Blockchain-based secure storage system for electric energy metering data

Abstract

The application relates to the technical field of electric energy data security, and discloses an electric energy metering data security storage system based on a blockchain. The system comprises five modules. An electric energy data acquisition module relies on partition electric energy metering equipment, counts metering data and equipment states, matches demand to calculate a difference value, and generates an original metering data set. A data feature analysis module analyzes electric energy consumption feature trends, sorts feature priorities, calculates related weights, and generates a data feature parameter set. A blockchain security verification module applies a consensus protocol, calibrates tampering risks and verification conflicts, adjusts verification intervals to generate a security verification result. A storage optimization module combines the result, counts storage nodes and data distribution, matches resource utilization rate and storage cycle, recombines resources to generate a storage optimization scheme. A real-time monitoring module monitors a network and data state based on the scheme, calculates a deviation, counts resources, generates a monitoring response result, and guarantees that data storage is safe and efficient.

Description

Technical Field

[0001] This invention relates to the field of power data security technology, specifically a blockchain-based power metering data security storage system. Background Technology

[0002] During the operation of the power system, electricity metering data serves as a crucial basis for power dispatch, electricity billing, and electricity consumption analysis. The security and integrity of its storage directly impact the stable development of the power industry and the vital interests of users. With the power system's transformation towards intelligence and digitalization, the number of electricity metering devices has increased significantly, resulting in massive amounts of metering data generated in real time. Traditional electricity metering data storage methods are gradually revealing numerous problems.

[0003] Traditional electricity metering data storage often employs a centralized storage architecture, storing large amounts of metering data on a few servers or databases. With this storage method, hardware failures, software vulnerabilities, network attacks, or human error on the storage servers can lead to the loss, corruption, or even tampering of significant amounts of metering data. Problems with electricity metering data not only affect the accuracy of power companies' statistics on user electricity consumption, leading to discrepancies in electricity billing, but can also interfere with power dispatching departments' assessment of power supply and demand balance, impacting the stable operation of the power system.

[0004] In traditional storage models, data verification often relies on manual checks or single software verification mechanisms, resulting in low efficiency and difficulty in comprehensively detecting anomalies or tampering traces in the data. With the continuous growth of measurement data, the workload of manual checks increases dramatically, easily leading to omissions; single software verification mechanisms also have limitations, unable to cope with complex tampering methods, and failing to guarantee the authenticity and reliability of the data. Furthermore, traditional storage systems lack flexibility in storage resource allocation, often allocating resources according to fixed patterns, failing to dynamically adjust storage resources based on actual data storage needs, usage frequency, storage cycles, etc. This leads to some storage resources being idle, while some data cannot be effectively preserved due to insufficient storage resources, resulting in wasted storage resources and affecting the overall efficiency of data storage.

[0005] Traditional storage systems have weak monitoring capabilities for the data storage process. They cannot keep track of changes in the blockchain network environment and the data operation status in real time, and cannot detect anomalies in the data storage process in a timely manner, such as data transmission delays and storage node failures. When problems occur, it is difficult to quickly locate the root cause of the problem and take effective countermeasures, which further exacerbates the security risks of data storage and fails to meet the current power industry's demand for high-quality storage of electricity metering data. Summary of the Invention

[0006] The purpose of this invention is to provide a blockchain-based secure storage system for electricity metering data to solve the problems mentioned in the background section.

[0007] To achieve the above objectives, the present invention provides a blockchain-based secure storage system for electricity metering data, the system comprising:

[0008] The power data acquisition module is used to statistically analyze metering data records and equipment status based on zoned power metering equipment, match equipment status with data requirements, calculate data acquisition differences, and generate raw metering data sets.

[0009] The data feature analysis module is used to analyze the trend of electricity consumption characteristics, sort the priority of features, calculate the feature utilization efficiency and fluctuation weight based on the original metering data set, and generate a set of data feature parameters.

[0010] The blockchain security verification module is used to identify data tampering risks and verification conflicts based on the data feature parameter set, apply the blockchain consensus protocol, adjust the security verification interval, and generate security verification results.

[0011] The storage optimization module is used to, based on the security verification results, statistically analyze the differences between storage time points and data distribution, match resource utilization with storage availability period, calculate the storage adjustment range, reorganize resource distribution, and generate a storage optimization scheme.

[0012] The real-time monitoring module is used to monitor changes in the blockchain network environment and data operation status based on the storage optimization scheme, calculate time deviation and environmental changes, count the remaining resources and operating hours, and generate monitoring response results.

[0013] Preferably, the original metering data set includes equipment status matching results and data acquisition difference analysis; the data feature parameter set includes feature priority sequence and fluctuation weight parameter; the security verification results include tampering risk identifier and verification conflict correction value; the storage optimization scheme includes storage allocation ratio and resource distribution reorganization results; and the monitoring response results include time deviation correction and environmental change parameters.

