Green energy transaction data monitoring system based on blockchain

Abstract

The present invention relates to the field of energy trading technology, specifically to a green energy trading data monitoring system based on blockchain, the system comprising: a data acquisition and correction module extracts transaction volume, transaction price and transaction time from the transaction records received by the blockchain node, and uses data cleaning to eliminate noise and abnormal values ​​in the records. In the present invention, the accuracy and transparency of green energy trading data are enhanced by comprehensively applying blockchain technology and advanced data processing methods, and the data is ensured not to be tampered with by blockchain technology, and the security and trust of the data are improved, thereby providing a solid foundation for market analysis. This precise and reliable data support makes the monitoring of market behavior more accurate, and by analyzing the changes in transaction frequency and the laws of price fluctuations, the periodicity of supply and demand changes is effectively predicted, and the understanding and prediction capabilities of market dynamics are enhanced, and the prediction model is adjusted in real time so that it can adapt to market changes in a timely manner.

Description

Technical Field

[0001] The present invention relates to the field of energy trading technology, and in particular to a green energy trading data monitoring system based on blockchain. Background Art

[0002] The energy trading technology field involves the use of advanced data processing and analysis tools to optimize, manage and predict the buying and selling of energy products such as electricity, natural gas and renewable energy. This field includes real-time data monitoring of energy markets, automation of trading strategies, risk management, and compliance tracking. Technology solutions may include blockchain technology to ensure transparency and security of transactions, big data analysis to predict market trends, and the application of artificial intelligence in automated trading and energy management decision support systems. In addition, with the increasing importance of sustainable development, green energy trading is also gradually increasing, and the application of technology is aimed at supporting environmentally friendly and economically sustainable energy solutions.

[0003] Among them, the green energy transaction data monitoring system refers to a system specially designed to monitor and analyze transaction data involving green energy resources (such as wind energy, solar energy, etc.). The purpose of this system is mainly to ensure the transparency, efficiency and compliance of energy transactions, while helping energy companies optimize the supply chain, reduce energy waste, and enhance market competitiveness by providing real-time data support. By monitoring market dynamics and transaction activities in real time, these systems help predict energy demand and price changes, thereby providing decision-making support for participants.

[0004] Existing technologies have obvious deficiencies in data processing and real-time analysis, especially when processing and integrating large-scale, complex transaction data, where processing delays and analytical inaccuracies are common. In the green energy market, such delays and inaccuracies may lead to misunderstandings of market trends, which in turn affect the formulation of trading strategies. For example, traditional data processing systems may not be able to promptly identify and respond to rapid changes in the market, such as supply and demand fluctuations and price adjustments, which leads to inaccurate forecasts and risk assessments and increases investment risks. The lack of real-time dynamically adjusted forecasting models also limits the effective prediction of future market behavior and reduces the adaptability and foresight of decision-making. These technical limitations are particularly evident in the highly volatile green energy market, which often leads to decision-making errors and lost opportunities. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and propose a green energy transaction data monitoring system based on blockchain.

[0006] In order to achieve the above purpose, the present invention adopts the following technical solution: A green energy transaction data monitoring system based on blockchain includes:

[0007] The data collection and correction module extracts transaction volume, transaction price and transaction time from the transaction records received by the blockchain node, uses data cleaning to eliminate noise and outliers in the records, smooths mutation points in the data, and establishes a transaction data baseline;

[0008] The market behavior analysis module monitors and records the increase and decrease of transaction frequency, the law of price fluctuations and the periodicity of supply and demand changes based on the transaction data baseline, maps the data changes to the differentiated nodes of the decision tree, analyzes the cause and effect relationship of market behavior through the interaction between differentiated nodes, and generates market behavior insights;

[0009] The dynamic prediction adjustment module combines the market behavior insights to adjust the structure of the decision tree model in real time, including the tree depth and node splitting criteria, to match current market trends and fluctuations, refine the decision tree model's prediction of future market dynamics, and generate market dynamics prediction results;

[0010] The risk assessment and early warning module uses the market dynamics prediction results to assess the transaction risk under the current market environment, sets a risk threshold, and sends a transaction risk notification to traders and managers when the market changes predicted by the model exceed the risk threshold.

