Diversified evaluation method and system for power system based on log fusion

Abstract

The invention provides an electric power system diversified evaluation method and system based on log fusion, and the method comprises the steps: obtaining logs corresponding to all kinds of index data in an electric power system, the index data comprises business index data, application index data, data interaction index data, basic software and hardware index data and infrastructure index data; performing log fusion on each log through a log fusion technology to obtain a fused log; according to various index data in the fusion log, performing diversified evaluation analysis on the power system to obtain a diversified evaluation result, the method comprises the steps of evaluating the economical efficiency and reliability of a power grid operation strategy, evaluating the application performance of power grid operation and maintenance management software and a scheduling decision support system, evaluating the data processing efficiency of a power grid, evaluating the basic operation and support capability of a power grid guarantee system, and evaluating the construction quality, maintenance state and upgrading capability of power facilities. And the accuracy and comprehensiveness of power system evaluation are improved.

Description

Technical Field

[0001] The present application relates to the field of power system software technology, and in particular to a method and system for diversified evaluation of a power system based on log fusion. Background Art

[0002] In the power industry, with the in-depth development of informatization and intelligence, multiple software systems are widely used to meet different business needs. These software systems, including but not limited to monitoring, scheduling, operation and maintenance, etc., will generate a large amount of log data during operation, which contains important information such as system operation status and abnormal alarms. However, since these software systems are often developed by different manufacturers and use different technical standards and architectures, there are large differences in the log data format, storage method and processing mechanism between them. This creates a problem: when interrelated alarm information appears between systems, due to the existence of information islands, one system may not be able to directly obtain the alarm data of another system, and thus cannot form a comprehensive alarm monitoring and analysis. Summary of the invention

[0003] In response to the above technical problems, the present application provides a diversified evaluation method and system for a power system based on log fusion, which conducts a comprehensive evaluation of the power system according to a variety of log data, thereby improving the accuracy and comprehensiveness of the evaluation of the power system.

[0004] In a first aspect, the present application provides a method for diversified evaluation of a power system based on log fusion, comprising:

[0005] Obtain logs corresponding to various indicator data in the power system, including business indicator data, application indicator data, data interaction indicator data, basic software and hardware indicator data, and infrastructure indicator data;

[0006] Performing log fusion on each of the logs using a preset log fusion technology to obtain a fused log;

[0007] According to various indicator data in the fusion log, the power system is subjected to diversified evaluation and analysis to obtain diversified evaluation results. The diversified evaluation and analysis include evaluating the economy and reliability of the power grid operation strategy, evaluating the application performance of the power grid operation and maintenance management software and the dispatching decision support system, evaluating the data processing efficiency of the power grid, evaluating the basic operation and support capabilities of the power grid security system, and evaluating the construction quality, maintenance status and upgrade capabilities of power facilities.

[0008] The embodiment of the present application provides a method for diversified evaluation of power systems based on log fusion, which obtains logs corresponding to various types of indicator data and uses log fusion technology to fuse the logs to obtain fused logs, effectively integrating indicator data of the business layer, application layer, data interaction layer, basic software and hardware layer, and infrastructure layer. Then, according to the various types of indicator data in the fused log, the power system is subjected to diversified evaluation and analysis to obtain diversified evaluation results, thereby achieving effective use of integrated data, and comprehensively evaluating the power system based on a variety of log data, thereby improving the accuracy and comprehensiveness of the evaluation of the power system.

[0009] Furthermore, the logs corresponding to various indicator data in the power system are obtained, including:

[0010] Obtaining a log corresponding to the business indicator data through the operation management software of the power system;

[0011] Obtaining logs corresponding to the application indicator data through the operation and maintenance management software and the dispatch decision support system of the power system;

[0012] Obtaining logs corresponding to the data interaction indicator data through data collection, processing and analysis tools of the power system;

[0013] Obtaining logs corresponding to the basic software and hardware indicator data through a monitoring tool of the power system;

[0014] The logs corresponding to the infrastructure indicator data are obtained through the management system of the power physical infrastructure and technical support facilities of the power system.

[0015] The embodiment of the present application provides a method for obtaining various types of logs. The logs corresponding to the business indicator data can be obtained through the operation and management software of the power system, which can track terminal results such as power supply reliability, customer satisfaction and economic operation. The logs corresponding to the application indicator data usually come from the operation and maintenance management software and the scheduling decision support system, which can provide information about system functionality, user experience and business process efficiency. The logs corresponding to the data interaction indicator data rely on data collection, processing and analysis tools, which ensure the accuracy and timeliness of information flow. The logs corresponding to the basic software and hardware indicator data can be obtained by monitoring the performance of hardware resources and software platforms. These monitoring tools can provide information such as system availability, resource utilization and maintenance efficiency. The logs corresponding to the infrastructure indicator data need to be collected through the management system of the power physical infrastructure and technical support facilities. These systems focus on the construction and operation of substations, transmission lines and data centers. By obtaining log information at all levels, the operation status of the power system can be reflected from all aspects, providing data support for the subsequent diversified evaluation of the power system.

[0016] In a possible implementation, performing log fusion on each of the logs by using a preset log fusion technology to obtain a fused log includes:

[0017] Determine a first log to be currently merged from each of the logs;

[0018] Calculate the affinity between the first log and each of the other logs using a preset text processing algorithm;

[0019] Selecting a log with the largest affinity from the other logs as the second log to be merged;

[0020] Determining a log fusion strategy according to the affinity between the first log and the second log;

[0021] Merging the first log and the second log into a third log according to a log fusion strategy;

[0022] The first log and the second log to be merged are determined again from each of the logs for merging until all the logs are merged to obtain the merged log.

[0023] The embodiment of the present application provides a method for log fusion of various types of logs, wherein log fusion refers to merging log information from multiple sources or multiple time periods into a unified log stream for easy analysis and processing. Affinity is an indicator for measuring the similarity or correlation between different logs. In the embodiment of the present application, each fusion selects the first log and the second log from each log for pairwise fusion until all logs are fused into a fusion log. In each fusion process, the first log to be fused is first determined from each log and the affinity between the first log and the other logs is calculated respectively, then the second log to be fused is determined according to each affinity and the log fusion strategy is determined according to the affinity, and finally the first log and the second log are fused according to the log fusion strategy, so as to achieve effective integration of different types of logs, and log information from different systems and different levels can be gathered in one place. By adopting a unified log format and standardized processing flow, the system can more comprehensively capture relevant information about the system operation status.

[0024] Furthermore, the using a preset text processing algorithm to calculate the affinity between the first log and each of the other logs includes:

[0025] Counting the frequency matching between the first log and each of the other logs, where the frequency matching is the frequency of occurrence of a certain keyword at a preset level;

[0026] Counting the time matching degree between the first log and each of the other logs, where the time matching degree is the frequency of API being accessed within a preset time window;

[0027] According to the frequency matching and time matching between the first log and each of the other logs, a preset affinity function is used to calculate the affinity between the first log and each of the other logs.