[0014] Preferably, the operation of the power data acquisition module includes: collecting power consumption records of the zone metering equipment, comprehensively statistically analyzing the status of the equipment in multiple zones and the matching degree with the data requirements, calculating the deviation between the total equipment records and the data requirements, and generating a data acquisition deviation set;

[0015] Analyze the data acquisition deviation set, adjust the allocation of device status and data acquisition, and establish a status and acquisition adjustment set;

[0016] Based on the aforementioned status and data collection adjustment set, the device status and data collection are compared to generate the original metering data set.

[0017] Preferably, the operation of the data feature analysis module includes: extracting energy consumption feature parameters based on the original metering data set, and establishing a feature change trend matrix;

[0018] The feature parameters are sorted using a priority model based on the feature change trend matrix, and the feature priority is adjusted in conjunction with the consumption trend.

[0019] Using the aforementioned feature priority and combined with actual fluctuation requirements, the feature utilization efficiency and fluctuation weight are calculated to generate the data feature parameter set.

[0020] Preferably, the operation of the blockchain security verification module includes: extracting feature priority sequences and fluctuation weight parameters from the data feature parameter set, analyzing the blockchain verification situation, and applying a consensus algorithm to determine verification requirements;

[0021] Based on the aforementioned verification requirements, all data tampering risks and verification conflicts are marked, and a risk and conflict index table is created.

[0022] Using the aforementioned risk and conflict index table, the security verification interval is recalculated, the verification plan is optimized, and the security verification result is generated.

[0023] Preferably, the operation of the storage optimization module includes: extracting tampering risk identifiers and verification conflict correction values ​​from the security verification results; analyzing data execution order and resource utilization efficiency based on the comparison of storage time and resource distribution; and generating a time node and data difference distribution table.

[0024] Using the aforementioned time nodes and data difference distribution table, resource utilization and storage availability cycles are matched, and based on the matching analysis of resource allocation and storage requirements, tasks with time conflicts and insufficient resources are identified.

[0025] Based on the aforementioned time conflict and resource shortage tasks, calculate the storage adjustment range and time node correction values, reorganize the resource distribution, and generate the aforementioned storage optimization scheme.

[0026] Preferably, the operation of the real-time monitoring module includes: extracting the storage allocation ratio and resource distribution reorganization results from the storage optimization scheme, combining the monitoring time nodes, using data analysis methods to determine the impact of environmental changes and data status on storage progress, and generating environmental and status analysis results;

[0027] Using the environmental and status analysis results, time deviation and environmental changes are calculated, and the remaining resources and operating hours are statistically analyzed. Through quantitative analysis methods, the resource allocation and time nodes that need to be adjusted are identified.

[0028] Based on the required adjustments to resource allocation and timeframes, adjust the resource allocation and timeframes to generate the monitoring response results.

[0029] Preferably, establishing the feature change trend matrix includes: extracting historical features from the energy consumption feature parameters, calculating the change trend values ​​within a specified time period, and arranging the change trend values ​​in chronological order to form a matrix.

[0030] Preferably, the application consensus algorithm for determining verification requirements includes: constructing a multi-dimensional verification space based on a feature priority sequence, calculating the clustering degree of historical tampering events in the verification space, and determining a set of early warning indicators;

[0031] Based on the early warning indicator set, real-time feature status is extracted, and location correlation is calculated in the verification space to generate risk correlation factors.

[0032] Preferably, the calculation of time deviation and environmental changes includes: determining the data signal distribution within a short time window based on environmental change parameters, analyzing the signal delay values ​​of adjacent nodes, and calculating the signal fluctuation factor;

[0033] Determine the frequency interference probability based on the signal fluctuation factor;

[0034] Based on the frequency interference probability, generate the nonlinear error probability;

[0035] Adjust the correction value at the time node based on the nonlinear error probability.

[0036] Compared with the prior art, the beneficial effects of the present invention are:

[0037] By using the power data acquisition module based on zoned power metering equipment, it can accurately record metering data and equipment status. Furthermore, it matches equipment status with data requirements, calculates data acquisition discrepancies, and ultimately generates a raw metering data set. This process makes the acquisition of raw metering data more targeted and accurate, avoiding acquisition deviations caused by mismatches between equipment status and data requirements. This provides a reliable raw data foundation for subsequent data processing and storage, ensuring the data quality of the entire storage system from the source of data generation.

[0038] The data feature analysis module, based on the original metering data set, analyzes the trends in electricity consumption characteristics, prioritizes features, calculates feature utilization efficiency and fluctuation weights, and generates a set of data feature parameters. Through in-depth mining and analysis of data features, the patterns and characteristics of electricity consumption can be clearly understood, along with the importance and utilization of different data features within the overall data system. These analytical results provide crucial reference for subsequent blockchain security verification, enabling more targeted verification around important data features and improving accuracy. They also provide data support for storage optimization, allowing for the rational allocation of storage resources based on data features.