[0011] As a further solution of the present invention, the steps of establishing the transaction data baseline are specifically as follows:

[0012] Receive data streams from blockchain nodes, filter transaction information, and use the formula:

[0013]

[0014] Generate a preliminary transaction dataset;

[0015] in, Represents a transaction tuple, represents the transaction volume, Represents the transaction price, represents the trading time;

[0016] The preliminary transaction data set is subjected to median filtering to eliminate outliers, using the formula:

[0017]

[0018] Generate a cleaned transaction dataset;

[0019] in, represents the filtered transaction volume, Represents the window size;

[0020] Exponential smoothing is applied to the cleaned transaction data set using the formula:

[0021]

[0022] Generate a volume baseline;

[0023] in, represents the smoothed transaction volume, Represents the smoothing coefficient.

[0024] As a further solution of the present invention, the steps of monitoring and recording the increase and decrease of transaction frequency, the law of price fluctuations and the periodicity of supply and demand changes are specifically:

[0025] Calculate the transaction data from the baseline The increase or decrease in transaction frequency , price fluctuations and supply and demand cyclicality , using the formula:

[0026]

[0027] and

[0028]

[0029] as well as

[0030]

[0031] Generate transaction feature change dataset;

[0032] in, Indicates the increase or decrease in transaction frequency. represents the percentage change in price, A periodic function representing changes in supply and demand, is the time variable, is the cycle length, Adjust the nonlinear response of transaction frequency change data to increase sensitivity to small changes. Adjusting for non-linear responses to price change data, Adjust the nonlinear response of supply and demand cycles and assist the model in identifying key change points in cyclical fluctuations;

[0033] The weighted combination method is applied to analyze the transaction feature change data set and calculate the influencing factors using the formula:

[0034]

[0035] Generate transaction behavior scores;

[0036] in, is the weight of the transaction frequency change, which determines the impact of the transaction frequency change on the total score. is the weight of price fluctuation, specifying the proportion of price change contribution to the total score, The weight of supply and demand cyclicality adjusts the importance of supply and demand changes in the overall model;

[0037] A significance analysis is performed based on the transaction behavior score, using a threshold determination method, and the formula is:

[0038]

[0039] Generate significance determination results of market behavior;

[0040] in, Represents the significant classification result, is the decision threshold.

[0041] As a further solution of the present invention, the step of analyzing the causal relationship of market behavior is specifically:

[0042] The significance determination results of the market behavior are mapped to the differentiation nodes of the decision tree using the formula:

[0043]

[0044] Generate decision tree node mapping results;

[0045] in, represents a decision tree node, Represents the results of the significance determination of market behavior, Represents the market status index, Represents environment variables, Represents historical behavior data;

[0046] Analyze the decision tree nodes Interactions and path dependencies between nodes, select two and For analysis, the formula is:

[0047]

[0048] Generate quantitative evaluation results of the relationships between nodes;

[0049] in, Represents the strength of the relationship between nodes. and Represents a differentiated decision tree node, Adjusting Nodes The impact on relationship strength, Adjustment Node The impact on relationship strength, The effect of moderating node differences on tie strength;

[0050] According to the quantitative evaluation results of the relationship between the nodes To conduct causal analysis, the formula is used:

[0051]

[0052] Generate causal analysis results;

[0053] in, Represents the result of causal relationship analysis, Represents the strength of the relationship between nodes. The influence of the square term of moderated relationship strength, The effect of the square root of moderated relationship strength.

[0054] As a further solution of the present invention, the steps of obtaining the market dynamics forecast results are specifically as follows:

[0055] Combined with the above market behavior insights, the new depth of the decision tree is calculated using the formula:

[0056]

[0057] Generate the current tree depth;

[0058] in, Represents the current tree depth, Represents the original tree depth, is the depth adjustment factor, Indicates the current market trend factor, Adjust thresholds for market trends;

[0059] According to market fluctuations, the updated node splitting criteria are calculated using the formula:

[0060]

[0061] Generate the current node splitting standard;

[0062] in, Represents the current node splitting standard, represents the original split standard, is the split standard adjustment factor, is the current market volatility index, is the fluctuation baseline;

[0063] Integrate the current tree depth and the current node splitting criteria, adjust the decision tree model structure, and use the formula:

[0064]

[0065] Generate an adjusted decision tree model;

[0066] in, represents the adjusted decision tree model, is the current tree depth, is the current node splitting standard;

[0067] Use the adjusted decision tree model to make detailed predictions on future market dynamics using the formula:

[0068]

[0069] Generate market dynamics forecast results;

[0070] in, Indicates the market dynamics forecast results, is the adjusted decision tree model, and data represents the market data used for prediction.