[0028] The embodiment of the present application provides a method for calculating the affinity of logs, and the similarity between logs is calculated by two indicators: frequency matching and time matching. Regarding frequency matching, in different logs, for example, if two instances ωx and ωy are matched, then the process paths corresponding to the two instances of ωx and ωy (usually referring to the sequence and path of a series of steps, stages or links experienced from the beginning to the end when completing a task or activity) appear the same number of times in the two logs. For one-to-many, many-to-one or many-to-many matching relationships between instances, a single instance pair is examined, and the number of occurrences of the corresponding process paths may be different, but from the global perspective of the individual, their number of occurrences should be similar in statistical terms. Frequency matching is to measure the similarity of the number of occurrences of process paths between all matching relationships in the individual. Regarding time matching, when two processes with business interaction are executed, there is an intersection in time. At this time intersection, one process is reflected in the log to trigger the execution of another process, that is, the duration interval of the two process instances that can be matched has a parallel part. Therefore, the affinity of an individual can be measured by measuring whether there is an overlap in the duration intervals of the process instances between all matching relationships. Furthermore, the embodiment of the present application uses the affinity function to further combine the frequency matching and the time matching to calculate the affinity between logs, thereby improving the rationality and accuracy of the affinity calculation, and thus improving the accuracy of log fusion.

[0029] Further, determining the log fusion strategy according to the affinity between the first log and the second log includes:

[0030] When the affinity is greater than or equal to a first threshold and less than 1, determining that the log fusion strategy is to directly merge the first log and the second log;

[0031] When the affinity is greater than or equal to the second threshold and less than the first threshold, determining the log fusion strategy to retain common log entries during the merging process and marking or annotating different parts;

[0032] When the affinity is less than a second threshold, the fusion of the first log and the second log is abandoned, the current first log is marked, and a new first log is selected.

[0033] The embodiment of the present application provides a method for selecting a log fusion strategy, which sets a first threshold and a second threshold in advance. When the affinity is greater than or equal to the first threshold and less than 1, it means that the two logs are very similar, and the first log can be directly merged with the second log; when the affinity is greater than or equal to the second threshold and less than the first threshold, it means that the two logs have a high affinity, but there are also some differences. Therefore, in the merging process, common log entries are retained, and different parts are marked or annotated; when the affinity is less than the second threshold, it means that the two logs are basically different and difficult to merge, so the fusion of the current first log and the second log is abandoned, and the first log is reselected. The embodiment of the present application provides three different fusion strategies based on the affinity of the first log and the second log. For high-affinity logs, direct fusion can be used to improve the efficiency of log fusion, while avoiding fusion between low-affinity logs, improving the accuracy of log fusion.

[0034] In a possible implementation, the evaluating the economy and reliability of the power grid operation strategy includes:

[0035] Extracting the business indicator data from the fusion log, the business indicator data including the number of power outages, the duration of power outages, the fault recovery time, user feedback information, complaint rate, satisfaction score, power cost, power sales revenue and maintenance cost;

[0036] According to the power outage times, power outage durations and fault recovery time, the average power outage time of the system, the average power outage frequency of the system and the average power outage time of users are calculated to evaluate the power supply reliability of the power system;

[0037] Analyze the number of complaints and their changing trends based on the user feedback information, complaint rate, and satisfaction score, and then evaluate the customer satisfaction of the power system;

[0038] Calculate the return on investment of each project based on the electricity cost, electricity sales income and maintenance costs, and then evaluate the economic benefits of the power system;

[0039] The economy and reliability of the grid operation strategy are evaluated in combination with the power supply reliability, customer satisfaction and economic benefits of the power system.

[0040] In an embodiment of the present application, business indicator data is extracted from the fusion log, and the power supply reliability, customer satisfaction and economic benefits of the power system are evaluated based on the business indicator data. The economy and reliability of the power grid operation strategy are comprehensively evaluated in combination with the above three indicators to achieve an evaluation of the business level of the power system.

[0041] In a possible implementation, the evaluation of the application performance of the power grid operation and maintenance management software and the dispatch decision support system includes:

[0042] Extracting the application indicator data from the fusion log, the application indicator data including system operation status, response time, error rate, user experience satisfaction data, user behavior pattern and system abnormality information;

[0043] Formulate a functional test plan based on the system operation status, response time and error rate to verify whether the system meets the preset functional requirements, thereby evaluating the system functionality of the power system;

[0044] According to the user experience satisfaction data, the user satisfaction is quantified using a Likert scale, and the average user satisfaction score and user satisfaction distribution are calculated, thereby evaluating the user experience satisfaction of the power system;

[0045] According to the user behavior pattern and system abnormality information, based on the preset key business process metrics, calculate the business process efficiency value of the power system;

[0046] Combined with the system functionality, user experience satisfaction and business process efficiency value of the power system, the application performance of the power grid operation and maintenance management software and the dispatching decision support system is evaluated.

[0047] In an embodiment of the present application, application indicator data is extracted from the fusion log, and the system functionality, user experience satisfaction, and business process efficiency value of the power system are evaluated based on the application indicator data. The application performance of the power grid operation and maintenance management software and the dispatching decision support system are comprehensively evaluated in combination with the above three indicators to achieve an evaluation of the application level of the power system.

[0048] In a possible implementation, the step of evaluating the data processing efficiency of a power grid includes:

[0049] Extracting the data interaction indicator data from the fusion log;

[0050] Verifying the integrity and accuracy of the data interaction indicator data through preset verification rules, and then evaluating the data processing stability of the power system;

[0051] According to the difference between the timestamps corresponding to the respective data in the data interaction indicator data, the real-time performance of data processing of the power system is evaluated;

[0052] Performing anomaly detection on the data interaction indicator data according to a preset security audit tool, thereby evaluating the data processing security of the power system;

[0053] The data processing efficiency of the power grid is evaluated by combining the data processing stability, data processing real-time performance and data processing security of the power system.

[0054] In an embodiment of the present application, data interaction indicator data is extracted from the fusion log, and the data processing stability, real-time performance and security of the power system are evaluated based on the data interaction indicator data. The data processing efficiency of the power grid is comprehensively evaluated in combination with the above three indicators to achieve an evaluation of the data interaction level of the power system.

[0055] In a possible implementation, the evaluation of the basic operation and support capabilities of the power grid security system includes:

[0056] Extracting the basic software and hardware indicator data from the fusion log, wherein the basic software and hardware indicator data includes historical operation data of software and hardware, hardware resource usage data, and maintenance record data;

[0057] Detecting potential faults and performance bottlenecks of the power system based on the historical operation data, and then evaluating the availability of the power system;

[0058] According to the hardware resource usage data, evaluating resource utilization of the power system through data analysis and visual display;

[0059] Analyzing the fault location, fault frequency and maintenance time according to the maintenance record data, and then evaluating the maintenance efficiency of the power system;

[0060] Combined with the availability, resource utilization and maintenance efficiency of the power system, the basic operation and support capabilities of the power grid security system are evaluated.

