A data mart trading platform based on power metering

Abstract

The present invention relates to the field of data transaction technology, specifically a data mart transaction platform based on electricity metering, the platform includes a data structure analysis module, an abnormal pattern positioning module, an emergency management module, a transaction pattern recognition module, a transaction flow configuration module, and a price dynamic adjustment module. In the present invention, by refining the data processing flow and enhancing transaction transparency and responsiveness, significant performance enhancement is provided, and by detailed monitoring of the seasonality and trend deviations of electricity data, the platform is allowed to identify data anomalies that do not conform to historical models in real time, and realize rapid identification and response to abnormal transactions, so that potential risk behaviors can be detected and analyzed in the early stages of formation, and by dynamically adjusting the transaction volume and path of the transaction node, the data flow is optimized to ensure that the market price of electricity data reflects the supply and demand changes in real time, improve the market's operating efficiency, enhance the market's adaptability to emergencies, and improve the rationality and economic benefits of electricity distribution.

Description

Technical Field

[0001] The present invention relates to the technical field of data transaction, and in particular to a data mart transaction platform based on power metering. Background Art

[0002] The field of data trading technology focuses on the sale and exchange of data resources through digital platforms, including data collection, processing, analysis and the construction of trading mechanisms. The field is committed to enhancing the value of data, ensuring the transparency and security of data transactions, and promoting market efficiency. Key technologies include data encryption, privacy protection, smart contracts, and the development of trading algorithms, aiming to create a reliable and efficient data market environment. The data trading platform is equipped with advanced analytical tools to support real-time data processing and optimize decision-making, enhance user experience and meet the needs of different industries.

[0003] The data mart trading platform is a data trading platform designed specifically for the power industry, allowing users to buy, sell or exchange data generated during power consumption and production. The platform's uses include promoting transparency of power data, supporting pricing mechanisms in the power market, and providing real-time data analysis to optimize power distribution and consumption. In this way, power suppliers and consumers can manage resources more efficiently, respond to market changes, and promote the implementation of renewable energy and energy efficiency projects.

[0004] Existing data mart trading platforms include a variety of security measures such as data encryption and privacy protection, and focus on improving the value of data and market efficiency, but existing technologies lack real-time data monitoring and dynamic response mechanisms for specific industries such as the power industry. Existing technologies focus more on long-term analysis and storage of data, but lack sensitivity and response speed to real-time data changes, including in an environment where demand changes rapidly in the power market. This deficiency leads to a lag in response to market fluctuations, thereby affecting the timeliness and accuracy of decision-making. Existing technologies are not flexible enough in data flow and price adjustment mechanisms, and it is difficult to quickly adapt to real-time changes in market supply and demand, resulting in uneven resource allocation, affecting the effective utilization of power resources and the overall health of the market. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a data mart trading platform based on electricity metering.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: A data mart trading platform based on power metering includes:

[0007] The data structure analysis module receives power transaction data and performs time series decomposition to distinguish seasonality, trends and random fluctuations, calculates key features of transaction data, and generates seasonal trend split results;

[0008] The abnormal pattern positioning module compares the residual fluctuations with the historical data based on the seasonal trend splitting results, identifies the transactions with excessive deviations as abnormal, monitors the changes in transaction cycle data, and generates abnormal pattern recognition results;

[0009] The emergency management module formulates a real-time response plan based on the abnormal pattern recognition results, adjusts the alarm level, outputs an alarm signal, updates monitoring parameters, and generates emergency execution detail record results;

[0010] The transaction pattern recognition module analyzes the abnormal pattern recognition results, identifies key transaction nodes and potential risk behaviors in the transaction network, analyzes transaction patterns and relationship density of transaction nodes, and generates risk behavior analysis results;

[0011] The transaction flow configuration module adjusts the data flow in the transaction network based on the risk behavior analysis results, optimizes the transaction volume and path of the transaction nodes, and generates transaction flow optimization results;

[0012] The price dynamic adjustment module uses the transaction flow optimization results to update and adjust the power data price according to market supply and demand, optimizes the pricing plan in real time, and generates a price adjustment completion record.

[0013] As a further scheme of the present invention, the seasonal trend splitting results include seasonal components, trend components, and random components; the abnormal pattern recognition results include abnormal fluctuation standards, historical comparison differences, and periodic deviation indicators; the emergency execution details record results include alarm activation status, emergency response strategy, and monitoring update data; the risk behavior analysis results include node behavior anomalies, relationship network density, and risk point identification; the transaction flow optimization results include flow adjustment strategy, node transaction optimization, and risk reduction measures; the price adjustment completion record includes adjusted price, response time efficiency, and supply and demand balance status.

[0014] As a further solution of the present invention, the data structure analysis module includes:

[0015] The time series decomposition submodule receives power transaction data, performs time series decomposition, sets initial decomposition parameters, performs seasonal, trend and random fluctuation decomposition, adjusts decomposition parameters, verifies the rationality of decomposition results, and obtains decomposition results of transaction data;

[0016] The feature calculation submodule extracts the fluctuation range of the transaction data based on the decomposition result of the transaction data, performs frequency change analysis, calculates periodic features, optimizes key features using adaptive parameters, and obtains key feature recognition results;

[0017] The periodic change analysis submodule performs seasonal trend analysis based on the key feature recognition results, splits seasonal changes, analyzes long-term trends, adjusts trend splitting parameters based on periodic features, verifies the accuracy of seasonal trends, and obtains seasonal trend splitting results.