[0071] As a further solution of the present invention, the steps of sending the transaction risk notification are specifically as follows:

[0072] According to the market dynamics forecast results, the risk level in the current market environment is calculated using the formula:

[0073]

[0074] generating risk level assessment values;

[0075] in, Represents the risk level assessment value, represents the risk sensitivity coefficient, Represents the future market forecast value, represents the benchmark market value;

[0076] Compare the risk level with the preset risk threshold using the formula:

[0077]

[0078] Generate risk comparison results;

[0079] in, represents the risk comparison result, Represents the risk level assessment value, represents the risk threshold;

[0080] Based on the risk comparison results, decide whether to send a transaction risk notification using the formula:

[0081]

[0082] Generate a notification sending status, if the notification sending status is 0, no operation is performed, if the notification sending status is 1, a transaction risk notification is sent;

[0083] in, Represents the notification sending status. Represents the risk comparison results.

[0084] Compared with the prior art, the advantages and positive effects of the present invention are:

[0085] In the present invention, the accuracy and transparency of green energy trading data are enhanced by the comprehensive application of blockchain technology and advanced data processing methods. Blockchain technology ensures that data is not tampered with, improves the security and trust of data, and thus provides a solid foundation for market analysis. This precise and reliable data support makes the monitoring of market behavior more accurate. By analyzing the changes in transaction frequency and the laws of price fluctuations, the cyclical nature of supply and demand changes can be effectively predicted, and the understanding and prediction capabilities of market dynamics can be enhanced. The prediction model is adjusted in real time so that it can adapt to market changes in a timely manner. This dynamic adaptation mechanism not only improves the accuracy of the prediction, but also greatly enhances the timeliness and accuracy of risk assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] Figure 1 is a system flow chart of the present invention;

[0087] Figure 2 A flowchart of the steps for establishing a transaction data baseline of the present invention;

[0088] Figure 3 A flowchart of the steps for the present invention to monitor and record the increase and decrease of transaction frequency, the law of price fluctuations and the periodicity of supply and demand changes;

[0089] Figure 4 A flowchart of the steps of analyzing the cause and effect relationship of market behavior in the present invention;

[0090] Figure 5 A flowchart of the steps for obtaining the market dynamics prediction results of the present invention;

[0091] Figure 6 This is a flow chart of the steps for sending a transaction risk notification according to the present invention. DETAILED DESCRIPTION

[0092] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention 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 invention and are not intended to limit the present invention.

[0093] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0094] Embodiment 1

[0095] See also Figure 1 , the green energy transaction data monitoring system based on blockchain includes:

[0096] The data collection and correction module extracts transaction volume, transaction price and transaction time from the transaction records received by the blockchain node, uses data cleaning to eliminate noise and outliers in the records, smooths mutation points in the data, and establishes a transaction data baseline;

[0097] The market behavior analysis module monitors and records the increase and decrease of transaction frequency, the law of price fluctuations and the periodicity of supply and demand changes based on the transaction data baseline, maps the data changes to the differentiated nodes of the decision tree, analyzes the cause and effect relationship of market behavior through the interaction between differentiated nodes, and generates market behavior insights;

[0098] The dynamic prediction adjustment module combines market behavior insights to adjust the structure of the decision tree model in real time, including the tree depth and node splitting criteria, to match current market trends and fluctuations, refine the decision tree model's prediction of future market dynamics, and generate market dynamics prediction results;

[0099] The risk assessment and early warning module uses market dynamics forecast results to assess transaction risks in the current market environment, sets risk thresholds, and sends transaction risk notifications to traders and managers when market changes predicted by the model exceed the risk threshold.

[0100] The transaction data baseline specifically includes basic transaction volume, benchmark transaction price, and standard transaction time. Market behavior insights include frequency increase and decrease patterns, price fluctuation patterns, and cyclical supply and demand changes. Market dynamics forecast results specifically refer to predicted market trends, expected fluctuation range, and future market status. Transaction risk notifications include risk level prompts and risk threshold alerts.

