Detailed Description
The following description of the embodiments of the present invention will be made apparent and fully in view of the accompanying drawings, in which some, but not all embodiments of the invention are shown. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings, are merely for convenience of describing the present invention and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In addition, the technical features of the different embodiments of the present invention described below may be combined with each other as long as they do not collide with each other.
For easy understanding, the following describes a specific flow of the embodiment of the present invention, referring to fig. 1, fig. 1 is a flowchart of a power quality tracing method based on a blockchain according to the embodiment of the present invention, and as shown in fig. 1, the method includes the following steps:
s101, collecting and preprocessing the power quality parameters of all links of power production, transmission, distribution and consumption in real time to obtain a standardized power quality data block;
s102, performing blockchain recording and consensus verification on the standardized power quality data blocks to obtain power quality blockchain data;
S103, performing intelligent contract deployment and automatic evaluation on the power quality block chain data to obtain a power quality evaluation result;
s104, carrying out visual processing on the power quality evaluation result through a multidimensional analysis algorithm to obtain a power quality analysis report and an abnormal alarm;
s105, performing multiparty collaborative diagnosis and treatment on the power quality abnormal problems in the abnormal alarm through a collaborative diagnosis platform to obtain a problem solution and a treatment record;
and S106, performing traceability analysis and trend prediction on the electric energy quality block chain data, the electric energy quality analysis report and the processing record to obtain an electric energy quality improvement suggestion and an optimization scheme.
The power quality tracing method based on the block chain is characterized in that intelligent sensors and monitoring equipment are deployed in all links of power production, transmission, distribution and consumption, and power quality parameters such as voltage, current, frequency, harmonic wave and power factor are collected in real time. The collected original data is subjected to preliminary screening and cleaning to remove obvious abnormal values and redundant information, and then standardized to unify formats and units. And adding metadata such as a time stamp, a geographic position, a device identifier and the like to each piece of data, and ensuring the traceability of the data. Through the early warning mechanism, when the acquired power quality parameter exceeds a preset threshold value, an alarm is triggered immediately. The preprocessed data is packed into standardized data blocks, each data block containing power quality information over a period of time. The standardized power quality data blocks are then uploaded to a specially designed power quality blockchain network. The network adopts a alliance chain structure, and participants comprise power production enterprises, power grid companies, power distribution units, large-scale power utilization enterprises, supervision departments and the like. When a new data block is submitted, nodes in the network first verify the integrity and legitimacy of the data. After verification passes, a modified proof of work (PoW) consensus algorithm is employed to decide which node to create a new block from. After the new block is created, it contains all the power quality data for a certain period of time, as well as the hash value, timestamp and other necessary information of the previous block. Other nodes verify the validity of the new block, and after consensus is reached, add it to the chain to form the power quality blockchain data.
A series of intelligent contracts are then deployed in the blockchain network, which are automatically executed according to preset power quality criteria and rules. The content of the intelligent contract comprises voltage deviation judgment, harmonic content evaluation, frequency fluctuation analysis, electric energy quality comprehensive evaluation and the like. The intelligent contract management system is responsible for deployment, updating and monitoring of contracts, and can dynamically adjust contract contents according to the latest power industry standards and regulations. When a new data block is added to the blockchain, the intelligent contract management system automatically scans the data content to identify whether a contract trigger condition is met. And if so, automatically executing the corresponding intelligent contract to obtain the power quality evaluation result. Based on the result of intelligent contract execution, the multidimensional power quality analysis system carries out deep analysis on power quality data from multiple dimensions such as time, space, parameter types and the like. Long-term trends and periodic changes in the power quality parameters are identified using time series analysis techniques. And (3) using a spatial clustering algorithm to find out the power quality difference and potential problem areas of different geographic areas. And adopting multivariate correlation analysis to explore the internal relation between different electric energy quality parameters. The analysis results are presented in an intuitive and interactive manner through a visualization platform, and various chart types such as thermodynamic diagrams, trend lines, scatter diagrams and the like are supported. Meanwhile, a Geographic Information System (GIS) is integrated, and the power quality data and the geographic position are associated to form a power quality map. The anomaly detection algorithm automatically identifies an abnormal power quality pattern based on historical data and a statistical model, and when an anomaly is detected, the system highlights in a visual interface and sends an alert to the relevant person.