[0039] The blockchain security verification module applies a blockchain consensus protocol, using a set of data feature parameters to identify data tampering risks and verification conflicts, adjusting the security verification interval, and generating security verification results. Blockchain technology itself possesses decentralized and immutable characteristics; combined with the application of a consensus protocol, it effectively prevents the risk of data tampering. By identifying tampering risks and verification conflicts, potential security vulnerabilities in the data can be identified in a timely manner. Adjusting the security verification interval according to the actual situation avoids both the waste of resources caused by overly frequent verification and the security vulnerabilities caused by excessively long verification intervals. This makes the security verification process more efficient and reasonable, effectively ensuring the security and integrity of measurement data during storage, ensuring that the data cannot be illegally tampered with, and maintaining the authenticity and reliability of the data.

[0040] Based on security verification results, the storage optimization module statistically analyzes the differences between storage time points and data distribution, matches resource utilization with storage availability, calculates the storage adjustment range, reorganizes resource distribution, and generates a storage optimization plan. This module fully considers data storage time, distribution, and storage resource utilization, and rationally adjusts the allocation of storage resources according to the storage needs of different data and the actual situation of storage resources. By reorganizing resource distribution, storage resources can be utilized more fully and rationally, avoiding the problems of idle or insufficient storage resources in traditional storage models, reducing storage resource waste. Simultaneously, it can allocate appropriate storage space and storage methods to different data based on the data's storage cycle and usage frequency, making data storage more flexible and improving the overall storage efficiency of the storage system.

[0041] The real-time monitoring module, based on a storage optimization scheme, monitors changes in the blockchain network environment and data operation status, calculates time deviations and environmental changes, and calculates remaining resources and operating hours, generating monitoring response results. Through real-time monitoring, it can promptly grasp the dynamic changes in the blockchain network environment and the operational status of data during storage and transmission. Once anomalies in the network environment or problems in data operation are detected, such as increased network latency or data transmission interruptions, it can quickly calculate time deviations and environmental changes, understand remaining resources and operating hours, and then generate corresponding monitoring response results. This enables staff to promptly identify anomalies in the storage system's operation, quickly take countermeasures, prevent problems from escalating, ensure the continuous and stable operation of the storage system, further improve the reliability and security of the entire system, and better meet the high requirements of the power industry for electricity metering data storage. Attached Figure Description

[0042] Figure 1 This is a timing diagram of the blockchain-based secure storage system for electricity metering data as described in this invention.

[0043] Figure 2 A flowchart defining the output results of each module;

[0044] Figure 3 A flowchart for the operation of the data feature analysis module;

[0045] Figure 4 A flowchart for the operation of the storage optimization module;

[0046] Figure 5 A flowchart for calculating time deviation and environmental changes. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] Please see Figure 1 This invention provides a blockchain-based secure storage system for electricity metering data, the system comprising:

[0049] The power data acquisition module, based on the statistical metering data records and equipment status of the zoned power metering equipment, matches equipment status with data requirements, calculates data acquisition differences, and generates a raw metering data set. The data feature analysis module, based on the raw metering data set, analyzes power consumption characteristics and trends, prioritizes features, calculates feature utilization efficiency and fluctuation weights, and generates a data feature parameter set. The blockchain security verification module, based on the data feature parameter set, applies the blockchain consensus protocol, identifies data tampering risks and verification conflicts, adjusts the security verification interval, and generates security verification results. The storage optimization module, based on the security verification results, statistically analyzes the differences between storage time nodes and data distribution, pairs resource utilization with storage availability cycles, calculates the storage adjustment range, reorganizes resource distribution, and generates a storage optimization scheme. The real-time monitoring module, based on the storage optimization scheme, monitors changes in the blockchain network environment and data operation status, calculates time deviations and environmental changes, statistically analyzes remaining resources and operating hours, and generates monitoring response results.

[0050] Example 1: See Figure 2 In a city power grid renovation project, the system connected metering equipment from 12 distribution zones, including three types of heterogeneous devices: old mechanical meters, new smart meters, and voltage monitors. When the power data acquisition module starts, it first collects power consumption records uploaded by each zone's devices: hourly pulse count data forwarded by the concentrator from mechanical meters, timestamped power consumption values ​​transmitted by smart meters every 15 minutes via LoRaWAN, and voltage and current waveform segments reported by the voltage monitors every 5 minutes. Simultaneously, it collects device status information: battery voltage values ​​from mechanical meters, signal strength and memory usage from smart meters, and sampling accuracy indicators from voltage monitors.

[0051] The matching process between equipment status and data requirements is represented by a dynamic mapping table. The system's preset data requirements include: 15-minute-level power data for load analysis, voltage and current waveforms for line loss calculation, and status parameters for equipment health monitoring. A mechanical meter is marked as "readability risk" due to its battery voltage being below a threshold, and its pulse data is only used for load analysis; a smart meter is marked as "transmission delay" due to weak signal strength, and its data is prioritized for line loss calculation; if a voltage monitor's sampling accuracy is abnormal, waveform acquisition is paused, and only status parameters are provided. The matching degree is calculated using a quantitative scoring system: 1 point for a complete match, 0.5 points for a partial match, and 0 points for no match. For example, a smart meter that simultaneously meets both load analysis (0.5 points, due to transmission delay) and equipment monitoring (1 point) has a total matching degree score of 1.5.