[0101] See also Figure 2 , the specific steps for establishing a transaction data baseline are:

[0102] Receive data streams from blockchain nodes, filter transaction information, and use the formula:

[0103]

[0104] Generate a preliminary transaction dataset;

[0105] in, Represents a transaction tuple, represents the transaction volume, Represents the transaction price, represents the trading time;

[0106] Median filtering is performed on the preliminary transaction data set to eliminate outliers using the formula:

[0107]

[0108] Generate a cleaned transaction dataset;

[0109] in, represents the filtered transaction volume, Represents the window size;

[0110] Apply exponential smoothing to the cleaned transaction data set using the formula:

[0111]

[0112] Generate a volume baseline;

[0113] in, represents the smoothed transaction volume, Represents the smoothing coefficient.

[0114] formula:

[0115] The formula is used to represent the transaction data received from the blockchain node, where It is A data tuple of a transaction.

[0116] in:

[0117] : Trading volume

[0118] : Transaction price

[0119] : Trading time

[0120] Derivation process:

[0121] Assume that the first transaction record received is unit, USD, . The transaction record is encapsulated into a tuple .

[0122] formula:

[0123] The formula is used to apply a median filter with the goal of removing outliers and smoothing the data.

[0124] in:

[0125] : Filtered transaction volume

[0126] : The size of the filter window, usually a small integer such as 1 or 2.

[0127] Derivation process:

[0128] Assumptions and three consecutive trading volumes ,in is an obvious outlier. Applying the median filter:

[0129]

[0130] In this example, It is the filtered trading volume, and the value can better reflect the normal trading level.

[0131] formula:

[0132] This formula is an exponential smoothing method used to iteratively smooth the data and establish a baseline for transaction data.

[0133] in:

[0134] Smoothed transaction volume

[0135] : Smoothing coefficient, between 0 and 1, such as 0.1 or 0.2.

[0136] Derivation process:

[0137] Assumptions and the previous smoothed volume , the current filtered transaction volume :

[0138] In this example, Represents the smoothed trading volume, which is closer to the historical trend.

[0139] See also Figure 3 The specific steps to monitor and record the increase and decrease in transaction frequency, the law of price fluctuations and the periodicity of supply and demand changes are:

[0140] Calculate the transaction data baseline The increase or decrease in transaction frequency , price fluctuations and supply and demand cyclicality , using the formula:

[0141]

[0142] and

[0143]

[0144] as well as

[0145]

[0146] Generate transaction feature change dataset;

[0147] in, Indicates the increase or decrease in transaction frequency. represents the percentage change in price, A periodic function representing changes in supply and demand, is the time variable, is the cycle length, Adjust the nonlinear response of transaction frequency change data to increase sensitivity to small changes. Adjusting for non-linear responses to price change data, Adjust the nonlinear response of supply and demand cycles and assist the model in identifying key change points in cyclical fluctuations;

[0148] The weighted combination method is used to analyze the transaction feature change data set and calculate the influencing factors using the formula:

[0149]

[0150] Generate transaction behavior scores;

[0151] in, is the weight of the transaction frequency change, which determines the impact of the transaction frequency change on the total score. is the weight of price fluctuation, specifying the proportion of price change contribution to the total score, The weight of supply and demand cyclicality adjusts the importance of supply and demand changes in the overall model;

[0152] The significance analysis is performed based on the transaction behavior score, using the threshold determination method, and the formula is:

[0153]

[0154] Generate significance determination results of market behavior;

[0155] in, Represents the significant classification result, is the decision threshold.

[0156] formula:

[0157]

[0158] in:

[0159] and It is the baseline value of transaction data at two consecutive moments.

[0160] It is a sensitivity parameter for adjusting the nonlinear response, which is used to adjust the model's response to small changes in transaction frequency.

[0161] Calculation example:

[0162] Assumptions Trading and Trading, Settings :

[0163]

[0164] in Indicates from time arrive The net value of the increase or decrease in transaction frequency.

[0165] formula:

[0166]

[0167] in:

[0168] It is a weight parameter of transaction frequency, price volatility and supply and demand cyclicality.

[0169] It is a calculated indicator of transaction frequency, price volatility and supply and demand cyclicality.