[0018] As a further solution of the present invention, the abnormal mode positioning module includes:

[0019] The fluctuation deviation calculation submodule extracts the remaining fluctuations based on the seasonal trend splitting results, sets the initial deviation threshold, compares the deviation between the threshold and the historical data, records the deviation value and adjusts the calculation parameters, and generates the fluctuation deviation measurement results;

[0020] The abnormal index calculation submodule extracts the fluctuation value of each transaction data based on the fluctuation deviation measurement result, calculates the abnormal index, sets the abnormal range, compares the abnormal index with the preset range, identifies the data exceeding the preset range as abnormal and records it, and obtains the abnormal detection result;

[0021] Based on the anomaly detection results, the cycle monitoring submodule adopts a time series analysis method to monitor the data changes of each transaction cycle, extract the data fluctuations within the cycle, analyze the changing trends of abnormal data, record the abnormal changes, and obtain abnormal pattern recognition results.

[0022] As a further solution of the present invention, the formula of the time series analysis method is as follows:

[0023]

[0024] in, is the predicted value at time t, μ is the constant term, φ is the autoregressive coefficient, y t-1 is the measured data at time t-1, θ is the moving average coefficient, ∈ t-1 is the error term at time t-1, α1 is the regression coefficient of the two-period lagged data, x t-2 is the data value at time t-2, β1 is the coefficient of data change, Δy t-1 is the change in the measured data at time t-1.

[0025] As a further solution of the present invention, the emergency management module includes:

[0026] The real-time response plan submodule formulates a real-time response plan based on the abnormal pattern recognition results, analyzes the type and scope of abnormal data, allocates processing resources, classifies abnormal data, determines priority processing items, arranges response steps, and generates an emergency management output plan;

[0027] The alarm adjustment submodule adjusts the platform alarm level based on the emergency management output plan, sets alarm parameters, classifies the alarm level, sends alarm information if an alarm signal is triggered, records the alarm trigger time and impact range, monitors the alarm effect, and generates an alarm configuration update result;

[0028] The parameter update submodule synchronously updates the monitoring parameters of the platform based on the alarm configuration update results, sets new monitoring indicators, performs parameter calibration, detects potential risks, records updated parameters, adjusts monitoring strategies, and generates emergency execution detail record results.

[0029] As a further solution of the present invention, the transaction pattern recognition module includes:

[0030] The key node identification submodule collects transaction network data based on the abnormal pattern recognition results, extracts transaction node information, classifies transaction nodes, analyzes the interaction frequency between transaction nodes, marks key nodes by comparing the differences in transaction volumes, and generates key transaction node identification results;

[0031] The relationship density analysis submodule analyzes the transaction patterns of the transaction nodes based on the key transaction node identification results, extracts the transaction frequency and transaction amount between the transaction nodes, calculates the relationship density, and classifies the relationships, evaluates the transaction closeness between the transaction nodes, and generates relationship strength evaluation data;

[0032] The potential risk assessment submodule uses the K-means clustering algorithm based on the relationship strength assessment data to analyze potential risk behaviors in the transaction network, detect abnormalities in transaction patterns, identify risky transaction nodes, evaluate the transaction behaviors of risky nodes, reveal potential problems, record risky behavior situations, and generate risky behavior analysis results.

[0033] As a further solution of the present invention, the formula of the K-means clustering algorithm is as follows:

[0034]

[0035] Among them, J is the sum of squares of clustering errors, k is the number of cluster centers, and n is the total number of samples. is the jth sample point in the i-th cluster, μ i is the i-th cluster center, w ij For sample points The weight coefficient, λ is the regularization parameter, and c is the centroid of the data set.

[0036] As a further solution of the present invention, the transaction flow configuration module includes:

[0037] The data flow adjustment submodule adjusts the data flow direction in the transaction network based on the risk behavior analysis results, analyzes the current transaction data flow direction, determines the data flow bottleneck, sets adjustment parameters, reconfigures the data flow direction, modifies the data path allocation, and monitors the data flow to obtain network liquidity adjustment data;

[0038] The transaction volume optimization submodule optimizes the transaction volume of the transaction nodes based on the network liquidity adjustment data, analyzes the transaction volume of the transaction nodes, identifies the load nodes, and redistributes the transaction volume, adjusts the transaction load, sets optimization parameters, and records the transaction volume distribution to generate optimized transaction distribution data;

[0039] The path optimization submodule optimizes the transaction path based on the optimized transaction distribution data, analyzes the transaction path between transaction nodes, identifies redundant paths, sets path optimization parameters, and generates transaction flow optimization results by adjusting path length and flow and recording transaction path information.