[0052] Data acquisition discrepancy analysis focuses on three dimensions: In terms of time, it compares the actual 15-minute acquisition interval of the smart meter with the required value by ±3 seconds; in terms of data integrity, it checks whether the voltage monitor waveform segments are missing cycle data; and in terms of accuracy, it verifies whether the kWh value after pulse conversion of the mechanical meter is within the ±0.5% error band. One acquisition revealed that: smart meter #3 in area A experienced a 28-second delay due to network congestion; the pulse counter of the mechanical meter in area B had a 0.8% system error; and the voltage monitor in area C was missing 5% of its waveform points. These discrepancies were recorded as structured data: {Device ID: DT-A3, Discrepancy Type: Time Delay, Discrepancy Value: +13 seconds}.

[0053] The generation of status and data acquisition adjustment sets adopts a closed-loop control mechanism. For the transmission delay of smart meters in Zone A, the system reduces its acquisition frequency to 30 minutes / time, while activating a backup communication node. For the counting error of mechanical meters in Zone B, a pulse compensation algorithm is initiated and the meters are marked as requiring on-site calibration. Voltage monitors in Zone C switch to redundant equipment for acquisition. The adjustment strategy is recorded as: {Equipment Group: Voltage Monitor - Zone C, Original Acquisition Mode: Single-point sampling by the main device, Adjusted Mode: Dual-device cross-validation}.

[0054] The final generated raw metering data set contains multi-dimensional information. Taking a certain time slice as an example: timestamp 2023-08-2014:00:00, device DT-B7 (smart meter), metered value 15.32kWh, status matching result {load analysis: 0.8, line loss calculation: 0.6}, data acquisition difference {time deviation: -2 seconds, integrity: 100%}. This set is transmitted through an encrypted channel with an attached digital signature. The signature contains the device serial number and acquisition batch number, forming an indivisible data unit. Throughout the entire acquisition cycle, the system processes data from 287 devices, generating a raw metering data set containing device status matching results and data acquisition difference analysis, providing structured input for subsequent modules.

[0055] Example 2: See Figure 3In a smart meter data analysis project implemented by a provincial power grid company, the system received a raw metering data set generated in Example 1. This set contained metering data collected from 350,000 smart meters across 8 cities in the province over a 72-hour period. The data feature analysis module initiated the processing flow. The module first extracted energy consumption characteristic parameters from the raw metering data set, including 12 dimensions of characteristic indicators such as average hourly electricity consumption, 15-minute load fluctuation, and daily electricity consumption variance. For the meter data of an industrial area, it was found that its average electricity consumption remained high during the day and dropped sharply at night; while residential meters showed a pattern of two peak electricity consumption periods, one in the morning and one in the evening. The module established a feature change trend matrix, dividing the 72 hours into 216 time units (each unit being 1 hour), with each time unit containing the numerical changes of 12 characteristic parameters. The rows of the matrix represent the time series, and the columns represent the feature dimensions, forming a 216×12 feature matrix.

[0056] The priority model employs a multi-factor weighted algorithm to rank feature parameters, considering three ranking dimensions: the degree of feature impact on grid stability (e.g., higher weight for load fluctuation values), the completeness of feature data (e.g., features with a data missing rate of less than 5% are prioritized), and the predictability of feature changes (e.g., features with lower variance are prioritized). The processing results show that load fluctuation values, peak electricity consumption, and valley electricity consumption are classified as high-priority features, while average power factor and voltage deviation are classified as medium- to low-priority features. When adjusting feature priorities based on consumption trends, an abnormal load fluctuation was observed in the electricity meters of a certain commercial area during midday; the system automatically elevated the priority of the load fluctuation feature for that period to the highest level.

[0057] The efficiency calculation of feature utilization employs a ratio analysis method between actual and theoretical values. For high-priority load fluctuation features, the system calculates the degree of matching between the actual collected fluctuation data and the expected values ​​of the power grid dispatch. Fluctuation weight calculation is based on historical data statistical analysis. For example, the fluctuation weight of a residential area's electricity meter during the evening peak hours is set to 0.85, while the fluctuation weight during the early morning hours is only 0.15. The final generated data feature parameter set includes a feature priority sequence and fluctuation weight parameters. The priority sequence uses a three-level classification (high, medium, low), and the fluctuation weight parameters are accurate to three decimal places.

[0058] The blockchain security verification module then begins operation. This module extracts feature priority sequences and fluctuation weight parameters from the data feature parameter set and compares and analyzes them with historical verification records in the blockchain network. The system employs an improved practical Byzantine fault-tolerant consensus algorithm, dynamically adjusting verification requirements based on feature priority: high-priority feature data requires consensus from 7 out of 9 verification nodes, medium-priority data requires 6 nodes, and low-priority data requires 5 nodes.

