participant, the module dynamically matches supply and demand through artificial intelligence algorithms;

[0041] The risk assessment and response module is used to build a risk assessment model for new energy power transactions, evaluate the real-time transaction data during the transaction between supply and demand parties, and automatically generate corresponding response strategies based on the assessment results.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] The present invention provides a new energy power trading method and system based on blockchain and artificial intelligence. By building a smart contract analysis model and connecting to external data sources, the smart contract terms are adjusted in real time, so that in the new energy power trading process, the contract terms can be automatically updated in time according to changes in the external environment; historical transaction data between participants is analyzed to obtain personalized transaction characteristics of each participant. The artificial intelligence decision model screens, combines and optimizes the smart contract terms based on personalized transaction characteristics, and generates exclusive contract terms for each participant that meet their needs; the artificial intelligence algorithm is used to dynamically match the supply and demand sides, quickly and efficiently connecting power suppliers and demanders; the constructed new energy power trading risk assessment model evaluates the real-time transaction data of the supply and demand transaction between participants, promptly discovers risks that may arise in the transaction process, automatically generates corresponding response strategies based on the assessment results, automatically triggers the deployment strategy, and ensures the stability and reliability of the transaction. The present invention improves the efficiency and risk control capabilities of new energy power trading by combining blockchain with artificial intelligence technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without any creative work.

[0045] Figure 1 A flowchart of a new energy power trading method based on blockchain and artificial intelligence is provided for an embodiment of the present invention;

[0046] Figure 2 A schematic diagram of the structure of a new energy power trading system based on blockchain and artificial intelligence is provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0047] The present invention will be further described in detail below with reference to the accompanying drawings and examples. It will be understood that the specific embodiments described herein are intended only to illustrate the present invention and are not intended to limit the present invention. It should also be noted that, for ease of description, the accompanying drawings only illustrate portions relevant to the present invention, not all structures.

[0048] Reference Figure 1 In a first aspect, this embodiment provides a new energy power trading method based on blockchain and artificial intelligence, the method comprising:

[0049] Analyze historical smart contracts through artificial intelligence algorithms, build smart contract analysis models, connect the smart contract analysis models with external data sources, and make real-time adjustments to existing smart contract terms.

[0050] By combining meteorological data and historical power generation data with artificial intelligence algorithms, real-time renewable energy power generation is predicted, and historical transaction data between participants is analyzed to obtain real-time prediction results and personalized transaction characteristics for each participant.

[0051] Smart contract terms are screened, combined, and optimized based on the participants' personalized transaction characteristics to generate exclusive contract terms that meet the participants' needs. Based on real-time prediction results and the exclusive contract terms required by each participant, supply and demand are dynamically matched through artificial intelligence algorithms.

[0052] Build a risk assessment model for new energy power transactions, evaluate the real-time transaction data during the transaction between supply and demand parties, and automatically generate corresponding response strategies based on the assessment results.

[0053] In this embodiment, artificial intelligence algorithms are used to analyze a large amount of historical smart contracts and related business data to build a smart contract analysis model, learning the logic of clauses, structural patterns, and common issues. By connecting the smart contract analysis model with external data sources, such as government regulatory websites, industry news platforms, and market databases, it can obtain real-time information on changes in laws and regulations, updated industry standards, and market trends. The smart contract analysis model automatically analyzes and proposes modifications to smart contract clauses. After review and approval by relevant parties, the smart contract is updated promptly. For example, when relevant tax regulations are adjusted, the smart contract analysis model immediately captures this change and automatically updates the tax calculation clause in the contract to ensure the legality and adaptability of the contract.