For detected abnormal problems, the collaborative problem diagnosis and treatment platform allows different parties to participate in the diagnosis and resolution process of the problem in common. The platform automatically generates a preliminary problem diagnosis report including description of the problem, possible reasons, scope of influence, etc. Then, a collaboration flow based on the smart contract is initiated. The interested party may submit its own diagnostic comments, evidence and solutions on the chain. Each submitted content is recorded on the blockchain, ensuring transparency and traceability of the process. Reputation-based incentive mechanisms encourage participants to provide valuable diagnostics and solutions. The intelligent decision support system comprehensively considers the opinion of each party, the historical cases and the current situation, and provides suggestions for the final solution. Finally, the power quality tracing system can trace the power quality condition of any node at any time point and analyze the historical change trend of the power quality condition. The intelligent search engine supports multidimensional, multi-conditional traceable queries. The knowledge graph system performs semantic association on information such as power quality data, problem diagnosis, processing schemes and the like, displays the relation between the data, and discovers potential rules and relations. The machine learning-based prediction model predicts future power quality trends using historical data, guiding preventive maintenance and system optimization. Continuous improvement management systems periodically analyze long-term trends in power quality, identify systematic problems and improvement opportunities, automatically generate improvement suggestions, and distribute to interested parties over a blockchain network.
For example, a certain electric company deploys the power quality traceability system in its distribution network. The system collects about 100 ten thousand pieces of power quality data including voltage, frequency, harmonic and other parameters in one month. Through preprocessing, the data is compressed and packed into 10000 standardized data blocks. These data blocks are recorded onto the blockchain to form a power quality blockchain containing 10000 blocks. The smart contract automatically performs 50000 evaluations, finding 500 potential power quality problems. The multidimensional analysis system generates a detailed monthly report comprising 30 trend graphs, 10 thermodynamic graphs, and 5 associated network graphs. Wherein the system recognizes that the voltage in one substation area fluctuates frequently, exceeding a standard threshold. The collaborative diagnosis platform sumps the power grid company, equipment manufacturer and the specialist of the power utilization enterprise for remote consultation, and determines that the problem is caused by an aged transformer through analysis of historical data and field inspection. Experts propose solutions on the platform to replace devices, which are recorded on the blockchain and executed. After one month, trend predictions for the system show a 15% improvement in voltage stability for this area, saving about 50 ten thousand yuan potential loss for the utility.
By executing the steps, the power quality parameters of all links of power production, transmission, distribution and consumption are comprehensively monitored by using a real-time acquisition and preprocessing technology, so that the timeliness and the accuracy of power quality data are ensured. By generating and processing the standardized power quality data blocks, the monitoring of the power quality not only covers each link, but also realizes the unification of data formats, thereby facilitating the subsequent analysis and comparison. Second, the introduction of blockchain technology provides security and transparency to the power quality data during data recording and consensus verification. The distributed account book characteristics of the blockchain enable the power quality data to be untampered once recorded, and all participants can access and verify the data in real time, so that the credibility of the system and the trust sense of users are improved. In addition, the deployment and automatic evaluation of the intelligent contracts greatly improve the efficiency of the power quality evaluation, the intelligent contracts can be automatically executed by setting clear evaluation rules, the risk of manual intervention is reduced, the data processing flow is accelerated, the power quality evaluation result can be rapidly fed back to related parties, and more efficient decision support is realized. The multi-dimensional analysis algorithm introduced in the scheme provides deep insight for power quality analysis, and complex power quality data can be presented to a user through visual processing, so that a power quality analysis report and an abnormal alarm which are easy to understand are formed. The visualization technology is not only convenient for the electric operators to monitor daily, but also provides visual analysis basis for the abnormal problem of the electric energy quality, and the power assisting is used for rapidly positioning and processing faults. Meanwhile, through the collaborative diagnosis platform, the collaborative diagnosis and processing of multiple parties can be realized aiming at the power quality problem in the abnormal alarm, and the information sharing and resource integration among the experts of the parties are promoted, so that the problem solving process is accelerated. The multiparty cooperation mechanism improves the overall response capability of the power system and ensures the stability and reliability of electric energy supply. The integrated power quality data set is formed by integrating the power quality block chain data, the analysis report and the processing record, so that not only can the time sequence reconstruction be performed, but also a power quality knowledge graph can be constructed through a relational network. The process enables the power quality management not only to be limited to simple records of historical data, but also can deeply analyze the relation among the data, and realize causal chain tracking, so that the source of the power quality problem is found. Through the clustering treatment of the problem types, the power system can more clearly know the existing problems and potential risk points, carry out scientific trend analysis and risk assessment, finally provide a clear direction and an optimization scheme for the power quality improvement, and improve the efficiency and the accuracy of the power quality tracing based on the blockchain.