[0059] During the analysis of blockchain verification, the system detected an anomaly in a batch of industrial electricity meter data: the feature priority sequence indicated that its load fluctuation value should be high priority, but in the actual verification records, this feature was classified as medium priority. The consensus algorithm determined that this batch of data needed to be re-verified, with verification requirements including increasing the number of verification nodes and extending the verification time window. The system flagged data tampering risks, with risk identifiers including risk level (high, medium, low), risk type (priority mismatch, data anomaly, etc.), and risk location. Simultaneously, verification conflicts were detected, such as multiple verification nodes disagreeing on the same data block.

[0060] When creating the risk and conflict index table, the system assigns a unique identifier to each risk event and records information such as risk characteristics, occurrence time, and related data blocks. The index table adopts a hierarchical structure: the first layer records a risk summary, and the second layer stores detailed conflict data. When recalculating the security verification interval, the system dynamically adjusts it based on the risk level: the verification interval for high-risk data is shortened to 5 minutes, for medium-risk data it remains at 15 minutes, and for low-risk data it is extended to once every 30 minutes. Optimizing the verification plan includes adjusting the allocation of verification nodes and modifying the consensus threshold. The final generated security verification result includes a tampering risk identifier and a verification conflict correction value. The tampering risk identifier uses color coding (red, yellow, green) to indicate the degree of risk, and the verification conflict correction value includes verification parameter adjustment suggestions and conflict resolution solutions.

[0061] Example 3: See Figure 4 The storage optimization module initiates the processing flow. First, it extracts tamper risk identifiers and verification conflict correction values ​​from the security verification results and maps them to the data storage management layer. Data blocks corresponding to high-risk tampering identifiers are marked as frozen, prohibiting write operations; medium- and low-risk data are assigned different access permission levels. Based on a comparative analysis of storage time and resource distribution, the module establishes a cross-reference system of time and space dimensions. In the time dimension, the system statistically analyzes the creation timestamp, last modification time, and planned storage period for each data block; in the spatial dimension, it records the distribution of data blocks across different storage nodes.

[0062] When analyzing data execution order and resource utilization efficiency, the system employs a multi-dimensional evaluation matrix. A batch of residential meter data had its storage priority downgraded due to a high verification conflict correction value; conversely, critical metering data from industrial areas received priority storage access due to a low risk of tampering. Resource utilization efficiency was assessed by calculating the correlation between storage throughput and capacity utilization, revealing storage space fragmentation issues at certain nodes. The generated time node and data difference distribution table includes the following fields: data block ID, planned storage time window, actual storage timestamp, storage node location, data integrity score, and resource utilization index.

[0063] Using time points and data difference distribution tables, the system performs a pairing analysis of resource utilization and storage availability period. Storage availability period is determined based on a combination of hardware lifespan, data importance, and access frequency. The pairing process identified several conflict points: the required storage availability period for a set of high-voltage substation monitoring data is 10 years, but the current allocated resource utilization is already close to the threshold; some low-voltage user data, although with short availability periods, have low resource utilization. Based on the matching analysis of resource allocation and storage needs, the system identified 19 time conflict points and 43 resource-insufficient tasks.

[0064] When calculating the storage adjustment range, the system uses a dynamic programming algorithm to determine the optimal adjustment strategy. The time-node correction value is calculated by combining historical data fluctuation patterns with the current resource status. During resource reconfiguration, the system performs the following operations: migrating high-risk data to storage nodes with higher security levels; implementing a sharding storage strategy for data with severe time conflicts; and adjusting the load balancing parameters of the storage nodes. The final storage optimization scheme includes storage allocation ratios and resource reconfiguration results. The storage allocation ratios are accurate to the percentage distribution of each type of data across different storage tiers, while the resource reconfiguration results record in detail the final storage location and access parameters of each data block.

[0065] The real-time monitoring module extracts storage allocation ratios and resource distribution reorganization results from the storage optimization scheme and compares them with the real-time status of the blockchain network. The monitoring timeframe is set to a 5-minute data collection cycle, with a full scan of network environment changes and data operation status within each cycle. Using multi-dimensional data analysis methods, the system tracks the deviation between storage progress and the planned timeline, and assesses the impact of environmental changes such as network latency and node failures on storage operations.

[0066] When determining the impact of environmental changes and data status on storage progress, the system establishes a causal relationship model. If a network bandwidth fluctuation causes data migration delays, the system records the correspondence between bandwidth changes and delay times; if a sudden failure of a storage node causes data access interruption, the system analyzes the failure mode and its scope of impact. The generated environmental and status analysis results include environmental parameter change curves, data status transition maps, and impact assessment matrices.

[0067] Using the results of environmental and condition analysis, the system calculates quantitative indicators of time deviation and environmental changes. Time deviation is calculated by comparing the planned timeline with the actual execution timeline, and the following formula is used for deviation assessment:

[0068]

[0069] in: This represents the comprehensive time deviation index. This represents the planned completion time of the i-th storage operation. Represents the actual completion time. The weighting coefficients are set based on the importance of the operations, where n is the total number of operations within the monitoring period. Environmental changes are quantified by monitoring indicators such as network latency volatility and node availability change rate.