[0061] In an embodiment of the present application, basic software and hardware indicator data are extracted from the fusion log, and the availability, resource utilization and maintenance efficiency of the power system are evaluated based on the basic software and hardware indicator data. The basic operation and support capabilities of the power grid security system are comprehensively evaluated in combination with the above three indicators, thereby realizing the evaluation of the basic software and hardware layer of the power system.

[0062] In a possible implementation, the evaluation of the construction quality, maintenance status and upgrade capability of power facilities includes:

[0063] Extracting the infrastructure indicator data from the fusion log, the infrastructure indicator data including operation status data of the substation, voltage data of the substation, current data of the substation, power data of the substation, load data of the transmission line, temperature data of the transmission line, and fault status information of the transmission line;

[0064] Analyze the infrastructure indicator data based on big data technology to identify the operation modes and trends of substations, transmission lines and data centers;

[0065] Based on GIS technology, the spatial location information of power grid facilities is combined with the infrastructure indicator data to determine the coverage of the power grid, the distribution of facilities and the potential impact area;

[0066] Based on sensors and IoT technology, obtain real-time status data of power facilities;

[0067] Predicting the remaining life, maintenance requirements and failure probability of each device of the power facility based on the real-time status data and the infrastructure indicator data through a preset prediction model;

[0068] The construction quality, maintenance status and upgrade capabilities of power facilities are evaluated based on the operating modes and trends of the substations, transmission lines and data centers, the coverage of the power grid, the distribution of facilities and potential impact areas, the remaining life, maintenance requirements and failure probability of each equipment in the power facilities.

[0069] In an embodiment of the present application, infrastructure indicator data is extracted from the fusion log, and data analysis is performed on the infrastructure indicator data using big data technology, GIS technology, sensor and Internet of Things technology, and machine learning technology, respectively, to obtain the operating modes and trends of the substations, transmission lines, and data centers, the coverage of the power grid, the distribution of facilities and potential impact areas, the remaining life of each equipment of the power facilities, maintenance requirements, and failure probabilities. The construction quality, maintenance status, and upgrade capabilities of the power facilities are comprehensively evaluated in combination with the above indicators, thereby realizing the evaluation of the infrastructure layer of the power system.

[0070] In a second aspect, accordingly, the present application provides a power system diversified assessment system based on log fusion, including an acquisition module, a log fusion module and an assessment module;

[0071] The acquisition module is used to acquire logs corresponding to various indicator data in the power system, wherein the indicator data includes business indicator data, application indicator data, data interaction indicator data, basic software and hardware indicator data, and infrastructure indicator data;

[0072] The log fusion module is used to perform log fusion on each of the logs through a preset log fusion technology to obtain a fused log;

[0073] The evaluation module is used to perform diversified evaluation and analysis on the power system according to various indicator data in the fusion log to obtain diversified evaluation results. The diversified evaluation and analysis includes evaluating the economy and reliability of the power grid operation strategy, evaluating the application performance of the power grid operation and maintenance management software and the dispatching decision support system, evaluating the data processing efficiency of the power grid, evaluating the basic operation and support capabilities of the power grid security system, and evaluating the construction quality, maintenance status and upgrade capabilities of power facilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 : A flow chart of a method for diversified evaluation of a power system based on log fusion provided in an embodiment of the present application.

[0075] Figure 2 : A schematic diagram of the hierarchical structure of indicator data in a diversified evaluation method of a power system based on log fusion provided in an embodiment of the present application.

[0076] Figure 3 : A flow chart of log fusion of a power system diversified assessment method based on log fusion provided in an embodiment of the present application.

[0077] Figure 4 : A flow chart of using affinity function to calculate affinity in a power system diversification assessment method based on log fusion provided in an embodiment of the present application.

[0078] Figure 5 : A structural diagram of a power system diversified assessment system based on log fusion provided in an embodiment of the present application. DETAILED DESCRIPTION

[0079] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0080] It should be noted that the step numbers in the text are only for the convenience of explanation of the specific embodiments and do not serve to limit the order in which the steps are executed. In the description of this application, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features.

[0081] Embodiment 1:

[0082] like Figure 1 As shown, embodiment 1 provides a power system diversified evaluation method based on log fusion, including steps S1-S3:

[0083] Step S1, obtaining logs corresponding to various indicator data in the power system, wherein the indicator data includes business indicator data, application indicator data, data interaction indicator data, basic software and hardware indicator data, and infrastructure indicator data;

[0084] Step S2: fusing the logs using a preset log fusion technology to obtain a fused log;

[0085] Step S3, performing diversified evaluation and analysis on the power system according to various indicator data in the fusion log to obtain diversified evaluation results, wherein the diversified evaluation and analysis includes evaluating the economy and reliability of the power grid operation strategy, evaluating the application performance of the power grid operation and maintenance management software and the dispatching decision support system, evaluating the data processing efficiency of the power grid, evaluating the basic operation and support capabilities of the power grid security system, and evaluating the construction quality, maintenance status and upgrade capabilities of power facilities.

[0086] The embodiment of the present application provides a method for diversified evaluation of power systems based on log fusion, which obtains logs corresponding to various types of indicator data and uses log fusion technology to fuse the logs to obtain fused logs, effectively integrating indicator data of the business layer, application layer, data interaction layer, basic software and hardware layer, and infrastructure layer. Then, according to the various types of indicator data in the fused log, the power system is subjected to diversified evaluation and analysis to obtain diversified evaluation results, thereby achieving effective use of integrated data, and comprehensively evaluating the power system based on a variety of log data, thereby improving the accuracy and comprehensiveness of the evaluation of the power system.

[0087] In a preferred embodiment, the index data of the power system is divided into multiple levels, such as Figure 2 As shown in the figure, it includes business layer, application layer, data interaction layer, basic software and hardware layer, and infrastructure layer. This division makes the indicator system more comprehensive and can integrate and optimize the power management system from multiple angles.

[0088] At the business level, the focus is on the operation and management of the power system and the quality of service. The indicators at this level are mainly used to evaluate the impact of power grid operation strategies on economy and reliability. Specifically, the indicators at the business level focus on the terminal results of power grid services, such as power supply reliability, customer satisfaction, and economic operation. These indicators can help understand the overall performance of the power system, so as to adjust and optimize the operation strategy.

[0089] At the application layer, the focus is on the performance of applications such as operation and maintenance management software and dispatch decision support systems. The indicators at this layer are mainly used to ensure that these tools can effectively support business operations. Specifically, the indicators at the application layer include system functionality, user experience, and the efficiency of business processes supported by the application. These indicators can help evaluate the performance of the application layer, thereby ensuring the efficient operation of the power management system.