[0040] As a further solution of the present invention, the price dynamic adjustment module includes:

[0041] The market supply and demand evaluation submodule evaluates the supply and demand status of the power market based on the transaction flow optimization results, collects current market supply and demand data, organizes the data, analyzes supply and demand changes, identifies supply and demand imbalances, sets evaluation parameters, verifies by adjusting the parameters, and obtains the market demand response status;

[0042] The real-time pricing scheme submodule formulates a real-time electricity pricing scheme based on the market demand response status, analyzes supply and demand data and extracts key indicators, sets pricing parameters, optimizes prices by adjusting parameters, updates market prices, records pricing scheme details, and generates pricing parameter adjustment results;

[0043] The price update monitoring submodule monitors the price adjustment process and extracts data based on the pricing parameter adjustment results, records the data of each adjustment, updates the price changes, verifies by comparing historical data, and generates a price adjustment completion record.

[0044] Compared with the prior art, the advantages and positive effects of the present invention are:

[0045] In the present invention, by refining the data processing flow and enhancing transaction transparency and responsiveness, significant performance enhancement is provided. By monitoring the seasonality and trend deviation of power data in detail, the platform is allowed to identify data anomalies that do not conform to the historical model in real time, thereby realizing rapid identification and response to abnormal transactions. This monitoring is not limited to identification, but also includes in-depth analysis of risks and combing of transaction patterns, so that any potential risk behavior can be detected and analyzed in the early stage of formation. By dynamically adjusting the transaction volume and path of the transaction node, the data flow is optimized to ensure that the market price of power data reflects the supply and demand changes in real time, improve the operating efficiency of the market, enhance the market's adaptability to emergencies, and improve the rationality and economic benefits of power distribution. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is a flow chart of the platform of the present invention;

[0047] Figure 2 It is a schematic diagram of the platform framework of the present invention;

[0048] Figure 3 It is a flow chart of the data structure analysis module of the present invention;

[0049] Figure 4 It is a flow chart of the abnormal mode positioning module of the present invention;

[0050] Figure 5 is a flow chart of the emergency management module of the present invention;

[0051] Figure 6 It is a flow chart of the transaction pattern recognition module of the present invention;

[0052] Figure 7 A flowchart of a transaction flow configuration module of the present invention;

[0053] Figure 8 This is a flow chart of the price dynamic adjustment module of the present invention. DETAILED DESCRIPTION

[0054] 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.

[0055] 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.

[0056] Embodiment 1

[0057] See also Figure 1 to Figure 2 , a data mart trading platform based on electricity metering includes:

[0058] The data structure analysis module receives power transaction data, divides the transaction data into seasonality, trend and random fluctuation through time series decomposition, calculates the key features of the transaction data, and generates seasonal trend split results;

[0059] The abnormal pattern location module uses the seasonal trend splitting results to calculate the abnormal index of trading data by comparing the deviation between the residual fluctuation and the historical data. If the deviation exceeds the preset range, it is marked as abnormal. It also uses the time series analysis method to monitor the data changes in each trading cycle and generate abnormal pattern recognition results.

[0060] The emergency management module formulates real-time response plans based on the abnormal pattern recognition results, adjusts the platform alarm level and outputs alarm signals, synchronously updates the platform's monitoring parameters and responds to potential risks, and generates emergency execution details record results;

[0061] The transaction pattern recognition module analyzes and identifies key transaction nodes and potential risk behaviors in the transaction network based on the abnormal pattern recognition results. By analyzing the transaction patterns and relationship density of key transaction nodes, it reveals potential problems of risk nodes and generates risk behavior analysis results.

[0062] The transaction flow configuration module adjusts the data flow in the transaction network based on the risk behavior analysis results, adjusts the transaction volume and path of the transaction nodes through the optimization algorithm, and generates transaction flow optimization results;

[0063] The dynamic price adjustment module uses the transaction flow optimization results to update and adjust the market price of electricity data, optimizes the pricing plan in real time according to the real-time market supply and demand status, and generates a price adjustment completion record.

[0064] The results of seasonal trend splitting include seasonal component, trend component, and random component. The results of abnormal pattern recognition include abnormal fluctuation standards, historical comparison differences, and periodic deviation indicators. The results of emergency execution details records include alarm activation status, emergency response strategy, and monitoring update data. The results of risk behavior analysis include node behavior anomalies, relationship network density, and risk point identification. The results of transaction flow optimization include flow adjustment strategy, node transaction optimization, and risk reduction measures. The price adjustment completion record includes adjusted price, response time efficiency, and supply and demand balance status.

[0065] Specifically, if Figure 2 , 3 As shown, the data structure analysis module includes:

[0066] The time series decomposition submodule receives power transaction data, performs time series decomposition, sets initial decomposition parameters, performs seasonal, trend and random fluctuation decomposition, adjusts decomposition parameters, verifies the rationality of decomposition results, and obtains the decomposition results of transaction data. The execution process is as follows;

[0067] The time series decomposition submodule first receives the power trading data and sets the initial decomposition parameters based on the preliminary analysis of the data, including seasonal cycle, trend window length, etc. Statistical methods are used to determine the seasonality, trend and random fluctuation components in the data, and a decomposition technique based on wavelet transform is used to improve the sensitivity to seasonality and trend by adjusting the wavelet coefficients. This method can explore the implicit patterns in the data in more detail. The decomposition parameters are adjusted at each step according to the principle of error minimization to ensure that the final decomposition results can truly reflect the intrinsic characteristics of the data. After multiple rounds of verification, the rationality of the decomposition results is confirmed, and the decomposition results of the power trading data are finally determined. The formula is as follows:

[0068] D t =β·(Y t -α·S t-p )+γ·∈ t

[0069] Among them, D t represents the decomposition result at time point t, Y t is the original transaction data, α and β are adaptive adjustment coefficients, optimized by data-driven methods, S t-p represents the seasonal component before p periods, γ is the coefficient used to adjust the impact of random fluctuations, ∈ t is a random error term.