[0070] When calculating remaining resources and operating hours, the system collects parameters such as remaining storage capacity, memory utilization, and CPU load for each storage node in real time. Operating hour records include data migration time, verification operation duration, and compression processing time. Through quantitative analysis, the system identifies 12 resource allocation schemes and 8 time node settings that require adjustment. Based on these parameters, the system performs dynamic adjustments: reallocating data to nodes with resource constraints; reordering operation processes with significant time deviations; and adding redundant backups to data blocks sensitive to environmental changes. During the adjustment process, the system adopts a gradual optimization strategy, adjusting only one parameter group at a time, observing the system response, and then making subsequent adjustments. The final monitoring response results include time deviation correction values ​​and environmental change parameters. The time deviation correction values ​​record the time compensation amount for each adjustment operation in detail, while the environmental change parameters include network status adjustment records and node resource configuration update schemes.

[0071] Throughout the process, the system successfully kept the time deviation of storage operations within acceptable limits, effectively handled three sudden network environment events, reallocated 15% of storage resources, and optimized the storage location of 28 data blocks. The monitoring response results generated by the real-time monitoring module were fed back to the system control center, providing a basis for decision-making regarding subsequent operational cycle adjustments.

[0072] Example 4: The process of determining requirements through the construction of a feature change trend matrix and the verification of consensus algorithms. In an electricity consumption behavior analysis system implemented by a regional power grid company, the system processes metering data uploaded by 420,000 smart meters within its jurisdiction, spanning a continuous 30-day period. The data feature analysis module initiates the feature change trend matrix construction process, extracting electricity consumption feature parameters from the original metering data set, including eight core features such as hourly electricity consumption, 15-minute load change rate, and daily peak-valley difference. Historical feature extraction uses a sliding window technique, with a window width set to 24 hours and a sliding step size of 1 hour. Trend calculations are performed on the feature data within each window. For example, for a residential user's smart meter, during the window from 08:00 on September 1, 2023 to 08:00 on September 2, 2023, the first-order difference value of hourly electricity consumption reflects the consumption changes in adjacent time periods, while the moving average represents the overall trend. The specified time period is determined to be 7 days (168 hours), and the feature change values ​​of the 168 time points are arranged in chronological order to form a feature change trend matrix. The rows of this matrix correspond to the time series, and the columns correspond to the feature dimensions. Each cell stores the change value of a specific feature at a specific time point. The table below shows a partial structure of this matrix:

[0073] Table 1: Feature change trend matrix segment (time window: 2023-09-01 to 2023-09-07).

[0074]

[0075] The blockchain security verification module constructs a three-dimensional verification space based on the feature priority sequence generated by the aforementioned matrix. The spatial coordinate axes correspond to: X-axis – load fluctuation priority (high: 0.8-1.0, medium: 0.4-0.7, low: 0-0.3), Y-axis – data integrity level (complete: 0.9-1.0, partially missing: 0.5-0.8, severely missing: 0-0.4), and Z-axis – time sensitivity (real-time: 0.7-1.0, near-real-time: 0.4-0.6, non-real-time: 0-0.3). The system extracts tampering event data from the blockchain history records over the past 6 months, including feature parameters from 32 data tampering events. Clustering analysis within the verification space reveals a high-risk area (coordinate range X: 0.85-0.95, Y: 0.88-0.92, Z: 0.75-0.85) containing 14 tampering events, accounting for 43.75% of the total events. The geometric center coordinates of the area were determined to be (0.91, 0.90, 0.80) using a density scanning algorithm, and the warning radius was set to 0.07 distance units, forming a core warning indicator set.

[0076] In the real-time data processing phase, the system extracts the real-time characteristic status of an industrial user's electricity meter: load fluctuation priority 0.92 (due to sudden startup of production equipment), data integrity 0.89 (minor packet loss), and time sensitivity 0.78 (used for real-time billing). This status is mapped to the coordinates (0.92, 0.89, 0.78) in the verification space. Location correlation is calculated using a spatial distance metric, determining the straight-line distance between this point and the center of the warning area to be 0.038 units (less than the warning radius of 0.07). Risk correlation factor generation uses an inverse distance algorithm, setting the maximum correlation distance to 0.1 units, resulting in a risk correlation factor of 0.82 (range 0-1, with higher values ​​indicating higher risk). This factor is appended to the metadata of the data block, triggering the enhanced processing flow of the blockchain verification mechanism.

[0077] During seven consecutive days of operation, the system processed 10.08 million feature data records (60,000 records per hour × 24 hours × 7 days), constructing 168 feature change trend matrices (generating a new matrix every hour). The consensus algorithm verification requirement determination process marked 217 high-risk data blocks, of which 183 were manually verified to be anomaly-free, achieving an accuracy rate of 84.3% for the verification space model. The dynamic update mechanism of the multi-dimensional verification space recalculates the clustering distribution every 24 hours, ensuring that the early warning indicator set adapts to changes in attack patterns. The application of risk correlation factors is reflected in the allocation of verification resources: data blocks with correlation factors higher than 0.7 use a 9-node consensus mechanism, those between 0.4 and 0.7 use a 7-node mechanism, and those below 0.4 use a 5-node basic verification mechanism. This system achieves quantitative assessment and graded response to security threats to electricity metering data.