[0054] Renewable energy generation is significantly affected by meteorological conditions, resulting in fluctuating and uncertain power generation. By combining meteorological data with historical power generation data, artificial intelligence algorithms can more accurately predict real-time renewable energy power generation. For example, by predicting wind speed or solar radiation intensity in advance, power generation can be predicted. This allows power generation companies to pre-emptively arrange transactions in the power market, avoiding contract defaults and financial losses caused by power generation not meeting expectations. Power market participants include power generation companies, power suppliers, and power users. By analyzing historical transaction data, the personalized trading characteristics of each participant can be identified. For example, customized trading plans and contract terms can be provided to different participants based on transaction price preferences, transaction volume requirements, and transaction time preferences, thereby optimizing the allocation of power resources.

[0055] By generating tailored contract terms for different participants based on their individual transaction characteristics, the needs of each participant can be precisely met. For example, cost-conscious electricity users can be provided with contract terms that include more favorable electricity prices; users with extremely high requirements for power supply stability can be provided with corresponding safeguards and compensation mechanisms. A risk assessment model is built to evaluate real-time transaction data during the transaction process, promptly identifying potential risk factors such as market price fluctuations, power supply interruptions, and transaction defaults. By monitoring and analyzing risks in real time and automatically generating corresponding response strategies based on the assessment results, the system can quickly respond to various risk scenarios and mitigate risk losses. For example, when significant market price fluctuations are detected, the system can automatically adjust the transaction price or volume to mitigate the impact of market risks on participants. When a potential power shortage is predicted, backup power sources can be activated in advance or load adjustments can be made to ensure power supply stability.

[0056] Real-time adjustments to existing smart contract terms, including:

[0057] Collect historical smart contract data, extract key information from the historical smart contract data, and build a smart contract knowledge graph based on the extracted key information.

[0058] Based on the smart contract knowledge graph, we can reason and learn the implicit relationships between clauses, potential logical loopholes, and possible improvement directions to build a smart contract analysis model.

[0059] Connect the smart contract analysis model with external data sources, analyze the real-time updated data from external data sources, and make real-time adjustments to existing smart contract terms based on the analysis results.

[0060] In this example, historical smart contract data is collected and a smart contract knowledge graph is constructed to more accurately understand and analyze key information within smart contract clauses. Based on the smart contract knowledge graph, implicit relationships between clauses are inferred and learned, thereby identifying potential logical vulnerabilities and ensuring compliance with relevant laws, regulations, and industry standards. Potential logical vulnerabilities can lead to problems such as financial losses and transaction disputes during contract execution. Proactively identifying and addressing these vulnerabilities can improve the security of smart contracts and reduce transaction risks. By connecting to external data sources and adjusting contract terms in real time, the smart contract analysis model ensures that smart contracts always meet the latest compliance requirements and avoids legal issues caused by non-compliance with clauses.

[0061] Specifically, through various channels related to new energy power trading, such as blockchain browsers, smart contract databases, and developer communities, a large amount of historical smart contract data, including smart contract code, transaction records, and development documentation, is obtained. The collected smart contract data is cleaned and formatted to remove invalid information and duplicate content, and the text format is standardized. Natural language processing tools and techniques, such as lexical analysis, syntactic analysis, and named entity recognition, are used to extract key information from the smart contract text. This extracted key information is annotated and categorized to form a structured data format. Based on the key information in the smart contract, the knowledge graph structure is designed, consisting of nodes and edges. Nodes represent entities such as contract parties and clauses, and edges represent relationships between entities, such as "include" and "associate."

[0062] Leveraging information from the knowledge graph, machine learning and reasoning algorithms are employed to learn the implicit relationships between smart contract clauses, potential logical loopholes, and possible improvement areas. For example, by analyzing the relationships between clauses, it is possible to identify certain combinations of clauses that could lead to inconsistencies or loopholes in contract execution. The smart contract analysis model is connected to external data sources related to renewable energy power transactions. Real-time data updates from these external data sources are analyzed, and combined with information from the smart contract knowledge graph, existing smart contract clauses can be adjusted in real time. For example, when market price fluctuations or changing weather conditions could affect renewable energy generation and transactions, pricing or delivery terms in the smart contract can be promptly modified based on the analysis results, allowing the contract to adapt to the new trading environment and requirements.