In a specific embodiment, the process of executing step S101 may specifically include the following steps:
(1) Sampling and digitizing the power quality parameters to obtain original power quality data, and performing time stamp marking on the original power quality data to obtain time sequence power quality data;
(2) Performing outlier detection and removal on the time series power quality data to obtain cleaned power quality data, and performing data format unification processing on the cleaned power quality data to obtain power quality data in a unified format;
(3) Carrying out data compression processing on the uniform-format power quality data to obtain compressed power quality data, and carrying out encryption processing on the compressed power quality data to obtain encrypted power quality data;
(4) The encryption power quality data is subjected to segmentation processing to obtain power quality data fragments, and the power quality data fragments are subjected to hash processing to obtain power quality data hash values;
(5) And carrying out combination processing on the hash value of the power quality data and the encrypted power quality data to obtain a power quality data packet, and carrying out checksum calculation on the power quality data packet to obtain a standardized power quality data block.
In particular, the sampling frequency needs to be high enough to capture the instantaneous change in power quality, typically set to hundreds of times the fundamental frequency of the grid. The digitized data is the original power quality data, including parameters such as voltage, current, frequency, harmonic wave, etc. A time stamp accurate to the order of milliseconds is then added to each piece of raw data, forming a time series of power quality data. The time stamp ensures the time sequence and traceability of the data, and lays a foundation for subsequent trend analysis and anomaly detection. Next, abnormal value detection and removal are performed on the time-series power quality data. Statistical methods, such as moving average and standard deviation methods, are used to identify and reject data points that deviate significantly from the normal range. For example, anomalies such as abrupt voltage changes or frequency dips may be marked and analyzed further. The data after the outlier removal is referred to as the washed power quality data. And then, carrying out format unification processing on the cleaned data, and converting the data with different sources and types into unified formats and units. The consistency of the data is ensured, and the subsequent analysis and processing are convenient.
In order to improve the data transmission and storage efficiency, the power quality data in a unified format is compressed. Compression algorithm selection requires a balance between compression rate and information retention, and common methods include wavelet transform compression or differential coding compression. The compressed data size is significantly reduced, but key power quality information is retained. The compressed data is then encrypted to protect sensitive information. The encryption algorithm adopts a high-strength symmetrical encryption method, such as AES (Advanced Encryption Standard), so that the security of data in the transmission and storage processes is ensured. The encrypted power quality data is subjected to segmentation processing through a sliding window algorithm, and the continuous data stream is segmented into data segments with fixed sizes. This segmentation approach facilitates parallel processing and flexible management. A hash value is calculated for each data segment, using a cryptographically secure hash algorithm such as SHA-256. The hash value is used as a checksum fingerprint of data integrity, and whether the data is tampered or not can be detected rapidly.
And finally, combining the hash value of the power quality data with the corresponding encrypted power quality data to form a power quality data packet. A checksum is calculated for each data packet, typically using a Cyclic Redundancy Check (CRC) algorithm. The addition of the checksum further enhances the integrity protection of the data. After this series of processing, a standardized power quality data block is obtained, which contains all the information of the original data and has the security, integrity and traceability. For example, a certain intelligent substation collects voltage data in one day, and the sampling frequency is 10kHz. The raw data volume is about 8.64 hundred million samples (24 hours x 3600 seconds x 10000 samples/second). Each data point has a time information of up to milliseconds after time stamping. Outlier detection found about 0.1% outliers, which were marked and temporarily removed from the main data stream. Data format unification converts voltage values into kilovolt (kV) units. The compression process reduces the amount of data to about 20% of the original, about 1.73 hundred million valid data points. The encrypted data is divided into 1728 data segments, each segment containing 10 ten thousand data points. And calculating the SHA-256 hash value for each fragment to obtain a 32-byte hash character string. Finally, 1728 data packets are combined into a complete standardized power quality data block, the total size of the data block is about 350MB, and the data block contains complete information of voltage quality of a transformer substation in one day, so that the authenticity and the integrity of data are ensured, and the efficiency of data processing and transmission is greatly improved.
In a specific embodiment, the process of executing step S102 may specifically include the following steps:
(1) Carrying out block structure encapsulation on the standardized power quality data block to obtain a candidate block, and carrying out digital signature processing on the candidate block to obtain a signed candidate block;
(2) Performing consensus verification on the signed candidate blocks to obtain verified blocks, and performing link processing on the verified blocks to obtain updated block chains;
(3) Performing distributed storage processing on the updated blockchain to obtain blockchain data stored by multiple nodes, and performing consistency check on the blockchain data stored by multiple nodes to obtain a consistency check result;
(4) Synchronizing the multi-node stored blockchain data according to the consistency check result to obtain synchronized blockchain data, and establishing an index of the synchronized blockchain data to obtain indexed blockchain data;
(5) And carrying out data integrity verification on the indexed blockchain data to obtain verified blockchain data, and taking the verified blockchain data as electric energy quality blockchain data.