[0070] Based on the decomposition results of transaction data, the feature calculation submodule extracts the fluctuation range of transaction data, performs frequency change analysis, calculates periodic features, and uses adaptive parameters to optimize key features. The execution process of obtaining key feature identification results is as follows;

[0071] The feature calculation submodule is based on the results of time series decomposition. The feature calculation submodule analyzes the fluctuation amplitude and frequency changes of the data and calculates the periodic characteristics of the data. The main periodic components are identified through dynamic spectrum analysis, and the weights of key features are adjusted using machine learning technology to optimize the prediction ability of the model. Adaptive parameters such as learning rate and regularization factor are used to optimize key feature extraction to adapt to the changes and complexity of the data, and the key feature identification results are output. The formula used is as follows:

[0072]

[0073] Among them, C k is the calculated kth eigenvalue, λ i is the weight of the ith data point, X i is the decomposed data value, μ is the mean of the data, σ is the standard deviation, ω i is the frequency variation parameter obtained through data analysis, t i is the time point of the time series.

[0074] The periodic change analysis submodule performs seasonal trend analysis based on the key feature recognition results, splits seasonal changes, analyzes long-term trends, adjusts trend splitting parameters based on periodic features, verifies the accuracy of seasonal trends, and obtains the execution process of seasonal trend splitting results as follows;

[0075] The periodic change analysis submodule conducts in-depth analysis of seasonal trends and long-term trends based on the key feature identification results. Seasonal changes are split through the mathematical model of the target, and long-term trends are identified and verified through regression analysis. The periodic features are combined with the trend splitting parameters, and the model parameters are optimized through an iterative method to ensure the accuracy of the seasonal trend. The seasonal trend splitting results are obtained using the following formula:

[0076]

[0077] Among them, T t is the seasonal trend analysis result at time t, κ j and δ j is the amplitude adjustment parameter obtained by data fitting, f j is the frequency parameter for splitting seasonal changes, and m is the number of periodic functions, which is dynamically adjusted according to the complexity of the data.

[0078] Specifically, if Figure 2 , 4 As shown, the abnormal mode positioning module includes:

[0079] The fluctuation deviation calculation submodule extracts the remaining fluctuations based on the seasonal trend splitting results, sets the initial deviation threshold, compares the deviation between the threshold and the historical data, records the deviation value and adjusts the calculation parameters, and generates the fluctuation deviation measurement results. The execution process is as follows;

[0080] The volatility deviation calculation submodule extracts the remaining volatility based on the seasonal trend splitting results. By setting the initial deviation threshold, which is calculated based on the standard deviation and average deviation of the historical volatility data, the current data is compared with the threshold, and the deviation value at each time point is recorded. The calculation parameters are adjusted according to the deviation size, such as adjusting the sliding window size and weight, to optimize future deviation detection and generate volatility deviation measurement results. The formula is as follows:

[0081]

[0082] Among them, B t is the fluctuation deviation measurement result at time point t, ΔX t is the deviation between current and historical data, θ is the deviation threshold, σ is the standard deviation of historical deviations, and K is the adjustment coefficient.

[0083] The abnormal index calculation submodule extracts the fluctuation value of each transaction data based on the fluctuation deviation measurement results, calculates the abnormal index, sets the abnormal range, compares the abnormal index with the preset range, identifies the data exceeding the preset range as abnormal and records it, and obtains the execution process of the abnormal detection result as follows;

[0084] The abnormal index calculation submodule extracts the fluctuation value of each transaction data point and calculates the abnormal index based on the fluctuation deviation measurement results, sets the abnormal range, and analyzes the data fluctuation based on multiple cycles. The calculated abnormal index is compared with the preset range. Data points that exceed the range will be marked as abnormal and recorded. The abnormal detection result is obtained with the following formula:

[0085]

[0086] Among them, A i is the abnormal indicator of transaction data point i; W i is the fluctuation value at that point, W avg is the average value of the fluctuations, W std is the standard deviation of the fluctuation value, and Z is the scaling factor used to adjust the sensitivity of anomaly identification.

[0087] Based on the anomaly detection results, the cycle monitoring submodule uses the time series analysis method to monitor the data changes of each transaction cycle, extract the data fluctuations within the cycle, analyze the change trend of abnormal data, record the abnormal changes, and obtain the abnormal pattern recognition results. The execution process is as follows;

[0088] The cycle monitoring submodule monitors the data changes within each transaction cycle, including the changing trends of abnormal data. By analyzing the fluctuation pattern of the data within the cycle, it records and identifies abnormal changes and obtains the abnormal pattern recognition results.