[0063] Generate contract terms tailored to the needs of participants, including:

[0064] Based on the smart contract knowledge graph, candidate contract terms that meet the needs of participants are screened according to their personalized transaction characteristics.

[0065] The candidate contract clauses are combined to obtain a draft contract.

[0066] The draft contract is input into the smart contract analysis model for evaluation, and the draft contract is optimized to generate exclusive contract terms that meet the needs of participants.

[0067] In this embodiment, based on the personalized transaction characteristics of the participants, clauses are filtered and combined from the smart contract knowledge graph to generate a dedicated contract that precisely meets their needs, thereby improving participant satisfaction. Both electricity users and power generation companies can obtain contract terms that meet their specific circumstances and requirements. For example, small electricity users can customize small-volume transactions and flexible billing contracts, while new energy power generation companies can customize contracts that include consumption guarantees and subsidy settlements. Natural language processing technology is used to quickly combine candidate clauses into drafts, eliminating the time-consuming and inefficient manual drafting process and accelerating the start-up of new energy power transactions. Automatically evaluate and optimize contract drafts, quickly identify and modify issues, and reduce the need for repeated manual review and modification, making the transaction process smoother and more efficient.

[0068] Specifically, natural language processing technology is used to combine and adjust the selected candidate clauses to generate a preliminary contract draft. This draft is then fed into a smart contract analysis model for a comprehensive assessment of clause completeness, logical consistency, regulatory compliance, and potential risks. For example, this assessment examines the existence of conflicting clauses, compliance with relevant laws and regulations, and coverage of all necessary transaction details. Based on this assessment, the draft contract is optimized and adjusted, including revisions to clause content, additions to missing clauses, and adjustments to the order of clauses, to ensure that the resulting unique contract terms meet the needs of participants and are legally and commercially feasible. The optimized draft contract is then reviewed again for accuracy and applicability, ultimately generating unique contract terms that meet the needs of participants. This contract is then deployed and executed on the smart contract platform. Once generated, the contract can be used in new energy power transactions, ensuring smooth transactions.

[0069] Optimize the contract draft, including:

[0070] The combined contract terms are represented as chromosomes, with each term as a gene. The value of the gene is whether the term is selected or the specific parameter value of the term.

[0071] A certain number of chromosomes are randomly generated as the initial population, and each chromosome represents a possible combination of contract terms.

[0072] Define a fitness function to measure the quality of the contract terms combination represented by each chromosome. Based on the value of the fitness function, a selection strategy is used to select chromosomes with a fitness value greater than the preset value from the current population as parent individuals to participate in crossover and mutation operations.

[0073] Repeat the selection, crossover, and mutation operations to generate a new generation of populations, calculate the fitness value of each individual in the new generation of populations, replace the current population with the new generation of populations, and continue iterative updates until the termination conditions are met, and output exclusive contract terms that meet the needs of the participants.

[0074] In this embodiment, there are numerous ways to combine contract terms, making it difficult for traditional methods to exhaust all possibilities and find the optimal solution. By representing contract terms as chromosomes and using the selection or nonselection of terms or parameter values as gene values, a genetic algorithm framework can be used to effectively explore the vast and complex combination space and discover potentially high-quality solutions. A fitness function is defined to quantify the quality of contract term combination solutions. Based on these fitness values, high-performing chromosomes are selected as parent individuals for genetic manipulation. This gradually eliminates inferior solutions while retaining and inheriting the characteristics of high-quality solutions. Crossover and mutation operations continuously introduce new features and diversity, preventing regression into local optima and ultimately outputting contract terms tailored to the needs of participants. Participants' needs are often complex and diverse, involving the synergistic interaction of multiple terms and parameters. Genetic algorithms can handle complex optimization problems with multiple objectives and constraints, comprehensively considering the interactions between different terms and their compatibility with participant needs, thereby generating a combination of contract terms that both meets individual requirements and is optimal overall.