Specifically, the data blocks are organized into a specific block structure, including a block header and a block body. The block header contains information such as version number, hash value of the previous block, timestamp, difficulty target and random number, and the block contains specific power quality data. The encapsulated structure is called a candidate block, which is the basic unit to be added to the blockchain. The candidate block is then digitally signed, typically using Elliptic Curve Digital Signature Algorithm (ECDSA). The source credibility and non-repudiation of the data are ensured, and the data are prevented from being tampered. The signed candidate block enters the consensus verification stage, which is a key step to ensure that all nodes in the blockchain network agree. In power quality traceability systems, a modified proof of work (PoW) consensus algorithm is employed. The improvement considers the credibility and history contribution of the nodes, ensures the decentralization, improves the efficiency and reduces the energy consumption. The verified block will be added to the end of the existing blockchain to form an updated blockchain. This process involves updating the pointer structure of the blockchain, ensuring that new blocks are cryptographically linked to previous blocks, maintaining the continuity and tamper resistance of the blockchain.
The updated blockchain data needs to be stored in a distributed manner among a plurality of nodes in the network. The storage mode enhances the redundancy and usability of the data and prevents the data from being lost due to single-point faults. In the storage process, the large-scale blockchain data is divided into smaller and manageable parts by adopting a slicing technology and is distributed to different nodes for storage. And immediately after the distributed storage is finished, consistency check is performed to ensure that the data on all nodes are kept consistent. And the consistency check adopts a Merkle tree structure, and the data difference among different nodes is compared rapidly. Based on the results of the consistency check, if the data is found to be inconsistent, a synchronization processing mechanism is started. The synchronization process uses an efficient data transfer protocol to transfer only inconsistent data blocks, rather than the entire blockchain, thereby saving bandwidth and time. After synchronization is completed, indexes are built for the block chain data, so that the data query efficiency is greatly improved. The index structure includes a time index, a geographical location index, a power quality parameter index, etc., so that a user can quickly locate and retrieve specific power quality data.
And finally, carrying out comprehensive data integrity verification on the indexed blockchain data. And re-calculating the hash value of each block by using a cryptographic hash function, and comparing the hash value with the stored hash value to ensure that the data is not tampered or damaged in the storage and processing processes. The block chain data passing verification becomes final power quality block chain data, and a reliable data basis is provided for subsequent power quality analysis and traceability. For example, 10 nodes in a blockchain network of a power company participate in the processing and storage of power quality data. During a day of operation, the system generated 144 standardized power quality data blocks (on average, one every 10 minutes). Each data block is approximately 2.5MB in size and contains detailed power quality parameters for that time period. The block structure encapsulation process organizes the data into candidate blocks, each block header size 80 bytes, containing metadata such as version number, time stamp, etc. The digital signature process adds 64 bytes of signature data for each candidate block. In the consensus verification stage, the improved PoW algorithm needs 30 seconds to finish verification of one block on average, and the efficiency is improved by 50% compared with the traditional PoW algorithm. After verification passes, new chunks are added to the blockchain, the chain length increases by 144 chunks, and the total data amount increases by about 360MB. The distributed storage process disperses the 360MB of newly added data across 10 nodes, each node adding an average of 36MB of storage burden. The consistency check finds that there are small differences in the data of the 2 nodes, and these differences are corrected within 5 seconds by the synchronization process. The index building process creates an index structure of about 1MB for the newly added data, and the data retrieval speed is remarkably improved. And finally, verifying the integrity to confirm that the data of all 144 newly added blocks are kept complete without tamper signs. The data processing and storage process of this day takes about 75 minutes in total, with most of the time being used for consensus verification and data synchronization, while the actual data transmission and storage time is only about 15% of the total time.