[0090] At the data interaction layer, the focus is on the efficiency of data collection, processing, and analysis. The indicators at this layer are mainly used to ensure the accuracy and timeliness of information flow to support the decision-making of upper-layer businesses and applications. Specifically, the indicators of the data interaction layer cover data integrity, real-time, security, and availability. These indicators can help evaluate the performance of the data interaction layer, thereby ensuring the smooth and accurate flow of information.

[0091] At the basic software and hardware layer, the focus is on the stability, performance, and scalability of hardware resources and software platforms. The indicators at this layer are mainly used to ensure the basic operation and support capabilities of the system. Specifically, the indicators of the basic software and hardware layer include system availability, resource utilization, maintenance efficiency, etc. These indicators can help evaluate the performance of the basic software and hardware layer, thereby ensuring the stable operation of the system.

[0092] At the infrastructure layer, the focus is on the construction and maintenance of the physical infrastructure and technical support facilities of the power system. The indicators at this layer are mainly used to evaluate the construction quality, maintenance status, and upgrade capabilities of the facilities. Specifically, the indicators at the infrastructure layer focus on the construction and operation status of substations, the length and capacity of transmission lines, and the scale and capacity of data centers. These indicators can help evaluate the performance of the infrastructure layer, thereby ensuring the physical support of the power system.

[0093] The embodiment of the present application can comprehensively evaluate and optimize the performance of the power management system by dividing the indicator system into multiple levels. The indicators of each level are interrelated and together constitute a complete indicator system.

[0094] Furthermore, in step S1, the logs corresponding to various indicator data in the power system are obtained, including:

[0095] Obtaining a log corresponding to the business indicator data through the operation management software of the power system;

[0096] Obtaining logs corresponding to the application indicator data through the operation and maintenance management software and the dispatch decision support system of the power system;

[0097] Obtaining logs corresponding to the data interaction indicator data through data collection, processing and analysis tools of the power system;

[0098] Obtaining logs corresponding to the basic software and hardware indicator data through a monitoring tool of the power system;

[0099] The logs corresponding to the infrastructure indicator data are obtained through the management system of the power physical infrastructure and technical support facilities of the power system.

[0100] In a preferred embodiment, the indicator data of the business layer can be obtained through the operation and management software of the power system, which can track terminal results such as power supply reliability, customer satisfaction and economic operation. The indicator data of the application layer usually comes from the operation and maintenance management software and the dispatch decision support system, which can provide information about system functionality, user experience and business process efficiency. The indicator data of the data interaction layer depends on data collection, processing and analysis tools, which ensure the accuracy and timeliness of information flow. The indicator data of the basic software and hardware layer can be obtained by monitoring the performance of hardware resources and software platforms. These monitoring tools can provide information such as system availability, resource utilization and maintenance efficiency. The indicator data of the infrastructure layer needs to be collected through the management systems of the power physical infrastructure and technical support facilities, which focus on the construction and operation of substations, transmission lines and data centers.

[0101] The embodiment of the present application provides a method for obtaining various types of logs. The logs corresponding to the business indicator data can be obtained through the operation and management software of the power system, which can track terminal results such as power supply reliability, customer satisfaction and economic operation. The logs corresponding to the application indicator data usually come from the operation and maintenance management software and the scheduling decision support system, which can provide information about system functionality, user experience and business process efficiency. The logs corresponding to the data interaction indicator data rely on data collection, processing and analysis tools, which ensure the accuracy and timeliness of information flow. The logs corresponding to the basic software and hardware indicator data can be obtained by monitoring the performance of hardware resources and software platforms. These monitoring tools can provide information such as system availability, resource utilization and maintenance efficiency. The logs corresponding to the infrastructure indicator data need to be collected through the management system of the power physical infrastructure and technical support facilities. These systems focus on the construction and operation of substations, transmission lines and data centers. By obtaining log information at all levels, the operation status of the power system can be reflected from all aspects, providing data support for the subsequent diversified evaluation of the power system.

[0102] In a possible implementation, in step S2, the logs are fused by using a preset log fusion technology to obtain a fused log, such as Figure 3 As shown, steps S201-S206 are included:

[0103] Step S201, determining the first log to be merged from the logs;

[0104] Step S202: Calculate the affinity between the first log and each of the other logs using a preset text processing algorithm;

[0105] Step S203: Select a log with the largest affinity from the other logs as the second log to be merged;

[0106] Step S204: determining a log fusion strategy according to the affinity between the first log and the second log;

[0107] Step S205: merging the first log and the second log into a third log according to a log fusion strategy;

[0108] Step S206: re-determine the first log and the second log to be merged from the logs, and merge them until all the logs are merged to obtain the merged log.

[0109] The embodiment of the present application provides a method for log fusion of various types of logs, wherein log fusion refers to merging log information from multiple sources or multiple time periods into a unified log stream for easy analysis and processing. Affinity is an indicator for measuring the similarity or correlation between different logs. In the embodiment of the present application, each fusion selects the first log and the second log from each log for pairwise fusion until all logs are fused into a fusion log. In each fusion process, the first log to be fused is first determined from each log and the affinity between the first log and the other logs is calculated respectively, then the second log to be fused is determined according to each affinity and the log fusion strategy is determined according to the affinity, and finally the first log and the second log are fused according to the log fusion strategy, so as to achieve effective integration of different types of logs, and log information from different systems and different levels can be gathered in one place. By adopting a unified log format and standardized processing flow, the system can more comprehensively capture relevant information about the system operation status.

[0110] Furthermore, in step S202, the using of a preset text processing algorithm to calculate the affinity between the first log and each of the other logs includes:

[0111] Counting the frequency matching between the first log and each of the other logs, where the frequency matching is the frequency of occurrence of a certain keyword at a preset level;

[0112] Counting the time matching degree between the first log and each of the other logs, where the time matching degree is the frequency of API being accessed within a preset time window;

[0113] According to the frequency matching and time matching between the first log and each of the other logs, a preset affinity function is used to calculate the affinity between the first log and each of the other logs.

[0114] The embodiment of the present application provides a method for calculating the affinity of logs, and the similarity between logs is calculated by two indicators: frequency matching and time matching. Regarding frequency matching, in different logs, for example, if two instances ωx and ωy are matched, then the process paths corresponding to the two instances of ωx and ωy (usually referring to the sequence and path of a series of steps, stages or links experienced from the beginning to the end when completing a task or activity) appear the same number of times in the two logs. For one-to-many, many-to-one or many-to-many matching relationships between instances, a single instance pair is examined, and the number of occurrences of the corresponding process paths may be different, but from the global perspective of the individual, their number of occurrences should be similar in statistical terms. Frequency matching is to measure the similarity of the number of occurrences of process paths between all matching relationships in the individual. Regarding time matching, when two processes with business interaction are executed, there is an intersection in time. At this time intersection, one process is reflected in the log to trigger the execution of another process, that is, the duration interval of the two process instances that can be matched has a parallel part. Therefore, the affinity of an individual can be measured by measuring whether there is an overlap in the duration intervals of the process instances between all matching relationships. Furthermore, the embodiment of the present application uses the affinity function to further combine the frequency matching and the time matching to calculate the affinity between logs, thereby improving the rationality and accuracy of the affinity calculation, and thus improving the accuracy of log fusion.