[0078] Example 5: See Figure 5 In the actual deployment at a provincial power grid dispatch center, the real-time monitoring module, based on the storage allocation ratio and resource distribution reorganization results extracted from the storage optimization scheme, combined with blockchain network environment monitoring data, performs refined calculations of time deviation and environmental changes. Environmental change parameters are derived from real-time reports from distributed monitoring nodes, including 12 categories of indicators such as network transmission latency, node response time, and packet loss rate. The short-time window is set at 300 milliseconds, with 500 data signal samples collected within each window. These signals come from 1,200 monitoring nodes deployed throughout the province.

[0079] When determining the data signal distribution within a short time window, the system employs a spectrum analysis method. One monitoring session revealed that within the time window from 14:30:00 to 14:30:00.300 on November 15, 2023, the signal strength of nodes in the northern region exhibited a bimodal distribution, mainly concentrated in the mid-to-high frequency band; while the signal strength of nodes in the southern region showed a single-peak distribution, concentrated in the low frequency band. This difference in distribution stems from the heterogeneity of the regional network infrastructure: the north uses a new generation of fiber optic networks, while the south still partially uses copper cable networks. When analyzing the signal delay values ​​of adjacent nodes, the system establishes a node topology mapping and calculates the signal transmission delay between each pair of adjacent nodes. For example, the delay value between node A7 and adjacent node B3 is 28 milliseconds, and the delay value between node C5 and adjacent node D9 is 42 milliseconds. A total of 8,500 pairs of adjacent node delay values ​​were calculated across the entire provincial network.

[0080] When calculating the signal fluctuation factor, the system statistically analyzes the variation of latency values ​​every 30 seconds. Within a certain period, the coefficient of variation for latency values ​​was 0.18 in the northern region, 0.31 in the southern region, and reached 0.45 in the coastal region due to weather conditions. The signal fluctuation factor is calculated using a weighted average method, incorporating these regional coefficients of variation, with weights allocated based on node importance and data traffic. The overall signal fluctuation factor for the entire provincial network is calculated to be 0.28.

[0081] To determine the frequency interference probability, the system retrospectively analyzed historical interference events over the past 72 hours. Pattern matching between the current signal fluctuation factor and historical data revealed that when the fluctuation factor was in the 0.25-0.35 range, the historical frequency interference probability was 23.7%. Combining this with parameters such as the current network load (78%) and ambient temperature (32℃), a conditional probability model calculated the current frequency interference probability to be 26.3%.

[0082] When generating the nonlinear error probability, the system employs a Gaussian process regression model from machine learning. The model input includes eight feature parameters, such as signal fluctuation factor, frequency interference probability, network load rate, and ambient temperature. Training data comes from network operation records over the past three months, containing 12 million labeled samples. The model output is a probability value between 0 and 1, representing the likelihood of nonlinear errors in the time synchronization system. The currently calculated nonlinear error probability is 0.34, primarily contributed by the signal fluctuation factor (weight 0.41) and network load rate (weight 0.33).

[0083] When adjusting the time node correction value, the system establishes an error compensation mechanism. The base time node comes from the consensus timestamp of the blockchain network, and the correction value is calculated based on the weighted integral of the nonlinear error probability. When the nonlinear error probability is below 0.2, a linear compensation mode is used; when the probability is between 0.2 and 0.5, a quadratic function compensation is used; and when the probability is above 0.5, an exponential compensation mode is activated. Currently, the probability value is 0.34, so the system selects the quadratic function compensation algorithm, generating a time node correction value of +47 milliseconds. This correction value is applied to the time synchronization system of the entire province's network, and the timestamps of all monitoring nodes and storage nodes are increased by an offset of 47 milliseconds.

[0084] Throughout the processing cycle, the system executes a complete calculation every 5 minutes, continuously monitoring changes in the network environment. Each calculation generates signal distribution data from 1,200 nodes, 8,500 sets of delay values, one global signal fluctuation factor, one frequency interference probability value, one nonlinear error probability value, and one time node correction value. This data is recorded on the blockchain, forming a traceable environmental adjustment history. Through this continuous environmental adaptation mechanism, the system ensures the time consistency of electricity metering data during storage and transmission, providing an accurate time reference for blockchain security verification.