In a specific embodiment, the process of executing step S103 may specifically include the following steps:
(1) Performing intelligent contract coding on a preset electric energy quality evaluation rule to obtain an electric energy quality evaluation intelligent contract, and performing security audit on the electric energy quality evaluation intelligent contract to obtain an audited intelligent contract;
(2) Performing blockchain deployment on the audited intelligent contracts to obtain deployed intelligent contracts, and performing trigger condition setting on the deployed intelligent contracts to obtain executable intelligent contracts;
(3) The method comprises the steps of monitoring power quality block chain data in real time through an event monitoring mechanism to obtain trigger execution conditions, and calling executable intelligent contracts according to the trigger execution conditions to obtain executing intelligent contracts;
(4) The method comprises the steps of carrying out parameter transfer on an intelligent contract in execution to obtain an intelligent contract instance with parameters, and executing the intelligent contract instance with parameters through a block chain virtual machine to obtain an intelligent contract execution result;
(5) Formatting the intelligent contract execution result to obtain formatted evaluation data, and taking the formatted evaluation data as an electric energy quality evaluation result.
Specifically, intelligent contract coding is performed on preset electric energy quality evaluation rules, and evaluation standards such as voltage deviation, harmonic content, frequency fluctuation and the like are converted into executable code logic. The encoding uses a smart contract specific language to ensure that contracts can run in an Ethernet Virtual Machine (EVM) or the like environment. After the encoding is completed, comprehensive security audit is carried out on the intelligent contract for evaluating the electric energy quality, including methods such as static code analysis, dynamic test and formal verification, so as to find potential loopholes and logic errors. The auditing process focuses on common security problems such as reentry attacks, integer overflows and the like, and ensures safe operation under various conditions. The audited intelligent contracts are then deployed into the blockchain network. The deployment process includes compiling the contract code into byte code and sending it to the blockchain network through special transactions. Once deployed successfully, the Smart break-in date obtains a unique address for subsequent calls and interactions. Triggering conditions are set for the deployed smart contracts, which define when the contracts are automatically executed. In a power quality traceability system, the triggering condition may include a specific time interval, a power quality parameter exceeding a preset threshold, etc. After setting the trigger condition, the intelligent contract becomes an executable state, and the intelligent contract is ready to respond to the event meeting the condition at any time.
And monitoring the power quality block chain data in real time through an event monitoring mechanism. This mechanism sets up multiple listening nodes in the blockchain network, continuously scanning the newly added blockdata. When data meeting the preset trigger conditions is detected, for example, voltage dip exceeds 10% or harmonic distortion rate exceeds 5%, trigger execution conditions are generated. Based on these conditions, the system automatically invokes the corresponding executable intelligent contract, converting it to an executing state. And carrying out parameter transfer on the intelligent contracts in execution, and transmitting real-time power quality data as input parameters into the contracts. Parameters may include specific voltage values, current values, frequency values, etc. The parameter transfer process ensures that contracts can be calculated and judged based on up-to-date data. The parameterized smart contract instance is then committed to execution in the blockchain virtual machine. The virtual machine provides an isolated execution environment for the intelligent contract, and ensures the safety and consistency of the execution process. In the execution process, the contract analyzes and evaluates the input electric energy quality data according to preset logic to obtain an execution result.
And finally, formatting the execution result of the intelligent contract. The original execution output is converted into a standardized format, so that subsequent storage, transmission and analysis are facilitated. The formatting process may include data type conversion, unit unification, structured encapsulation, and the like. The processed data is used as a final power quality evaluation result, and a basis is provided for power quality tracing and analysis.
For example, an electric utility company deploys an intelligent contract for assessing the voltage quality of a distribution network. Contract encodings contain evaluation logic for voltage deviations, fluctuations and flicker. The security audit process discovers and remedies a potential integer overflow vulnerability. After deployment, the contract is set to trigger automatically once per hour, and execution is triggered when a voltage deviation exceeding + -7% is detected. In one execution, the event listening mechanism captures a voltage value of 207V (nominal voltage 220V) for a distribution line, triggering contract execution. The contract receives this voltage value as an input parameter and calculates a voltage deviation of-5.91%. The virtual machine executes the contract logic to determine that the voltage deviation, while outside the normal range (+ -5%), has not reached a severity level (+ -10%). The execution results include the percentage deviation, duration, and severity ratings. The formatting process converts these results into JSON format, containing time stamps, location information, specific parameter values, and evaluation conclusions. This result is recorded on the blockchain and a low level alarm is triggered to alert the service personnel to the voltage condition of the line.
In a specific embodiment, the process of executing step S104 may specifically include the following steps:
(1) Performing data classification processing on the power quality evaluation result to obtain classified power quality data, and performing trend analysis on the classified power quality data through a time sequence analysis algorithm to obtain power quality trend data;
(2) Carrying out geographical distribution analysis on the power quality trend data to obtain power quality geographical distribution characteristics, and carrying out thermodynamic diagram generation processing on the power quality geographical distribution characteristics to obtain a power quality thermodynamic diagram;
(3) Performing parameter association analysis on the classified power quality data to obtain a power quality parameter association matrix, and performing network diagram generation processing on the power quality parameter association matrix to obtain a power quality association network diagram;
(4) Performing abnormal mode identification on the power quality trend data to obtain power quality abnormal data, and performing threshold comparison processing on the power quality abnormal data to obtain an abnormal alarm;
(5) And carrying out integrated processing on the power quality trend data, the power quality thermodynamic diagram, the power quality association network diagram and the abnormal alarm to obtain a comprehensive analysis result, and taking the comprehensive analysis result as a power quality analysis report.