[0115] In a preferred embodiment, when dealing with the problem of API cluster load statistics, the statistical method of frequency matching and time matching between logs is usually as follows:

[0116] 1. Frequency match statistics: Frequency match refers to the frequency of a keyword appearing at a given level, which is usually counted through the following steps:

[0117] 1) Log parsing: First, parse each record in the log file and extract the URL address of each API.

[0118] 2) Hierarchical decomposition: Split each URL address by slash ( / ) to obtain strings at each level.

[0119] 3) Keyword matching: For each URL, check whether the string at the specified level completely matches the query keyword (case sensitive).

[0120] 4) Counting statistics: For each successfully matched URL, a counter is increased. This process will traverse all log records.

[0121] 5) Result output: Finally, output the total frequency of the keyword appearing at the specified level.

[0122] 2. Time matching statistics: Time matching usually refers to the frequency of API access within a specific time window. The statistical method is as follows:

[0123] 1) Time parsing: Extract the timestamp of each request from the log. This usually involves parsing the date and time information in the log.

[0124] 2) Time window definition: Define a time window, such as counting by minutes, hours, or days.

[0125] 3) Time matching: Compare the timestamp of each request with the time window to determine which time window it belongs to.

[0126] 4) Count within window: For each time window, count the number of times the API is accessed within this window.

[0127] 5) Result output: Output the access frequency in each time window, or calculate the average access frequency in the time window.

[0128] 3. Combine frequency and time statistics: In some cases, you may need to consider both frequency and time matching. Here are some possible steps:

[0129] Log preprocessing: First, preprocess the logs to extract the URL and timestamp. Time window segmentation: Split the time into multiple windows according to the required time resolution. Hierarchical and keyword matching: In each time window, perform hierarchical decomposition of the URL and match the keywords. Frequency and time statistics: Count the frequency of keywords in each time window and record them. Result integration: Integrate the statistical results of each time window and output complete frequency and time matching statistics.

[0130] Through these steps, the frequency of API access and the access patterns in different time windows can be effectively counted, thereby providing data support for the load balancing of the server cluster.

[0131] In a preferred embodiment, Figure 4 As shown, the functional expression of the affinity is:

[0132] f=α 1 ∑AOF+α 2 ∑OLT

[0133] Among them, AOF refers to the frequency matching in different logs. For example, if two instances ωx and ωy are matched, then the process paths corresponding to the two instances ωx and ωy (usually refers to the sequence and path of a series of steps, stages or links from the beginning to the end when completing a task or activity) appear the same number of times in the two logs. For a one-to-many, many-to-one or many-to-many matching relationship between instances, the number of occurrences of the corresponding process paths may be different when examining a single instance pair, but from the global perspective of the individual, their number of occurrences should be statistically similar; frequency matching is to measure the similarity of the number of occurrences of process paths between all matching relationships in the individual. OLT refers to the time matching when two processes with business interaction are executed, there is an intersection in time. At this time intersection, one process is reflected in the log to trigger the execution of another process, that is, the duration intervals of the two matching process instances have parallel parts. Therefore, the affinity of the individual can be measured by measuring whether there is an overlapping part in the duration intervals of the process instances between all matching relationships. α1 and α2 are weight factors.

[0134] Furthermore, in step S204, determining a log fusion strategy according to the affinity between the first log and the second log includes:

[0135] When the affinity is greater than or equal to a first threshold and less than 1, determining that the log fusion strategy is to directly merge the first log and the second log;

[0136] When the affinity is greater than or equal to the second threshold and less than the first threshold, determining the log fusion strategy to retain common log entries during the merging process and marking or annotating different parts;

[0137] When the affinity is less than a second threshold, the fusion of the first log and the second log is abandoned, the current first log is marked, and a new first log is selected.

[0138] The embodiment of the present application provides a method for selecting a log fusion strategy, which sets a first threshold and a second threshold in advance. When the affinity is greater than or equal to the first threshold and less than 1, it means that the two logs are very similar, and the first log can be directly merged with the second log; when the affinity is greater than or equal to the second threshold and less than the first threshold, it means that the two logs have a high affinity, but there are also some differences. Therefore, in the merging process, common log entries are retained, and different parts are marked or annotated; when the affinity is less than the second threshold, it means that the two logs are basically different and difficult to merge, so the fusion of the current first log and the second log is abandoned, and the first log is reselected. The embodiment of the present application provides three different fusion strategies based on the affinity of the first log and the second log. For high-affinity logs, direct fusion can be used to improve the efficiency of log fusion, while avoiding fusion between low-affinity logs, improving the accuracy of log fusion.

[0139] In a preferred embodiment, the specific implementation process of step S2 is:

[0140] 1) Calculate affinity:

[0141] First, the affinity between different logs is calculated using an appropriate algorithm (such as affinity function, cosine similarity, Jaccard similarity, edit distance, etc.) The affinity value is usually between 0 and 1, where 1 means they are exactly the same and 0 means they are completely different.

[0142] 2) Determine the fusion strategy: Determine the log fusion strategy based on the affinity value. Common strategies include:

[0143] Complete fusion: When the affinity is close to 1, the two logs can be directly merged. Partial fusion: When the affinity is high but not completely the same, the common log entries can be retained and the different parts can be marked or annotated. No fusion: When the affinity is low, you can choose not to merge or just display them side by side.

[0144] 3) Log preprocessing: Before fusion, the logs are preprocessed, including time standardization, format unification, and removal of redundant information.

[0145] 4) Log fusion steps:

[0146] a) Sorting: Sort the logs according to their affinity values, and prioritize the logs with the highest affinity.

[0147] b) Merge log entries: Merge log entries by following the steps below: For identical log entries, keep only one. For partially identical log entries, merge the common parts and highlight or annotate the different parts. For completely different log entries, keep them separately and add tags in the log to distinguish the source.

[0148] c) Time synchronization: Ensure that the merged logs are continuous and consistent on the timeline.

[0149] d) Information supplement: During the fusion process, if it is found that some log entries are missing important information, it can be supplemented from other logs.

[0150] 5) Log review: After the integration is completed, the merged logs are reviewed to ensure the accuracy and completeness of the information. Check for duplicate, missing or erroneous information and make necessary corrections.

[0151] 6) Storage and indexing: The merged logs are stored in a unified log system and indexed to facilitate subsequent query and analysis.

[0152] 7) Monitoring and optimization: Continuously monitor the effect of log fusion and adjust the fusion strategy according to the actual situation. Regularly update and optimize the log fusion algorithm to adapt to changes in log data. Through the above steps, different logs can be effectively fused according to affinity, improving the quality and efficiency of log analysis.