[0170] Calculation example:

[0171] Assumptions , and it is known that , and the assumption and :

[0172]

[0173] in Indicates a score that reflects the dynamics of trading behavior.

[0174] formula:

[0175]

[0176] in:

[0177] is the threshold used to determine significance.

[0178] Calculation example:

[0179] Assumptions , known :

[0180]

[0181] in Indicates that under the current scoring conditions, the market behavior is significant.

[0182] See also Figure 4 , the steps to analyze the cause and effect of market behavior are as follows:

[0183] Map the significance judgment results of market behavior to the differentiation nodes of the decision tree using the formula:

[0184]

[0185] Generate decision tree node mapping results;

[0186] in, represents a decision tree node, Represents the results of the significance determination of market behavior, Represents the market status index, Represents environment variables, Represents historical behavior data;

[0187] Analyzing decision tree nodes Interactions and path dependencies between nodes, select two and For analysis, the formula is:

[0188]

[0189] Generate quantitative evaluation results of the relationships between nodes;

[0190] in, Represents the strength of the relationship between nodes. and Represents a differentiated decision tree node, Adjustment Node The impact on relationship strength, Adjustment Node The impact on relationship strength, The effect of moderating node differences on tie strength;

[0191] Quantitative evaluation results based on the relationship between nodes To conduct causal analysis, the formula is used:

[0192]

[0193] Generate causal analysis results;

[0194] in, Represents the result of causal relationship analysis, Represents the strength of the relationship between nodes. The influence of the square term of moderated relationship strength, The effect of the square root of moderated relationship strength.

[0195] formula:

[0196]

[0197] Detailed explanation and examples:

[0198] : The significance determination result of market behavior. The assumed value is 1, indicating high significance.

[0199] : Market status index. The assumed value is 0.8, which means the current market status is relatively active.

[0200] : Environmental variable. The assumed value is 0.5, reflecting a moderate external environmental impact.

[0201] : Historical behavior data. The assumed value is 0.7, indicating a high consistency of behavior obtained from historical data.

[0202] Calculation process:

[0203] Assume that the mapping function map is calculated by weighted summation ,So:

[0204]

[0205] The value (0.82) indicates that the node of the decision tree is a significant market behavior in a high activity state.

[0206] formula:

[0207]

[0208] Detailed explanation and examples:

[0209] : Adjustment coefficient, assumed to be 1.

[0210] : Assumption and .

[0211] Calculation process:

[0212]

[0213] The value (1.637) indicates the strength of the relationship between two nodes, with higher values ​​indicating stronger associations.

[0214] formula:

[0215]

[0216] Detailed explanation and examples:

[0217] : Adjustment coefficient, assumed to be 1.

[0218] : Assumption .

[0219] Calculation process:

[0220]

[0221] The value (3.96) represents the analysis result of causal relationship, and higher values ​​indicate stronger causal relationship.

[0222] See also Figure 5 , the specific steps for obtaining market dynamics forecast results are:

[0223] Combined with the market behavior insights, the new depth of the decision tree is calculated using the formula:

[0224]

[0225] Generate the current tree depth;

[0226] in, Represents the current tree depth, Represents the original tree depth, is the depth adjustment factor, Indicates the current market trend factor, Adjust thresholds for market trends;

[0227] According to market fluctuations, the updated node splitting criteria are calculated using the formula:

[0228]

[0229] Generate the current node splitting standard;

[0230] in, Represents the current node splitting standard, represents the original split standard, is the split standard adjustment factor, is the current market volatility index, is the fluctuation baseline;

[0231] Integrate the current tree depth and the current node splitting criteria, adjust the decision tree model structure, and use the formula:

[0232]

[0233] Generate an adjusted decision tree model;

[0234] in, represents the adjusted decision tree model, is the current tree depth, is the current node splitting standard;

[0235] Use the adjusted decision tree model to make detailed predictions on future market dynamics using the formula:

[0236]

[0237] Generate market dynamics forecast results;

[0238] in, Indicates the market dynamics forecast results, is the adjusted decision tree model, and data represents the market data used for prediction.

[0239] Assume that the depth of the current decision tree is =10, market trend factor 1.2, market trend adjustment threshold is 1.0, depth adjustment factor is 0.5.