Specifically, the electric energy quality evaluation result is carefully analyzed through data classification processing, so that the data can be classified into different categories according to different electric energy quality indexes such as voltage fluctuation, frequency deviation, harmonic content and the like, and classified electric energy quality data can be obtained. The classification processing aims at defining the characteristics of different types of power quality problems, so that the subsequent data analysis is more targeted, and the overall data processing efficiency is improved. After the classified data are obtained, the time sequence analysis algorithm is adopted to carry out trend analysis on the classified data, and the change rule of the electric energy quality can be mined. For example, the time series analysis may use an autoregressive moving average (ARMA) model to set a suitable time window, calculate an expected trend of future power quality changes based on historical data, and generate power quality trend data. And then, carrying out geographical distribution analysis on the power quality trend data, so that the distribution characteristics of the power quality of different areas can be revealed. By introducing a Geographic Information System (GIS) technology, using geographic coordinates of the power quality monitoring points, the trend data is mapped into a geographic space, so that geographic distribution characteristics of power quality are obtained, and a thermodynamic diagram is further generated. The thermodynamic diagram visually displays the quality of the electric energy of different areas through the change of the color depth, and is helpful for a decision maker to quickly identify the problem area. For example, if a power quality thermodynamic diagram of a certain area shows more red areas, it indicates that the power quality of the area is generally low, and attention and improvement are required.
When the parameter association analysis is carried out on the power quality data, a power quality parameter association matrix is formed, and the matrix reveals the relation among different power quality parameters. For example, the relationship between the voltage fluctuation and the harmonic content can be judged by calculating the pearson correlation coefficient between the parameters, so as to obtain a matrix formed by the parameter relationship. On the basis, the power quality parameter association matrix is visualized, a power quality association network diagram is generated, nodes in the network diagram represent power quality parameters, and the intensity of edges represents the association degree between the parameters. Not only can help identify parameters which mainly affect the quality of the electric energy, but also can provide guidance for further optimization measures. The process of identifying abnormal modes of the power quality trend data is also of great importance. By setting a reasonable threshold value and utilizing a machine learning algorithm such as a Support Vector Machine (SVM) or an isolated forest algorithm to perform anomaly detection on the power quality trend data, potential power quality problems can be effectively identified. For example, when the voltage fluctuation exceeds a set threshold, the system will trigger an anomaly alarm to prompt the relevant staff to intervene and process in time.
And finally, integrating the power quality trend data, the power quality thermodynamic diagram, the power quality association network diagram and the abnormal alarm to form a comprehensive analysis result. The comprehensive analysis result is used as a power quality analysis report and provides reference for related decisions. Through the analysis, the formed report not only contains the current state of the power quality, but also covers potential risks and improvement directions, and guides the optimization and improvement of the power quality.
For example, assuming that in a certain evaluation, the classification of the voltage fluctuation data shows that 50% of the samples have a fluctuation amplitude greater than + -5% (set normal range), the fluctuation is found to be ascending in a specific period of time after trend analysis, in combination with geographical distribution analysis, the fluctuation is found to be concentrated in several important power supply areas, and the thermodynamic diagram shows that these areas are poor in power quality. Through parameter correlation analysis, the fluctuations are found to have higher correlation with harmonic content, and an alarm is triggered in abnormal pattern recognition. After the data are integrated, the formed power quality analysis report indicates the important improvement direction for the power company, so that the power company is prompted to take targeted measures, and finally, the continuous improvement of the power quality is realized.
In a specific embodiment, the process of executing step S105 may specifically include the following steps:
(1) Performing problem classification processing on the abnormal alarm to obtain classified abnormal problems, and performing preliminary diagnosis analysis on the classified abnormal problems to obtain a preliminary diagnosis report;
(2) The method comprises the steps of performing multiparty sharing processing on a preliminary diagnosis report through a blockchain intelligent contract to obtain shared diagnosis information, and performing authority management on the shared diagnosis information to obtain accessible diagnosis data;
(3) Collecting accessible diagnostic data in multiple ways to obtain multiple-way diagnostic opinion sets, and summarizing the multiple-way diagnostic opinion sets through a consensus algorithm to obtain comprehensive diagnostic results;
(4) Performing similar case matching on the comprehensive diagnosis result to obtain a candidate solution set, and evaluating and screening the candidate solution set through a multi-criterion decision algorithm to obtain a problem solution;
(5) And recording the execution process of the problem solution in real time to obtain execution process data, and performing blockchain storage processing on the execution process data to obtain a processing record.