[0153] In a possible implementation, in step S3, evaluating the economy and reliability of the power grid operation strategy includes:

[0154] Extracting the business indicator data from the fusion log, the business indicator data including the number of power outages, the duration of power outages, the fault recovery time, user feedback information, complaint rate, satisfaction score, power cost, power sales revenue and maintenance cost;

[0155] According to the power outage times, power outage durations and fault recovery time, the average power outage time of the system, the average power outage frequency of the system and the average power outage time of users are calculated to evaluate the power supply reliability of the power system;

[0156] Analyze the number of complaints and their changing trends based on the user feedback information, complaint rate, and satisfaction score, and then evaluate the customer satisfaction of the power system;

[0157] Calculate the return on investment of each project based on the electricity cost, electricity sales income and maintenance costs, and then evaluate the economic benefits of the power system;

[0158] The economy and reliability of the grid operation strategy are evaluated in combination with the power supply reliability, customer satisfaction and economic benefits of the power system.

[0159] In an embodiment of the present application, business indicator data is extracted from the fusion log, and the power supply reliability, customer satisfaction and economic benefits of the power system are evaluated based on the business indicator data. The economy and reliability of the power grid operation strategy are comprehensively evaluated in combination with the above three indicators to achieve an evaluation of the business level of the power system.

[0160] In a preferred embodiment, the evaluation of the economy and reliability of the power grid operation strategy is specifically as follows:

[0161] Collect power supply reliability data from fusion logs, including the number of power outages, power outage duration, fault recovery time, etc. Collect customer satisfaction survey data, including user feedback, complaint rate, satisfaction score, etc. Obtain financial data related to economic operation, such as power cost, power sales revenue, maintenance costs, etc. Clean the collected data to remove duplicate, erroneous or invalid data. Standardize the data to ensure that data from different sources and in different formats can be compared and analyzed under the same framework.

[0162] 1) Power supply reliability assessment: Calculate the average power outage time per user per year to reflect the reliability level of the system power supply. Count the average number of power outages per user per year to assess the frequency of system power outages. Calculate the average duration of each power outage event to reflect the system's response and recovery capabilities to faults.

[0163] 2) Customer satisfaction assessment: Conduct statistical analysis on customer satisfaction survey data to understand users’ overall satisfaction with power supply services and their evaluation of specific aspects. Analyze the number of complaints and their changing trends to identify potential service problems and areas for improvement.

[0164] 3) Economic evaluation: Evaluate the economic feasibility of the operation strategy by comparing financial data such as electricity costs, electricity sales revenue and maintenance costs. Calculate the return on investment and evaluate the economic feasibility of specific investment or improvement projects.

[0165] In a possible implementation, in step S3, the evaluating the application performance of the power grid operation and maintenance management software and the dispatch decision support system includes:

[0166] Extracting the application indicator data from the fusion log, the application indicator data including system operation status, response time, error rate, user experience satisfaction data, user behavior pattern and system abnormality information;

[0167] Formulate a functional test plan based on the system operation status, response time and error rate to verify whether the system meets the preset functional requirements, thereby evaluating the system functionality of the power system;

[0168] According to the user experience satisfaction data, the user satisfaction is quantified using a Likert scale, and the average user satisfaction score and user satisfaction distribution are calculated, thereby evaluating the user experience satisfaction of the power system;

[0169] According to the user behavior pattern and system abnormality information, based on the preset key business process metrics, calculate the business process efficiency value of the power system;

[0170] Combined with the system functionality, user experience satisfaction and business process efficiency value of the power system, the application performance of the power grid operation and maintenance management software and the dispatching decision support system is evaluated.

[0171] In an embodiment of the present application, application indicator data is extracted from the fusion log, and the system functionality, user experience satisfaction, and business process efficiency value of the power system are evaluated based on the application indicator data. The application performance of the power grid operation and maintenance management software and the dispatching decision support system are comprehensively evaluated in combination with the above three indicators to achieve an evaluation of the application level of the power system.

[0172] In a preferred embodiment, the evaluation of the application performance of the power grid operation and maintenance management software and the dispatch decision support system is specifically:

[0173] Use automated monitoring tools (such as Zabbix, Prometheus, etc.) to collect key performance indicators (KPIs) such as system operation status, response time, and error rate in real time. Collect user experience satisfaction data through questionnaires, online scoring systems, or user interviews. Analyze system logs, extract user behavior patterns and system abnormality information, and use them to evaluate system stability and user behavior characteristics. Clean the collected data to remove noise and outliers to ensure data consistency and accuracy.

[0174] By formulating a functional test plan, verify whether the system meets the preset functional requirements, and use indicators such as test case coverage and defect density to quantify functionality. Use the Likert Scale to quantify user satisfaction and calculate indicators such as average score and satisfaction distribution. Define metrics for key business processes, such as task processing time, process completion rate, resource utilization, etc., and evaluate efficiency by comparing actual values ​​with target values.

[0175] In a possible implementation, in step S3, evaluating the data processing efficiency of the power grid includes:

[0176] Extracting the data interaction indicator data from the fusion log;

[0177] Verifying the integrity and accuracy of the data interaction indicator data through preset verification rules, and then evaluating the data processing stability of the power system;

[0178] According to the difference between the timestamps corresponding to the respective data in the data interaction indicator data, the real-time performance of data processing of the power system is evaluated;

[0179] Performing anomaly detection on the data interaction indicator data according to a preset security audit tool, thereby evaluating the data processing security of the power system;

[0180] The data processing efficiency of the power grid is evaluated by combining the data processing stability, data processing real-time performance and data processing security of the power system.

[0181] In an embodiment of the present application, data interaction indicator data is extracted from the fusion log, and the data processing stability, real-time performance and security of the power system are evaluated based on the data interaction indicator data. The data processing efficiency of the power grid is comprehensively evaluated in combination with the above three indicators to achieve an evaluation of the data interaction level of the power system.

[0182] In a preferred embodiment, the data processing efficiency of the power grid is evaluated as follows:

[0183] 1) Data integrity assessment: Verify the integrity and accuracy of data through preset verification rules (such as data range, format verification, etc.). Use hash functions, checksums and other methods to detect whether data has been tampered with or lost during transmission. For missing data, predict and complete it based on historical data, business rules or statistical models to ensure data integrity.

[0184] 2) Real-time evaluation: Record timestamps at each stage of data generation, transmission, and processing, and evaluate the real-time performance of data by the difference in timestamps. Use streaming processing frameworks such as Apache Kafka and Spark Streaming to achieve real-time data collection, processing, and analysis, and improve the real-time performance of data processing.

[0185] 3) Security assessment: Use encryption protocols such as SSL / TLS to encrypt data transmission to ensure the security of data during transmission. At the same time, encrypt and store sensitive data to prevent data leakage. Implement strict access control policies, such as role-based access control (RBAC), to ensure that only authorized users can access sensitive data. Record data access, modification and other operation logs, conduct regular audits and anomaly detection through security audit tools, and promptly discover and deal with potential security threats.