[0089] The formula for the time series analysis method is as follows:

[0090]

[0091] in, is the predicted value at time t, μ is the constant term, φ is the autoregressive coefficient, y t-1 is the measured data at time t-1, θ is the moving average coefficient, ∈ t-1 is the error term at time t-1, α1 is the regression coefficient of the two-period lagged data, x t-2 is the data value at time t-2, β1 is the coefficient of data change, Δy t-1 is the change in the measured data at time t-1.

[0092] The execution process is as follows:

[0093] Determine the constant term μ and initialize it, select an appropriate autoregressive coefficient φ to reflect the influence of the previous period's data, and calculate the measured data y at time t-1 t-1 Substitute it into the formula, select the moving average coefficient θ to calculate the error term ∈ at time t-1 t-1 Substitute it into the formula, and then determine the regression coefficient α1 of the two-period lagged data to reflect the impact of the data value at time t-2, and obtain the data value x at time t-2 t-2 Substitute it into the formula, select the coefficient β1 of the data change to reflect the impact of data change, and calculate the change Δy of the measured data at time t-1 t-1 Substitute it into the formula and finally get the predicted value of time t through weighted calculation

[0094] Specifically, if Figure 2 , 5 As shown, the emergency management module includes:

[0095] The real-time response plan submodule formulates a real-time response plan based on the abnormal pattern recognition results, analyzes the type and scope of abnormal data, allocates processing resources, classifies abnormal data, determines priority processing items, and arranges response steps to generate the execution process of the emergency management output plan as follows;

[0096] The real-time response plan submodule analyzes the type and range of abnormal data based on the abnormal pattern recognition results, and allocates corresponding processing resources for different abnormal types. The abnormal data is classified and processed, and the high-priority processing items are determined. The response steps are arranged according to the determined priorities, and the emergency management output plan is generated to ensure a fast and effective response. The formula is as follows:

[0097]

[0098] Among them, R t represents the real-time response resource allocation ratio at time t, V i is the severity score of the i-th abnormality, P i is the processing priority of this type of exception, T i is the amount of available processing resources.

[0099] The alarm adjustment submodule adjusts the platform alarm level based on the emergency management output plan, sets alarm parameters, classifies the alarm level, sends alarm information if an alarm signal is triggered, records the alarm trigger time and impact range, monitors the alarm effect, and generates the alarm configuration update result. The execution process is as follows;

[0100] The alarm adjustment submodule is based on the emergency management output plan. This module adjusts the platform's alarm level, sets alarm parameters, and classifies according to different alarm levels. If an alarm signal is triggered, the platform will send an alarm message and record the alarm trigger time and impact range. At the same time, it monitors the effect of the alarm to optimize future alarm configurations, generate alarm configuration update results, and dynamically adjust the alarm level through the following formula:

[0101]

[0102] Among them, L t is the alarm level at time t, α is an adjustment factor used to balance the alarm sensitivity, S j is the severity of the jth alarm, U j is the urgency assessment of the alert for the platform.

[0103] The parameter update submodule updates the monitoring parameters of the platform based on the alarm configuration update results, sets new monitoring indicators, performs parameter calibration, detects potential risks, and records the updated parameters, adjusts the monitoring strategy, and generates the emergency execution details record results. The execution process is as follows;

[0104] The parameter update submodule is responsible for synchronously updating the platform's monitoring parameters, setting new monitoring indicators and calibrating parameters. It detects potential risks, records updated parameters, adjusts monitoring strategies, and ensures that the platform's monitoring can respond to future anomalies or risks in a timely manner, and generates emergency execution details to record results, using the following formula:

[0105]

[0106] Among them, K t is the parameter calibration result at time t, β k is the adjustment weight of the kth parameter, D k is the monitoring data point, γ k is the risk adjustment factor, O k Is the old monitoring indicator value.

[0107] Specifically, if Figure 2 , 6 As shown, the transaction pattern recognition module includes:

[0108] The key node identification submodule collects transaction network data based on the abnormal pattern recognition results, extracts transaction node information, classifies transaction nodes, analyzes the interaction frequency between transaction nodes, and marks key nodes by comparing the differences in transaction volumes. The execution process of generating key transaction node identification results is as follows;

[0109] The key node identification submodule collects and analyzes transaction network data based on the abnormal pattern recognition results, extracts transaction node information, classifies nodes, analyzes the interaction frequency between transaction nodes, and marks key nodes by comparing the transaction volume differences of each node. It generates key transaction node identification results through the following formula:

[0110]

[0111] Among them, N i is the criticality score of the ith transaction node, V ij is the transaction volume of node i to node j, is the average transaction volume of all nodes to node j, σ j is the standard deviation of trading volume, which is used to assess the dispersion of trading volume.