[0240] formula

[0241]

[0242] calculate

[0243]

[0244] in Indicates that the adjusted tree depth has increased slightly, reflecting a small response to market trends.

[0245] Assume that the current node splitting standard is 0.05, the market volatility index is 0.8, fluctuating baseline The split standard adjustment coefficient is 0.7. is 2.

[0246] formula

[0247]

[0248] calculate

[0249]

[0250]

[0251] in It indicates that the current node splitting criteria has been raised, indicating that the model has become more sensitive to small changes in the data, matching the increased market volatility.

[0252] See also Figure 6 , the specific steps for sending transaction risk notifications are:

[0253] According to the market dynamics forecast results, the risk level in the current market environment is calculated using the formula:

[0254]

[0255] generating risk level assessment values;

[0256] in, Represents the risk level assessment value, represents the risk sensitivity coefficient, Represents the future market forecast value, represents the benchmark market value;

[0257] Compare the risk level with the preset risk threshold using the formula:

[0258]

[0259] Generate risk comparison results;

[0260] in, represents the risk comparison result, Represents the risk level assessment value, represents the risk threshold;

[0261] Based on the risk comparison results, decide whether to send a transaction risk notification using the formula:

[0262]

[0263] Generate notification sending status. If the notification sending status is 0, no operation will be performed. If the notification sending status is 1, a transaction risk notification will be sent.

[0264] in, Represents the notification sending status. Represents the risk comparison results.

[0265] formula:

[0266]

[0267] Parameter Description:

[0268] Risk level assessment value

[0269] : Risk sensitivity coefficient, assumed to be 0.5, which is a general coefficient used to adjust the sensitivity of risk assessment.

[0270] Future market forecast

[0271] Benchmark market value

[0272] Derivation process:

[0273] Assumed benchmark market value =100, the future market forecast value is 120. Then:

[0274]

[0275] This means that according to the current parameter settings, the market risk level assessment value is 0.1, indicating that there is a certain risk level relative to the benchmark market value. risk of change.

[0276] formula:

[0277]

[0278] Parameter Description:

[0279] : Risk comparison results

[0280] : The resulting risk level

[0281] : Risk threshold, set to 0.05, represents the maximum risk threshold that the market can tolerate.

[0282] Derivation process:

[0283] Known :

[0284]

[0285] The result indicates that the current market risk exceeds the set risk threshold and therefore action is required.

[0286] formula:

[0287]

[0288] Parameter Description:

[0289] : Notification sending status, 1 means sending notification, 0 means not sending.

[0290] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. The green energy transaction data monitoring system based on blockchain is characterized by: The system comprises: The data collection and correction module extracts transaction volume, transaction price and transaction time from the transaction records received by the blockchain node, uses data cleaning to eliminate noise and outliers in the records, smooths mutation points in the data, and establishes a transaction data baseline; The market behavior analysis module monitors and records the increase and decrease of transaction frequency, the law of price fluctuations and the periodicity of supply and demand changes based on the transaction data baseline, maps the data changes to the differentiated nodes of the decision tree, analyzes the cause and effect relationship of market behavior through the interaction between differentiated nodes, and generates market behavior insights; The dynamic prediction adjustment module combines the market behavior insights to adjust the structure of the decision tree model in real time, including the tree depth and node splitting criteria, to match current market trends and fluctuations, refine the decision tree model's prediction of future market dynamics, and generate market dynamics prediction results; The risk assessment and early warning module uses the market dynamics prediction results to assess the transaction risk under the current market environment, sets a risk threshold, and sends a transaction risk notification to traders and managers when the market changes predicted by the model exceed the risk threshold.

2. The green energy transaction data monitoring system based on blockchain according to claim 1 is characterized in that: The steps for establishing the transaction data baseline are specifically as follows: Receive data streams from blockchain nodes, filter transaction information, and use the formula: Generate a preliminary transaction dataset; in, Represents a transaction tuple, represents the transaction volume, Represents the transaction price, represents the trading time; The preliminary transaction data set is subjected to median filtering to eliminate outliers, using the formula: Generate a cleaned transaction dataset; in, represents the filtered transaction volume, Represents the window size; Exponential smoothing is applied to the cleaned transaction data set using the formula: Generate a volume baseline; in, represents the smoothed transaction volume, Represents the smoothing coefficient.