[0186] 4) Comprehensive evaluation method:

[0187] Indicator scoring method: Set scoring criteria and weights for each evaluation indicator, score each indicator based on the collected data, and then sum the weighted sum to obtain a comprehensive evaluation result. Data analysis and mining: Use big data analysis and mining technology to conduct in-depth analysis of massive operation and maintenance data to discover potential problems and patterns, providing strong support for the evaluation. Visual display: Display the evaluation results through visualization methods such as charts and dashboards, so that decision makers can intuitively and quickly understand the performance and existing problems of the data interaction module.

[0188] In a possible implementation, in step S3, the step of evaluating the basic operation and support capability of the power grid security system includes:

[0189] Extracting the basic software and hardware indicator data from the fusion log, wherein the basic software and hardware indicator data includes historical operation data of software and hardware, hardware resource usage data, and maintenance record data;

[0190] Detecting potential faults and performance bottlenecks of the power system based on the historical operation data, and then evaluating the availability of the power system;

[0191] According to the hardware resource usage data, evaluating resource utilization of the power system through data analysis and visual display;

[0192] Analyzing the fault location, fault frequency and maintenance time according to the maintenance record data, and then evaluating the maintenance efficiency of the power system;

[0193] Combined with the availability, resource utilization and maintenance efficiency of the power system, the basic operation and support capabilities of the power grid security system are evaluated.

[0194] In an embodiment of the present application, basic software and hardware indicator data are extracted from the fusion log, and the availability, resource utilization and maintenance efficiency of the power system are evaluated based on the basic software and hardware indicator data. The basic operation and support capabilities of the power grid security system are comprehensively evaluated in combination with the above three indicators, thereby realizing the evaluation of the basic software and hardware layer of the power system.

[0195] In a preferred embodiment, the evaluation of the basic operation and support capabilities of the power grid security system is specifically as follows:

[0196] 1) System availability assessment: Deploy a monitoring system to collect the operating status of software and hardware in real time, and combine log analysis technology to detect and record potential failures or performance bottlenecks, so as to assess the availability of the system. Compare the actual operating data with the pre-set service level agreement (SLA) to assess whether the system meets or exceeds the established availability standards. Perform fault simulation and recovery tests regularly to assess the system's recovery capability and recovery time in the event of sudden failures, and ensure that the system can resume normal operation in the shortest possible time.

[0197] 2) Resource utilization evaluation: Use performance monitoring tools (such as Nagios, Zabbix, etc.) to collect the usage of key resources such as CPU, memory, disk IO, etc., and evaluate resource utilization and bottlenecks through data analysis and visualization. According to resource utilization, use virtualization technology (such as VMware, Hyper-V, etc.) or containerization technology (such as Docker, Kubernetes, etc.) to dynamically allocate and schedule resources to improve resource utilization. By simulating high load conditions in actual business scenarios, evaluate the resource consumption and performance indicators of the system under different loads, and provide data support for resource optimization.

[0198] 3) Maintenance efficiency assessment: Use automated operation and maintenance tools (such as Ansible, Puppet, etc.) to implement configuration management, deployment automation, monitoring and early warning functions, reduce manual intervention, and improve maintenance efficiency. Establish a complete knowledge base and document management system to record solutions to common problems and maintenance experience, so that maintenance personnel can quickly locate and solve problems. By introducing operation and maintenance management frameworks such as ITIL, standardize maintenance processes, clarify the responsibilities and standards of each link, and improve the efficiency and quality of maintenance work.

[0199] In a possible implementation, in step S3, the evaluation of the construction quality, maintenance status and upgrade capability of the power facilities includes:

[0200] Extracting the infrastructure indicator data from the fusion log, the infrastructure indicator data including operation status data of the substation, voltage data of the substation, current data of the substation, power data of the substation, load data of the transmission line, temperature data of the transmission line, and fault status information of the transmission line;

[0201] Analyze the infrastructure indicator data based on big data technology to identify the operation modes and trends of substations, transmission lines and data centers;

[0202] Based on GIS technology, the spatial location information of power grid facilities is combined with the infrastructure indicator data to determine the coverage of the power grid, the distribution of facilities and the potential impact area;

[0203] Based on sensors and IoT technology, obtain real-time status data of power facilities;

[0204] Predicting the remaining life, maintenance requirements and failure probability of each device of the power facility based on the real-time status data and the infrastructure indicator data through a preset prediction model;

[0205] The construction quality, maintenance status and upgrade capabilities of power facilities are evaluated based on the operating modes and trends of the substations, transmission lines and data centers, the coverage of the power grid, the distribution of facilities and potential impact areas, the remaining life, maintenance requirements and failure probability of each equipment in the power facilities.

[0206] In an embodiment of the present application, infrastructure indicator data is extracted from the fusion log, and data analysis is performed on the infrastructure indicator data using big data technology, GIS technology, sensor and Internet of Things technology, and machine learning technology, respectively, to obtain the operating modes and trends of the substations, transmission lines, and data centers, the coverage of the power grid, the distribution of facilities and potential impact areas, the remaining life of each equipment of the power facilities, maintenance requirements, and failure probabilities. The construction quality, maintenance status, and upgrade capabilities of the power facilities are comprehensively evaluated in combination with the above indicators, thereby realizing the evaluation of the infrastructure layer of the power system.

[0207] In a preferred embodiment, the evaluation of the construction quality, maintenance status and upgrade capability of the power facilities is specifically as follows:

[0208] 1) Supervisory Control and Data Acquisition (SCADA): The SCADA system is used to collect data from key nodes in the power grid in real time, including the operating status, voltage, current, power and other parameters of the substation, as well as the load, temperature, fault status and other information of the transmission line. The data is transmitted to the central control room through remote communication for analysis by the evaluation system.

[0209] 2) Big data analysis: Store, process and analyze the massive amount of infrastructure data collected to identify the operating patterns and trends of key facilities such as substations, transmission lines and data centers. Apply data mining and machine learning algorithms to identify potential operating risks, performance bottlenecks and failure modes.

[0210] 3) Geographic Information System (GIS): Combined with GIS technology, the spatial location information of power grid facilities is combined with attribute data (such as the capacity of substations and the specifications of transmission lines) for visualization. GIS can intuitively evaluate the coverage of the power grid, the distribution of facilities and the potential impact areas, and provide decision support for upgrades and maintenance.

[0211] 4) Condition monitoring and prediction: Use sensors and Internet of Things (IoT) technology to monitor the condition of power facilities in real time and collect key indicators such as vibration, temperature, humidity, etc. By establishing a prediction model, the remaining life, maintenance requirements and failure probability of the equipment are predicted to formulate a preventive maintenance plan.

[0212] In a preferred embodiment, the method further includes real-time monitoring of various indicator data, and issuing an alarm when the indicator value of the indicator data exceeds the normal range. For example, for the evaluation of the reliability of power supply at the business layer, after obtaining the relevant value, its operating status can be judged by comparing it with the preset threshold, and an alarm can be issued when the value is abnormal, thereby ensuring the stable operation of the platform.