[0085] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0086] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A blockchain-based secure storage system for electricity metering data, characterized in that, The system includes: The power data acquisition module is used to statistically analyze metering data records and equipment status based on zoned power metering equipment, match equipment status with data requirements, calculate data acquisition differences, and generate raw metering data sets. The data feature analysis module is used to analyze the trend of electricity consumption characteristics, sort the priority of features, calculate the feature utilization efficiency and fluctuation weight based on the original metering data set, and generate a set of data feature parameters. The blockchain security verification module is used to identify data tampering risks and verification conflicts based on the data feature parameter set, apply the blockchain consensus protocol, adjust the security verification interval, and generate security verification results. The storage optimization module is used to, based on the security verification results, statistically analyze the differences between storage time points and data distribution, match resource utilization with storage availability period, calculate the storage adjustment range, reorganize resource distribution, and generate a storage optimization scheme. The real-time monitoring module is used to monitor changes in the blockchain network environment and data operation status based on the storage optimization scheme, calculate time deviation and environmental changes, count the remaining resources and operating hours, and generate monitoring response results. The original measurement data set includes equipment status matching results and data acquisition difference analysis; The operation of the power data acquisition module includes: collecting power consumption records of the power metering equipment in the zones, comprehensively calculating the status of the equipment in multiple zones and the matching degree with the data requirements, calculating the deviation between the total records of the equipment and the data requirements, and generating a data acquisition deviation set. Analyze the data acquisition deviation set, adjust the allocation of device status and data acquisition, and establish a status and acquisition adjustment set; Based on the aforementioned status and data collection adjustment set, the device status and data collection are compared to generate the original metering data set.

2. The blockchain-based secure storage system for electricity metering data according to claim 1, characterized in that, The data feature parameter set includes a feature priority sequence and a fluctuation weight parameter; the security verification result includes a tampering risk identifier and a verification conflict correction value; the storage optimization scheme includes a storage allocation ratio and a resource distribution reorganization result; and the monitoring response result includes a time deviation correction and an environmental change parameter.

3. The blockchain-based secure storage system for electricity metering data according to claim 2, characterized in that, The operation of the data feature analysis module includes: extracting energy consumption feature parameters based on the original metering data set, and establishing a feature change trend matrix; The feature parameters are sorted using a priority model based on the feature change trend matrix, and the feature priority is adjusted in conjunction with the consumption trend. Using the aforementioned feature priority and combined with actual fluctuation requirements, the feature utilization efficiency and fluctuation weight are calculated to generate the data feature parameter set.

4. The blockchain-based secure storage system for electricity metering data according to claim 3, characterized in that, The operation of the blockchain security verification module includes: extracting feature priority sequences and fluctuation weight parameters from the data feature parameter set, analyzing the blockchain verification situation, and applying a consensus algorithm to determine the verification requirements; Based on the aforementioned verification requirements, all data tampering risks and verification conflicts are marked, and a risk and conflict index table is created. Using the aforementioned risk and conflict index table, the security verification interval is recalculated, the verification plan is optimized, and the security verification result is generated.

5. The blockchain-based secure storage system for electricity metering data according to claim 4, characterized in that, The operation of the storage optimization module includes: extracting tampering risk identifiers and verification conflict correction values ​​from the security verification results; analyzing data execution order and resource utilization efficiency based on the comparison of storage time and resource distribution; and generating a time node and data difference distribution table. Using the aforementioned time nodes and data difference distribution table, resource utilization and storage availability cycles are matched, and based on the matching analysis of resource allocation and storage requirements, tasks with time conflicts and insufficient resources are identified. Based on the aforementioned time conflict and resource shortage tasks, calculate the storage adjustment range and time node correction values, reorganize the resource distribution, and generate the aforementioned storage optimization scheme.

6. The blockchain-based secure storage system for electricity metering data according to claim 5, characterized in that, The operation of the real-time monitoring module includes: extracting the storage allocation ratio and resource distribution reorganization results from the storage optimization scheme, combining the monitoring time nodes, using data analysis methods to determine the impact of environmental changes and data status on storage progress, and generating environmental and status analysis results. Using the environmental and status analysis results, time deviation and environmental changes are calculated, and the remaining resources and operating hours are statistically analyzed. Through quantitative analysis methods, the resource allocation and time nodes that need to be adjusted are identified. Based on the required adjustments to resource allocation and timeframes, adjust the resource allocation and timeframes to generate the monitoring response results.

7. The blockchain-based secure storage system for electricity metering data according to claim 3, characterized in that, The establishment of the feature change trend matrix includes: extracting historical features from the power consumption feature parameters, calculating the change trend values ​​within a specified time period, and arranging the change trend values ​​in chronological order to form a matrix.

8. The blockchain-based secure storage system for electricity metering data according to claim 4, characterized in that, The application consensus algorithm for determining verification requirements includes: constructing a multi-dimensional verification space based on a feature priority sequence, calculating the clustering degree of historical tampering events in the verification space, and determining a set of early warning indicators; Based on the early warning indicator set, real-time feature status is extracted, and location correlation is calculated in the verification space to generate risk correlation factors.

9. The blockchain-based secure storage system for electricity metering data according to claim 6, characterized in that, The calculation of time deviation and environmental changes includes: determining the data signal distribution within a short time window based on environmental change parameters, analyzing the signal delay values ​​of adjacent nodes, and calculating the signal fluctuation factor; Determine the frequency interference probability based on the signal fluctuation factor; Based on the frequency interference probability, generate the nonlinear error probability; Adjust the correction value at the time node based on the nonlinear error probability.

Images