3. The green energy transaction data monitoring system based on blockchain according to claim 2 is characterized in that: The steps of monitoring and recording the increase and decrease of transaction frequency, the law of price fluctuations and the periodicity of supply and demand changes are specifically as follows: Calculate the transaction data from the baseline The increase or decrease in transaction frequency , price fluctuations and supply and demand cyclicality , using the formula: and as well as Generate transaction feature change dataset; in, Indicates the increase or decrease in transaction frequency. represents the percentage change in price, A periodic function representing changes in supply and demand, is the time variable, is the cycle length, Adjust the nonlinear response of transaction frequency change data to increase sensitivity to small changes. Adjusting for non-linear responses to price change data, Adjust the nonlinear response of supply and demand cycles and assist the model in identifying key change points in cyclical fluctuations; The weighted combination method is applied to analyze the transaction feature change data set and calculate the influencing factors using the formula: Generate transaction behavior scores; in, is the weight of the transaction frequency change, which determines the impact of the transaction frequency change on the total score. is the weight of price fluctuation, specifying the proportion of price change contribution to the total score, The weight of supply and demand cyclicality adjusts the importance of supply and demand changes in the overall model; A significance analysis is performed based on the transaction behavior score, using a threshold determination method, and the formula is: Generate significance determination results of market behavior; in, Represents the significant classification result, is the decision threshold.

4. The green energy transaction data monitoring system based on blockchain according to claim 3 is characterized in that: The steps of analyzing the causal relationship of market behavior are specifically as follows: The significance determination results of the market behavior are mapped to the differentiation nodes of the decision tree using the formula: Generate decision tree node mapping results; in, represents a decision tree node, Represents the results of the significance determination of market behavior, Represents the market status index, Represents environment variables, Represents historical behavior data; Analyze the decision tree nodes Interactions and path dependencies between nodes, select two and For analysis, the formula is: Generate quantitative evaluation results of the relationships between nodes; in, Represents the strength of the relationship between nodes. and Represents a differentiated decision tree node, Adjusting Nodes The impact on relationship strength, Adjusting Nodes The impact on relationship strength, The effect of moderating node differences on tie strength; According to the quantitative evaluation results of the relationship between the nodes To conduct causal analysis, the formula is used: Generate causal analysis results; in, Represents the result of causal relationship analysis. Represents the strength of the relationship between nodes. The influence of the square term of moderated relationship strength, The effect of the square root of moderated relationship strength.

5. The green energy transaction data monitoring system based on blockchain according to claim 4 is characterized in that: The steps for obtaining the market dynamics forecast results are specifically as follows: Combined with the above market behavior insights, the new depth of the decision tree is calculated using the formula: Generate the current tree depth; in, Represents the current tree depth, represents the original tree depth, is the depth adjustment factor, Indicates the current market trend factor, Adjust thresholds for market trends; According to market fluctuations, the updated node splitting criteria are calculated using the formula: Generate the current node splitting standard; in, Represents the current node splitting standard, represents the original split standard, is the split standard adjustment factor, is the current market volatility index, is the fluctuation baseline; Integrate the current tree depth and the current node splitting criteria, adjust the decision tree model structure, and use the formula: Generate an adjusted decision tree model; in, represents the adjusted decision tree model, is the current tree depth, is the current node splitting standard; Use the adjusted decision tree model to make detailed predictions on future market dynamics using the formula: Generate market dynamics forecast results; in, Indicates the market dynamics forecast results, is the adjusted decision tree model, and data represents the market data used for prediction.

6. The green energy transaction data monitoring system based on blockchain according to claim 5 is characterized in that: The specific steps of sending the transaction risk notification are: According to the market dynamics forecast results, the risk level in the current market environment is calculated using the formula: generating risk level assessment values; in, Represents the risk level assessment value, represents the risk sensitivity coefficient, Represents the future market forecast value, represents the benchmark market value; Compare the risk level with the preset risk threshold using the formula: Generate risk comparison results; in, represents the risk comparison result, Represents the risk level assessment value, represents the risk threshold; Based on the risk comparison results, decide whether to send a transaction risk notification using the formula: Generate a notification sending status, if the notification sending status is 0, no operation is performed, if the notification sending status is 1, a transaction risk notification is sent; in, Represents the notification sending status. Represents the risk comparison results.

Images