[0112] The relationship density analysis submodule analyzes the transaction patterns of transaction nodes based on the key transaction node identification results, extracts the transaction frequency and transaction amount between transaction nodes, calculates the relationship density, and classifies the relationships to evaluate the transaction closeness between transaction nodes. The execution process of generating relationship strength evaluation data is as follows;

[0113] The relationship density analysis submodule analyzes the transaction patterns between transaction nodes based on the key transaction node identification results, extracts the transaction frequency and transaction amount between nodes, calculates the relationship density, involves relationship classification, and evaluates the closeness between transaction nodes to generate relationship strength evaluation data. The calculation formula for relationship density is:

[0114]

[0115] Among them, D ij is the relationship density between node i and node j, F ij is the transaction frequency, M ij is the transaction amount, M max is the maximum amount of all transactions during the observation period, used to normalize the impact of transaction amounts.

[0116] The potential risk assessment submodule uses the K-means clustering algorithm based on the relationship strength assessment data to analyze the potential risk behaviors in the transaction network, detect abnormalities in the transaction pattern, identify risky transaction nodes, evaluate the transaction behaviors of risky nodes, reveal potential problems, record risky behaviors, and generate risky behavior analysis results. The execution process is as follows;

[0117] The potential risk assessment submodule uses the K-means clustering algorithm to analyze potential risk behaviors in the transaction network based on the relationship strength assessment data. It detects abnormalities in transaction patterns, identifies risky transaction nodes, evaluates the transaction behaviors of these nodes, reveals potential problems, records risky behaviors, and generates risky behavior analysis results.

[0118] The formula of K-means clustering algorithm is as follows:

[0119]

[0120] Among them, J is the sum of squares of clustering errors, k is the number of cluster centers, and n is the total number of samples. is the jth sample point in the i-th cluster, μ i is the i-th cluster center, w ij For sample points The weight coefficient, λ is the regularization parameter, and c is the centroid of the data set.

[0121] The execution process is as follows:

[0122] Determine the number of cluster centers k and the total number of samples n, and calculate each sample point To its cluster center μ i The sum of the distance squares is used to calculate the weight coefficient w of each sample point ij , use the contribution of sample points to the cluster center to adjust the distance sum of squares, then introduce the regularization parameter λ to control the distribution of cluster centers and calculate the cluster center μ i The sum of the squares of the distances to the centroid c of the entire data set is calculated by the weight coefficient w ij and regularization parameter λ to optimize the clustering process, weight coefficient w ijBased on the similarity calculation of the sample points, the regularization parameter λ is determined through cross-validation, the minimized clustering error value J is calculated, and the minimized clustering error value is generated.

[0123] Specifically, if Figure 2 , 7 As shown, the transaction flow configuration module includes:

[0124] The data flow adjustment submodule adjusts the data flow in the trading network based on the risk behavior analysis results, analyzes the current transaction data flow, determines the data flow bottleneck, sets adjustment parameters, reconfigures the data flow, modifies the data path allocation, and monitors the data flow to obtain the execution process of network liquidity adjustment data as follows;

[0125] The data flow adjustment submodule adjusts the data flow in the trading network based on the results of risk behavior analysis. First, analyze the current transaction data flow, identify the bottleneck area in the data flow, set adjustment parameters such as traffic capacity and priority, and reconfigure the data flow to eliminate bottlenecks and optimize overall network performance. Data flow is tracked through real-time monitoring to ensure the effectiveness of the adjustment, and the network liquidity adjustment data is obtained through the following formula:

[0126]

[0127] Among them, F ij is the adjusted flow from node i to node j, C i j is the capacity of the current connection, L ij is the current load, C max is the maximum capacity of the network.

[0128] The transaction volume optimization submodule optimizes the transaction volume of transaction nodes based on the network liquidity adjustment data, analyzes the transaction volume of transaction nodes, identifies load nodes, and redistributes the transaction volume, adjusts the transaction load, sets optimization parameters, and records the transaction volume distribution. The execution process of generating optimized transaction distribution data is as follows;

[0129] The transaction volume optimization submodule optimizes the transaction volume of transaction nodes based on the network liquidity adjustment data, analyzes the transaction volume of each node, identifies nodes with excessive load, and redistributes the transaction volume to balance the load of the entire network. It sets optimization parameters such as transaction redirection ratio and load balancing coefficient to optimize the transaction volume distribution, and finally records the optimized transaction distribution to generate optimized transaction distribution data. The optimization formula is:

[0130]

[0131] Among them, Q i is the optimized transaction volume of node i, Q avgis the average transaction volume of all nodes, Q k is the transaction volume of other nodes, and n is the total number of nodes.

[0132] The path optimization submodule optimizes the transaction path based on the optimized transaction distribution data, analyzes the transaction path between transaction nodes, identifies redundant paths, sets path optimization parameters, adjusts the path length and flow, and records the transaction path information to generate the transaction flow optimization results. The execution process is as follows;

[0133] The path optimization submodule optimizes the transaction path based on the optimized transaction distribution data, analyzes the transaction path between transaction nodes, and identifies and eliminates redundant paths. Set path optimization parameters, such as path shortest weight and flow distribution strategy, optimize transaction flow by adjusting path length and flow, record transaction path information, and generate transaction flow optimization results. The path optimization formula is as follows:

[0134]

[0135] Among them, P ij is the optimized path from node i to node j, α k is the optimization weight of path k, d ijk is the distance from i to j on path k, D max is the maximum allowed path length.