[0075] Forecasting of real-time renewable energy power generation, including:

[0076] Real-time meteorological data and historical power generation data of the corresponding new energy power generation device are obtained, where the historical power generation data includes historical operation data and corresponding historical meteorological data. Feature extraction is performed on the historical operation data and the corresponding historical meteorological data to obtain historical first power generation feature data and historical second power generation feature data.

[0077] The correlation characteristic data between the historical first power generation characteristic data and the historical second power generation characteristic data are extracted, and the historical first power generation characteristic data, the historical second power generation characteristic data and the correlation characteristic data are input into a preset power prediction model for training to obtain a power prediction model for the new energy power plant.

[0078] The power prediction model includes a historical power generation data analysis sub-model and a power prediction sub-model. Real-time meteorological data and the latest historical power generation data of the corresponding new energy power generation device are input into the power prediction model to predict the power generation data of the new energy power plant in the future target time period.

[0079] In this embodiment, by combining real-time meteorological data with historical power generation data from renewable energy power generation devices, various factors influencing power generation are more comprehensively captured. Feature extraction is performed on historical operating data and historical meteorological data separately, and correlation algorithms are used to extract correlation feature data between them, more accurately identifying key factors that significantly affect power generation. Renewable energy power generation is highly volatile and uncertain due to meteorological conditions. By acquiring real-time meteorological data and combining it with historical data for forecasting, the impact of meteorological changes on power generation is promptly reflected, thereby helping power dispatch departments better plan power production and distribution, optimize grid operations, and reduce power supply instability caused by the volatility of renewable energy power generation. By inputting the extracted feature data into a preset power prediction model for training, the model parameters are continuously optimized, improving the model's predictive performance. Furthermore, the power prediction model is decomposed into a historical power generation data analysis sub-model and a power prediction sub-model to more efficiently process different types of input data and improve the model's overall performance.

[0080] Specifically, real-time meteorological data, including wind speed, wind direction, temperature, humidity, and solar radiation, is acquired from meteorological stations, satellites, and other channels. Historical power generation data, including power generation, equipment operating status, and fault records, is obtained from the monitoring systems of renewable energy power generation devices. After cleaning, denoising, addressing missing values, and converting the collected data, feature extraction is performed on both historical operating data and historical meteorological data. Correlation algorithms, such as the Pearson correlation coefficient and the Spearman rank correlation coefficient, are used to calculate correlations between historical power generation feature data and extract correlation features that significantly influence power generation. A power prediction model is constructed based on a neural network, and the extracted feature data is input into the power prediction model for training. In actual operation, real-time meteorological data and the latest historical power generation data for the corresponding renewable energy power generation devices are fed into the trained power prediction model to predict power generation data for the renewable energy power plant within a target future time period. For example, hourly power generation predictions for the next 24 hours can be used to provide a reference for power dispatch.

[0081] Analyze historical transaction data between participants, including:

[0082] Obtain complete transaction records between participants from the new energy power trading platform, analyze the transaction records, and construct a transaction network diagram. Use network analysis to identify participants' trading strategies, trading preferences, risk appetite, and credit status;

[0083] Clustering and grouping participants based on their trading strategies, trading preferences, risk appetite, and credit status;

[0084] Perform correlation analysis on the transaction characteristics of participants in the completed cluster grouping to obtain the personalized transaction characteristics of each participant.

[0085] In this example, network analysis is used to analyze historical transaction data and construct a transaction network diagram. This identifies transaction strategies, preferences, risk appetite, and credit profiles, visually displaying the transaction relationships between participants. A clustering algorithm is used to group participants, enabling refined management of participant groups. Based on the clustering results, an association rule mining algorithm is used to conduct in-depth analysis of the transaction characteristics of each cluster group, thereby obtaining personalized transaction characteristics for each participant, providing a basis for personalized services and precision marketing.