Specifically, a machine learning algorithm, such as a Support Vector Machine (SVM) or decision tree, is used to automatically classify alarms according to categories such as voltage anomalies, harmonic interference, frequency deviations, etc. The classified abnormal problems immediately enter a preliminary diagnosis analysis stage, and each type of problems is rapidly evaluated by using an expert system and a rule base to generate a preliminary diagnosis report. This report contains descriptions of the problem, possible causes, and preliminary advice. Next, the preliminary diagnostic report is subject to multiparty sharing processing via blockchain intelligence contracts. The smart contracts automatically encrypt and distribute diagnostic reports to predefined stakeholders, such as utility companies, equipment suppliers, and regulatory authorities. In the sharing process, a role-based access control (RBAC) mechanism is adopted to strictly manage the rights of the shared diagnostic information. Different roles are given different access rights, ensuring that sensitive information is only visible to a particular user, thereby obtaining accessible diagnostic data.
The collection of multiple opinion regarding the accessible diagnostic data is the next key step. Through collaboration tools on the blockchain platform, the parties are allowed to submit their diagnostic comments and suggestions. Each opinion is recorded on the blockchain, ensuring transparency and traceability of the process. The collected multiparty diagnosis opinion sets are then summarized by a consensus algorithm. The consensus algorithm adopted in the method is different from the block chain consensus, but is a decision fusion algorithm based on expert weights, the experience and the historical accuracy of each expert are considered, the different findings are weighted and averaged, and finally the comprehensive diagnosis result is obtained. The comprehensive diagnosis results then enter a similar case matching stage. The current problem is compared to records in the historical case base using case-based reasoning (CBR) techniques. The matching process takes multiple features, such as problem types, symptom descriptions, environmental factors, etc., into account, and screens out the most relevant historical cases by calculating similarity scores to form a candidate solution set. And evaluating and screening the candidate schemes through a multi-criterion decision algorithm. The algorithm comprehensively considers a plurality of factors such as feasibility, cost, time efficiency, potential risk and the like of the schemes, calculates a comprehensive score for each scheme, and finally selects an optimal problem solution.
Finally, the execution of the selected problem solution is recorded in real time. This includes recording details of each execution step, the resources used, the obstacles encountered, and the actual effects. The execution process data is uploaded to the blockchain network in real time, and becomes a non-tamperable blockchain record after encryption and consensus verification. This approach ensures transparency and traceability of the overall problem-solving process, providing valuable data support for subsequent analysis and optimization.
For example, a certain distribution network presents serious voltage fluctuation problems in one operation. The system first classifies this anomaly alarm as a "voltage quality problem" and presumes, through preliminary diagnostic analysis, that it may be due to sudden load changes or grid equipment failure. This preliminary diagnostic report is shared to the grid operators, equipment manufacturers, and local authorities through smart contracts. The expert of the grid operator accesses the complete report through the rights management system, whereas the device manufacturer can only see part of the information related to it. In the multiparty opinion collection phase, three-party experts respectively put forward their opinion that the grid operators suspect to be a distribution transformer tap switching problem, the equipment manufacturers consider that the Static Var Compensator (SVC) control system that they may produce is abnormal, and the regulatory authorities recommend to check if there are large industrial users to use electricity illegally. The opinions are summarized through a consensus algorithm, the algorithm gives different weights according to the past problem solving accuracy of each party, and finally the obtained comprehensive diagnosis result tends to be the most likely to be a problem of the distribution transformer. The system then matches to 3 similar historical cases in a case base, involving transformer tap switching failure, SVC failure, and large customer sudden load, respectively. The multi-criterion decision algorithm comprehensively evaluates the solutions of the three conditions, and finally selects the solution for checking and maintaining the tap switching mechanism of the transformer in consideration of maintenance difficulty, influence range and solution time.
During execution, each operational step of the maintenance team is recorded from power down, inspection of the tap switch mechanism, discovery of mechanical stuck, replacement of components, to final restoration of power. These data are uploaded to the blockchain in real time, forming a complete problem-solving process record. The whole process lasts for 4 hours, the problem of voltage fluctuation is finally successfully solved, and the voltage quality index is restored to the national standard range.