[0136] Specifically, if Figure 2 , 8 As shown, the price dynamic adjustment module includes:

[0137] The market supply and demand evaluation submodule evaluates the supply and demand status of the power market based on the transaction flow optimization results, collects the current market supply and demand data, organizes the data, analyzes the supply and demand changes, identifies the supply and demand imbalance, sets the evaluation parameters, verifies by adjusting the parameters, and obtains the execution process of the market demand response status as follows;

[0138] The market supply and demand evaluation submodule evaluates the supply and demand status of the power market based on the transaction flow optimization results. First, the supply and demand data of the current market are collected and sorted out, and the areas of supply and demand imbalance are identified by analyzing the data changes. The evaluation parameters, such as the supply and demand elasticity coefficient, are set to analyze the market's ability to respond to price changes. The supply and demand balance status is verified by adjusting the parameters, and the market demand response status is obtained. The formula is as follows:

[0139]

[0140] Among them, S t is the market supply and demand assessment result, D i is the demand in the ith region, S iis the supply in the ith region, n is the number of evaluation regions, and β is the supply-demand balance adjustment coefficient.

[0141] The real-time pricing scheme submodule formulates the real-time power pricing scheme based on the market demand response status, analyzes the supply and demand data and extracts key indicators, sets pricing parameters, optimizes prices by adjusting parameters, updates market prices, records pricing scheme details, and generates pricing parameter adjustment results. The execution process is as follows;

[0142] The real-time pricing scheme submodule formulates a real-time electricity pricing scheme based on the market demand response status, analyzes supply and demand data and extracts key indicators such as demand peak and base load, sets pricing parameters, including price elasticity and supply and demand balance coefficient, and optimizes prices by adjusting parameters to respond to market changes. The details of the pricing scheme are recorded and the pricing parameter adjustment results are generated. The formula is as follows:

[0143]

[0144] Among them, P t is the adjusted price, P0 is the base price, D t is the current total demand, S t is the current total supply, and α is the price adjustment factor that determines the sensitivity of price to changes in supply and demand.

[0145] The price update monitoring submodule monitors the price adjustment process and extracts data based on the pricing parameter adjustment results, records the data of each adjustment, updates the price changes, verifies by comparing historical data, and generates the execution process of price adjustment completion records as follows;

[0146] The price update monitoring submodule is responsible for monitoring the price adjustment process and recording the detailed data of each price adjustment. It verifies the validity of the price adjustment by analyzing the updated price data and comparing it with the historical price data. It generates the price adjustment completion record through quantitative analysis and calculation of the data. The formula used is as follows:

[0147]

[0148] Where ΔP is the percentage change in price, P new is the new adjusted price, P old is the price before adjustment.