[0086] Specifically, complete transaction records between participants are obtained from the database of the new energy power trading platform, covering information such as the transaction parties, time, electricity consumption, price, and whether there was a default. The transaction records are then de-noised, missing values are filled, and format conversion is performed to ensure data quality and consistency, laying the foundation for subsequent analysis. A trading network diagram is constructed with participants as nodes and transaction relationships as edges. Edge weights are assigned based on transaction frequency, electricity consumption, and other factors, visually demonstrating trading activity and closeness. Network analysis methods are used to identify participants' trading strategies and preferences. Node degree centrality and betweenness centrality are calculated to identify participants with active trading and intermediary advantages. Community discovery algorithms are used to reveal trading groups, reflecting trading preferences and patterns. Furthermore, credit assessment models are used to analyze creditworthiness, identifying participants with high default risk. A clustering algorithm is then selected to cluster participants based on their trading strategies, preferences, risk appetite, and creditworthiness. Apriori or FP-Growth algorithms are used to discover association rules among participants' transaction characteristics. Based on these association rules, personalized transaction features are extracted, such as whether a participant has large trading volume, conservative risk appetite, and high creditworthiness, forming a comprehensive and personalized profile.

[0087] Dynamically matching supply and demand through artificial intelligence algorithms, including:

[0088] Build a transaction matching model, input the exclusive contract terms required by both the supply and demand sides into the transaction matching model, and output the initial matching plan.

[0089] Monitor market dynamics and changes in the situations of both parties to the transaction in real time, dynamically adjust the matching strategy based on the initial matching plan, and search for the optimal matching plan according to preset rules.

[0090] In this embodiment, by constructing a transaction matching model, comprehensively considering the needs of both supply and demand sides, and inputting exclusive contract terms, an initial matching solution can be quickly generated. The introduction of a reinforcement learning agent enables the matching process to adapt to market dynamics and changes between transaction parties in real time, continuously adjusting the matching strategy, thereby maintaining efficient matching capabilities in an ever-changing market environment. The new energy power market is highly uncertain and dynamic, and the needs and market conditions of both supply and demand sides may change at any time. The dynamic matching mechanism can quickly respond to these changes, improve the market's adaptability and flexibility, and promote its healthy development.

[0091] Specifically, the transaction matching model and reinforcement learning agent are integrated into the new energy power trading platform to form a complete dynamic matching system. This system obtains market dynamics data and real-time information from both parties involved in the transaction, such as electricity price fluctuations and changes in demand from both supply and demand sides. The reinforcement learning agent monitors market data and information from both parties in real time, making decisions based on the current state and pre-set rules, and dynamically adjusting the initial matching plan. The system then pushes the new matching plan to both parties to facilitate the transaction. The adjusted matching results are also recorded for further training and optimization of the agent.

[0092] Automatically generate corresponding response strategies, including:

[0093] Establish a transaction risk assessment model, use the real-time transaction data and risk indicators between participants based on the optimal matching plan as model input, conduct real-time assessment of the risk of each transaction, and output the risk assessment results.

[0094] Analyze and mine historical transaction cases and response strategies to build a risk response strategy library.

[0095] Based on the risk assessment results, combined with the risk response strategy library and real-time market conditions, the corresponding response strategy is automatically generated and automatically executed through the smart contract.

[0096] In this embodiment, a transaction risk assessment model is established through machine learning algorithms, assessing the risk of each transaction in real time. Cluster analysis and association rule mining are also used to build a risk response strategy library, enabling real-time monitoring of transaction risks and rapid response, thereby improving transaction security and stability and reducing risk losses. Response strategies are automatically generated and executed based on real-time market conditions, ensuring timely and accurate responses and improving transaction efficiency. Furthermore, by analyzing and mining historical transaction cases and response strategies, comprehensive and effective solutions can be provided for various risk scenarios, enhancing the system's risk response capabilities.

[0097] Reference Figure 2In a second aspect, this embodiment provides a new energy power trading system based on blockchain and artificial intelligence, the system comprising:

[0098] The smart contract optimization module is used to analyze historical smart contracts through artificial intelligence algorithms, build smart contract analysis models, connect the smart contract analysis models with external data sources, and make real-time adjustments to existing smart contract terms.