In a specific embodiment, the process of executing step S106 may specifically include the following steps:
(1) Integrating the electric energy quality block chain data, the electric energy quality analysis report and the processing record to obtain a comprehensive electric energy quality data set, and performing time sequence reconstruction on the comprehensive electric energy quality data set to obtain time-sequence electric energy quality data;
(2) Building a relational network of the time-ordered power quality data to obtain a power quality knowledge graph, and performing semantic analysis on the power quality knowledge graph to obtain power quality semantic association data;
(3) Carrying out causal chain tracking on the power quality semantic association data to obtain power quality problem root data, and carrying out clustering treatment on the power quality problem root data to obtain power quality problem type distribution;
(4) Trend analysis is carried out on the time-ordered power quality data to obtain a power quality development trend prediction result, and risk assessment is carried out on the power quality development trend prediction result to obtain a potential risk point identification result;
(5) And comprehensively analyzing the type distribution of the power quality problems and the identification result of potential risk points to obtain a power quality improvement direction, and carrying out scheme generation processing on the power quality improvement direction to obtain a power quality improvement suggestion and an optimization scheme.
Specifically, by sorting the power quality blockchain data, the first step of integration is to combine the data with the power quality analysis report and the process record to form a comprehensive power quality data set. The data set not only contains the power quality data from different links, but also covers related analysis results and processing information, thereby providing comprehensive basic data for subsequent analysis. By flushing and formatting this information, consistency and usability of the data is ensured, e.g. using a uniform time stamp and data format to process the data, making it suitable for time series analysis. And then, carrying out time sequence reconstruction on the comprehensive power quality data set so as to analyze the change condition of the power quality along with time. In the process, a time sequence algorithm is utilized to construct the power quality data with time sequence characteristics by carrying out interpolation, smoothing and trend analysis on the power quality parameters at different time points. The reconstruction process can reveal long-term trends and short-term fluctuations of the power quality, so as to provide basic data for subsequent relational network construction.
When a relational network is constructed, a correlation analysis method in graph theory is adopted to generate a power quality knowledge graph. By analysing the correlation between the power quality parameters, interactions between different parameters, such as the relationship between voltage fluctuations and load variations, can be found. After the knowledge graph is constructed, semantic analysis is further carried out, potential power quality associated data are extracted, and the data can be used for identifying potential sources of power quality problems to form a complete power quality knowledge system. And when the root data is identified, tracking the occurrence cause of the power quality problem by adopting a causal chain tracking method. This process involves extracting relevant parameters and events from identified power quality problems and analyzing the interactions between these factors by building a causal relationship model. For example, if an abnormality in the power quality is found, it may be caused by a sudden increase in load during a certain period. By clustering the factors, the distribution of the power quality problem types can be obtained, and the most common power quality problems and the occurrence conditions thereof can be identified.
When the power quality trend analysis is performed, firstly, based on time-series power quality data, regression analysis or a time sequence prediction model is adopted to analyze the power quality trend. These models can capture trends and seasonal fluctuations in the data. For example, if the data indicates that the power quality fluctuates in summer due to an increase in air conditioning load, a change in power quality of several months in the future can be predicted. In addition, risk assessment is performed on the predicted results to identify potential risk points, such as the risk of power quality failure during certain peak load periods. And finally, comprehensively analyzing the type distribution of the power quality problems and the identification result of potential risk points to form a power quality improvement direction. This process identifies links that need to be optimized by correlating the identified power quality problems with predicted risk points. For example, if power quality fluctuations are found to be concentrated mainly in a specific period of time, specific improvement suggestions may be made by optimizing load management or device upgrades. For these improvements, an optimization scheme of the system is formed, and the stability and reliability of the power quality in the future are ensured.
For example, a certain analysis shows that during peak summer hours, the voltage fluctuation rate is 15%, and after analysis of the integrated power quality data set, the load during that period was found to increase by 30% on average. By establishing a causal link, it is determined that voltage fluctuations are mainly caused by load increases, at which time the impact of load peaks on power quality can be reduced by implementing load management strategies, e.g. optimizing the distribution of peak loads or introducing intelligent scheduling systems. The process not only improves the stability of the electric energy quality, but also provides data support for the operation of the electric power system, and finally forms a practical improvement scheme.
The above examples are provided for illustrating the technical aspects of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the examples, it should be understood that modifications and equivalents may be made to the specific embodiments of the present invention without departing from the spirit and scope of the present invention, and it should be construed as being encompassed by the scope of the claims of the present invention.