[0149] 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. A data mart trading platform based on electricity metering, characterized in that: The platform includes: The data structure analysis module receives power transaction data and performs time series decomposition to distinguish seasonality, trends and random fluctuations, calculates key features of transaction data, and generates seasonal trend split results; The data structure analysis module includes: The time series decomposition submodule receives power transaction data, performs time series decomposition, sets initial decomposition parameters, performs seasonal, trend and random fluctuation decomposition, adjusts decomposition parameters, verifies the rationality of decomposition results, and obtains decomposition results of transaction data; The feature calculation submodule extracts the fluctuation range of the transaction data based on the decomposition result of the transaction data, performs frequency change analysis, calculates periodic features, optimizes key features using adaptive parameters, and obtains key feature recognition results; The periodic change analysis submodule performs seasonal trend analysis based on the key feature recognition results, splits seasonal changes, analyzes long-term trends, adjusts trend splitting parameters based on periodic features, verifies the accuracy of seasonal trends, and obtains seasonal trend splitting results; The abnormal pattern positioning module compares the residual fluctuations with the historical data based on the seasonal trend splitting results, identifies the transactions with excessive deviations as abnormal, monitors the changes in transaction cycle data, and generates abnormal pattern recognition results; The abnormal mode positioning module includes: The fluctuation deviation calculation submodule extracts the remaining fluctuations based on the seasonal trend splitting results, sets the initial deviation threshold, compares the deviation between the threshold and the historical data, records the deviation value and adjusts the calculation parameters, and generates the fluctuation deviation measurement results; The abnormal index calculation submodule extracts the fluctuation value of each transaction data based on the fluctuation deviation measurement result, calculates the abnormal index, sets the abnormal range, compares the abnormal index with the preset range, identifies the data exceeding the preset range as abnormal and records it, and obtains the abnormal detection result; Based on the anomaly detection results, the cycle monitoring submodule uses a time series analysis method to monitor the data changes of each transaction cycle, extract the data fluctuations within the cycle, analyze the change trend of abnormal data, record the abnormal changes, and obtain the abnormal pattern recognition results; The formula of the time series analysis method is as follows: in, is the predicted value at time t, μ is the constant term, φ is the autoregressive coefficient, y t-1 is the measured data at time t-1, θ is the moving average coefficient, ∈ t-1 is the error term at time t-1, α1 is the regression coefficient of the two-period lagged data, x t-2 is the data value at time t-2, β1 is the coefficient of data change, Δy t-1 is the change in the measured data at time t-1; The emergency management module formulates a real-time response plan based on the abnormal pattern recognition results, adjusts the alarm level, outputs an alarm signal, updates monitoring parameters, and generates emergency execution detail record results; The emergency management module includes: The real-time response plan submodule formulates a real-time response plan based on the abnormal pattern recognition results, analyzes the type and scope of abnormal data, allocates processing resources, classifies abnormal data, determines priority processing items, arranges response steps, and generates an emergency management output plan; The alarm adjustment submodule adjusts the platform alarm level based on the emergency management output plan, sets alarm parameters, classifies the alarm level, sends alarm information if an alarm signal is triggered, records the alarm trigger time and impact range, monitors the alarm effect, and generates an alarm configuration update result; The parameter update submodule synchronously updates the monitoring parameters of the platform based on the alarm configuration update result, sets new monitoring indicators, performs parameter calibration, detects potential risks, and records the updated parameters, adjusts the monitoring strategy, and generates emergency execution detail record results; The transaction pattern recognition module analyzes the abnormal pattern recognition results, identifies key transaction nodes and potential risk behaviors in the transaction network, analyzes transaction patterns and relationship density of transaction nodes, and generates risk behavior analysis results; The transaction pattern recognition module includes: The key node identification submodule collects transaction network data based on the abnormal pattern recognition results, extracts transaction node information, classifies transaction nodes, analyzes the interaction frequency between transaction nodes, marks key nodes by comparing the differences in transaction volumes, and generates key transaction node identification results; The relationship density analysis submodule analyzes the transaction patterns of the transaction nodes based on the key transaction node identification results, extracts the transaction frequency and transaction amount between the transaction nodes, calculates the relationship density, and classifies the relationships, evaluates the transaction closeness between the transaction nodes, and generates relationship strength evaluation data; The potential risk assessment submodule uses the K-means clustering algorithm based on the relationship strength assessment data to analyze potential risk behaviors in the transaction network, detect abnormalities in the transaction pattern, identify risky transaction nodes, evaluate the transaction behaviors of risky nodes, reveal potential problems, record risky behaviors, and generate risky behavior analysis results; The formula of the K-means clustering algorithm is as follows: Among them, J is the sum of squares of clustering errors, k is the number of cluster centers, and n is the total number of samples. is the jth sample point in the i-th cluster, μ i is the i-th cluster center, w ij For sample points The weight coefficient, λ is the regularization parameter, and c is the centroid of the data set; The transaction flow configuration module adjusts the data flow in the transaction network based on the risk behavior analysis results, optimizes the transaction volume and path of the transaction nodes, and generates transaction flow optimization results; The transaction flow configuration module includes: The data flow adjustment submodule adjusts the data flow direction in the transaction network based on the risk behavior analysis results, analyzes the current transaction data flow direction, determines the data flow bottleneck, sets adjustment parameters, reconfigures the data flow direction, modifies the data path allocation, and monitors the data flow to obtain network liquidity adjustment data; The transaction volume optimization submodule optimizes the transaction volume of the transaction nodes based on the network liquidity adjustment data, analyzes the transaction volume of the transaction nodes, identifies the load nodes, and redistributes the transaction volume, adjusts the transaction load, sets optimization parameters, and records the transaction volume distribution to generate optimized transaction distribution data; The path optimization submodule optimizes the transaction path based on the optimized transaction distribution data, analyzes the transaction path between transaction nodes, identifies redundant paths, sets path optimization parameters, adjusts the path length and flow, and records the transaction path information to generate transaction flow optimization results; The price dynamic adjustment module uses the transaction flow optimization results to update and adjust the power data price according to market supply and demand, optimizes the pricing plan in real time, and generates a price adjustment completion record; The price dynamic adjustment module includes: The market supply and demand evaluation submodule evaluates the supply and demand status of the power market based on the transaction flow optimization results, collects current market supply and demand data, organizes the data, analyzes supply and demand changes, identifies supply and demand imbalances, sets evaluation parameters, verifies by adjusting the parameters, and obtains the market demand response status; The real-time pricing scheme submodule formulates a real-time electricity pricing scheme based on the market demand response status, analyzes supply and demand data and extracts key indicators, sets pricing parameters, optimizes prices by adjusting parameters, updates market prices, records pricing scheme details, and generates pricing parameter adjustment results; The price update monitoring submodule monitors the price adjustment process and extracts data based on the pricing parameter adjustment results, records the data of each adjustment, updates the price changes, verifies by comparing historical data, and generates a price adjustment completion record.

2. The data mart trading platform based on electricity metering according to claim 1 is characterized by: The seasonal trend splitting results include seasonal components, trend components, and random components; the abnormal pattern recognition results include abnormal fluctuation standards, historical comparison differences, and periodic deviation indicators; the emergency execution detail record results include alarm activation status, emergency response strategies, and monitoring update data; the risk behavior analysis results include node behavior anomalies, relationship network density, and risk point identification; the transaction flow optimization results include flow adjustment strategies, node transaction optimization, and risk reduction measures; the price adjustment completion record includes adjusted prices, response time efficiency, and supply and demand balance status.

Images