[0099] The real-time market monitoring module is used to predict the real-time renewable energy power generation power by combining meteorological data and historical power generation data through artificial intelligence algorithms, and analyze the historical transaction data between participants to obtain real-time prediction results and personalized transaction characteristics of each participant.

[0100] The smart contract matching module is used to screen, combine and optimize smart contract terms based on the personalized transaction characteristics of participants to generate exclusive contract terms that meet the needs of participants. Based on real-time prediction results and the exclusive contract terms required by each participant, it dynamically matches supply and demand parties through artificial intelligence algorithms.

[0101] The risk assessment and response module is used to build a risk assessment model for new energy power transactions, evaluate the real-time transaction data during the transaction between supply and demand parties, and automatically generate corresponding response strategies based on the assessment results.

[0102] In this embodiment, in the new energy power trading system, the smart contract optimization module first constructs a smart contract analysis model by analyzing historical smart contracts and connects to external data sources to adjust existing smart contract terms in real time. The real-time market monitoring module combines meteorological data and historical power generation data to predict real-time new energy power generation, while analyzing participants' historical transaction data to identify each participant's personalized transaction characteristics. The smart contract matching module screens, combines, and optimizes smart contract terms based on personalized transaction characteristics, generates exclusive contract terms, and uses artificial intelligence algorithms to dynamically match supply and demand parties based on real-time prediction results and exclusive contract terms. Finally, the risk assessment and response module constructs a risk assessment model, evaluates transaction data in real time, and automatically generates response strategies. These four modules work closely together to form an automated and efficient new energy power trading process.

[0103] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A new energy power trading method based on blockchain and artificial intelligence, characterized in that: The method comprises: Analyze historical smart contracts through artificial intelligence algorithms, build smart contract analysis models, connect smart contract analysis models with external data sources, and make real-time adjustments to existing smart contract terms; By combining meteorological data and historical power generation data with artificial intelligence algorithms, real-time renewable energy power generation is predicted. Historical transaction data between participants is analyzed to obtain real-time prediction results and personalized transaction characteristics for each participant. Smart contract terms are screened, combined, and optimized based on the participants' personalized transaction characteristics to generate exclusive contract terms that meet the participants' needs. Based on real-time prediction results and the exclusive contract terms required by each participant, supply and demand are dynamically matched through artificial intelligence algorithms. Build a risk assessment model for new energy power transactions, evaluate the real-time transaction data during the transaction between supply and demand parties, and automatically generate corresponding response strategies based on the assessment results.

2. The new energy power trading method based on blockchain and artificial intelligence according to claim 1 is characterized in that: Real-time adjustments to existing smart contract terms, including: Collect historical smart contract data, extract key information from the historical smart contract data, and build a smart contract knowledge graph based on the extracted key information; Based on the smart contract knowledge graph, we can reason and learn the implicit relationships between clauses, potential logical loopholes, and possible improvement directions, thereby building a smart contract analysis model. Connect the smart contract analysis model with external data sources, analyze the real-time updated data from external data sources, and make real-time adjustments to existing smart contract terms based on the analysis results.

3. The new energy power trading method based on blockchain and artificial intelligence according to claim 2 is characterized in that: Generate contract terms tailored to the needs of participants, including: Based on the smart contract knowledge graph, candidate contract terms that meet the needs of participants are screened according to their personalized transaction characteristics; Combine candidate contract clauses to obtain a draft contract; The draft contract is input into the smart contract analysis model for evaluation, and the draft contract is optimized to generate exclusive contract terms that meet the needs of participants.

4. The new energy power trading method based on blockchain and artificial intelligence according to claim 3 is characterized in that: Optimize the contract draft, including: The combined contract terms are represented as chromosomes, with each term as a gene, and the value of the gene is whether the term is selected or the specific parameter value of the term; A certain number of chromosomes are randomly generated as the initial population, each chromosome represents a possible combination of contract terms; Define a fitness function to measure the quality of the contract terms combination represented by each chromosome. Based on the value of the fitness function, a selection strategy is used to select chromosomes with a fitness value greater than the preset value from the current population as parent individuals to participate in crossover and mutation operations. Repeat the selection, crossover, and mutation operations to generate a new generation of populations, calculate the fitness value of each individual in the new generation of populations, replace the current population with the new generation of populations, and continue iterative updates until the termination conditions are met, and output exclusive contract terms that meet the needs of the participants.

5. The new energy power trading method based on blockchain and artificial intelligence according to claim 1 is characterized in that: Forecasting of real-time renewable energy power generation, including: Acquire real-time meteorological data and historical power generation data of the corresponding new energy power generation device, the historical power generation data including historical operation data and corresponding historical meteorological data, perform feature extraction on the historical operation data and the corresponding historical meteorological data, and obtain historical first power generation feature data and historical second power generation feature data; Extracting correlation feature data between the historical first power generation feature data and the historical second power generation feature data, inputting the historical first power generation feature data, the historical second power generation feature data and the correlation feature data into a preset power prediction model for training, and obtaining a power prediction model for the new energy power plant; The power prediction model includes a historical power generation data analysis sub-model and a power prediction sub-model. Real-time meteorological data and the latest historical power generation data of the corresponding new energy power generation device are input into the power prediction model to predict the power generation data of the new energy power plant in the future target time period.

6. The new energy power trading method based on blockchain and artificial intelligence according to claim 1 is characterized in that: Analyze historical transaction data between participants, including: Obtain complete transaction records between participants from the new energy power trading platform, analyze the transaction records, and construct a transaction network diagram. Use network analysis to identify participants' trading strategies, trading preferences, risk appetite, and credit status; Clustering and grouping participants based on their trading strategies, trading preferences, risk appetite, and credit status; Perform correlation analysis on the transaction characteristics of participants in the completed cluster grouping to obtain the personalized transaction characteristics of each participant.

7. The new energy power trading method based on blockchain and artificial intelligence according to claim 1 is characterized in that: Dynamically matching supply and demand through artificial intelligence algorithms, including: Build a transaction matching model, input the exclusive contract terms required by both the supply and demand sides into the transaction matching model, and output the initial matching plan; Monitor market dynamics and changes in the situations of both parties to the transaction in real time, dynamically adjust the matching strategy based on the initial matching plan, and search for the optimal matching plan according to preset rules.

8. The new energy power trading method based on blockchain and artificial intelligence according to claim 1 is characterized in that: Automatically generate corresponding response strategies, including: Establish a transaction risk assessment model, using real-time transaction data and risk indicators between participants based on the optimal matching plan as model input, conduct real-time risk assessment of each transaction, and output risk assessment results; Analyze and mine historical transaction cases and response strategies to build a risk response strategy library; Based on the risk assessment results, combined with the risk response strategy library and real-time market conditions, the corresponding response strategy is automatically generated and automatically executed through the smart contract.

9. A new energy power trading system based on blockchain and artificial intelligence, characterized by: The system comprises: Smart contract optimization module, which is used to analyze historical smart contracts through artificial intelligence algorithms, build smart contract analysis models, connect smart contract analysis models with external data sources, and make real-time adjustments to existing smart contract terms; The real-time market monitoring module uses artificial intelligence algorithms to combine meteorological data and historical power generation data to predict real-time renewable energy power generation. It also analyzes historical transaction data between participants to obtain real-time prediction results and personalized transaction characteristics for each participant. The smart contract matching module is used to screen, combine, and optimize smart contract terms based on the personalized transaction characteristics of participants to generate exclusive contract terms that meet the needs of participants. Based on real-time prediction results and the exclusive contract terms required by each participant, the module dynamically matches supply and demand through artificial intelligence algorithms; The risk assessment and response module is used to build a risk assessment model for new energy power transactions, evaluate the real-time transaction data during the transaction between supply and demand parties, and automatically generate corresponding response strategies based on the assessment results.

Images