[0213] For example, a large enterprise has a complex power supply system that needs to ensure 24-hour uninterrupted power supply to support its business operations. In order to ensure power supply reliability, the enterprise introduces a power supply reliability observation system based on the method provided in this application and sets preset thresholds, such as the maximum allowable voltage fluctuation range, the minimum current supply level, etc.

[0214] In this system, the monitoring module will obtain the values ​​of various indicators of the power supply system in real time, such as voltage, current, frequency, etc., and compare these values ​​with the preset thresholds. If the value of an indicator exceeds the preset threshold, such as the voltage fluctuation exceeds the allowable range, the monitoring module will immediately identify this abnormal situation.

[0215] Next, the patrol module will intervene further. The patrol module not only monitors the indicator values ​​in real time, but also inspects each component of the power supply system to check for potential faults or hidden dangers. When the monitoring module finds an abnormality, the patrol module will quickly locate the location of the abnormality and evaluate its impact on the entire power supply system.

[0216] If the patrol module deems the abnormal situation serious and may affect the stable operation of the power supply system, it will immediately trigger the alarm mechanism. The alarm can be sent in a variety of ways, such as SMS, email, system notification, etc., so that relevant personnel can quickly understand the abnormal situation and take necessary measures.

[0217] Through the collaboration of monitoring and patrol modules, enterprises can ensure the stable operation of the power supply system, promptly detect and resolve potential faults or hidden dangers, and thus ensure the normal operation of their business.

[0218] In a preferred embodiment, in step S1, in addition to the system log, the method can also directly obtain various types of indicator data that are manually entered and automatically entered. The manually entered indicator data generally includes, but is not limited to: unstructured data, such as handwritten notes, voice recordings, images and video content; data that requires professional knowledge judgment; and data that requires personal experience or subjective judgment. The automatically entered indicator data generally includes structured data, such as spreadsheets, database records, data submitted by online forms; and paper document data that can be converted into electronic format through scanning, OCR technology, etc.

[0219] The automatically entered index data mainly comes from the data interaction between the power system and external systems, including but not limited to power management systems, operation and maintenance monitoring tools, and various business systems. The power system exchanges data with other systems through relational databases, non-relational databases, API interfaces, or message queues. Such a design not only ensures the accuracy and timeliness of the data, but also improves the flexibility and scalability of the system.

[0220] Embodiment 2:

[0221] like Figure 5 As shown, the second embodiment provides a power system diversified assessment system based on log fusion, including an acquisition module 10, a log fusion module 20 and an assessment module 30;

[0222] The acquisition module 10 is used to acquire logs corresponding to various indicator data in the power system, wherein the indicator data includes business indicator data, application indicator data, data interaction indicator data, basic software and hardware indicator data, and infrastructure indicator data;

[0223] The log fusion module 20 is used to perform log fusion on each of the logs through a preset log fusion technology to obtain a fused log;

[0224] The evaluation module 30 is used to perform diversified evaluation and analysis on the power system according to various indicator data in the fusion log to obtain diversified evaluation results. The diversified evaluation and analysis includes evaluating the economy and reliability of the power grid operation strategy, evaluating the application performance of the power grid operation and maintenance management software and the dispatching decision support system, evaluating the data processing efficiency of the power grid, evaluating the basic operation and support capabilities of the power grid security system, and evaluating the construction quality, maintenance status and upgrade capabilities of power facilities.

[0225] The more detailed working principle and step flow of this embodiment can refer to, but are not limited to, the relevant records of Embodiment 1.

[0226] The specific embodiments described above further describe the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the scope of protection of the present application. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A power system diversified assessment method based on log fusion, characterized in that: include: Obtain logs corresponding to various indicator data in the power system, including business indicator data, application indicator data, data interaction indicator data, basic software and hardware indicator data, and infrastructure indicator data; Performing log fusion on each of the logs using a preset log fusion technology to obtain a fused log; According to various indicator data in the fusion log, the power system is subjected to diversified evaluation and analysis to obtain diversified evaluation results. The diversified evaluation and analysis include evaluating the economy and reliability of the power grid operation strategy, evaluating the application performance of the power grid operation and maintenance management software and the dispatching decision support system, evaluating the data processing efficiency of the power grid, evaluating the basic operation and support capabilities of the power grid security system, and evaluating the construction quality, maintenance status and upgrade capabilities of power facilities.

2. A method for diversified evaluation of power systems based on log fusion as claimed in claim 1, characterized in that: The log fusion is performed on each of the logs by using a preset log fusion technology to obtain a fused log, including: Determine a first log to be currently merged from each of the logs; Calculate the affinity between the first log and each of the other logs using a preset text processing algorithm; Selecting a log with the largest affinity from the other logs as the second log to be merged; Determining a log fusion strategy according to the affinity between the first log and the second log; Merging the first log and the second log into a third log according to a log fusion strategy; The first log and the second log to be merged are determined again from each of the logs for merging until all the logs are merged to obtain the merged log.

3. A method for diversified evaluation of power systems based on log fusion as claimed in claim 2, characterized in that: The using a preset text processing algorithm to calculate the affinity between the first log and each of the other logs includes: Counting the frequency matching between the first log and each of the other logs, where the frequency matching is the frequency of occurrence of a certain keyword at a preset level; Counting the time matching degree between the first log and each of the other logs, where the time matching degree is the frequency of API being accessed within a preset time window; According to the frequency matching and time matching between the first log and each of the other logs, a preset affinity function is used to calculate the affinity between the first log and each of the other logs.

4. A method for diversified evaluation of power systems based on log fusion as claimed in claim 2, characterized in that: The determining the log fusion strategy according to the affinity between the first log and the second log includes: When the affinity is greater than or equal to a first threshold and less than 1, determining that the log fusion strategy is to directly merge the first log and the second log; When the affinity is greater than or equal to the second threshold and less than the first threshold, determining the log fusion strategy to retain common log entries during the merging process and marking or annotating different parts; When the affinity is less than a second threshold, the fusion of the first log and the second log is abandoned, the current first log is marked, and a new first log is selected.

5. The method for diversified evaluation of power system based on log fusion according to claim 1, characterized in that: The economic and reliability evaluation of the grid operation strategy includes: Extracting the business indicator data from the fusion log, the business indicator data including the number of power outages, the duration of power outages, the fault recovery time, user feedback information, complaint rate, satisfaction score, power cost, power sales revenue and maintenance cost; According to the power outage times, power outage durations and fault recovery time, the average power outage time of the system, the average power outage frequency of the system and the average power outage time of users are calculated to evaluate the power supply reliability of the power system; Analyze the number of complaints and their changing trends based on the user feedback information, complaint rate, and satisfaction score, and then evaluate the customer satisfaction of the power system; Calculate the return on investment of each project based on the electricity cost, electricity sales income and maintenance costs, and then evaluate the economic benefits of the power system; The economy and reliability of the grid operation strategy are evaluated in combination with the power supply reliability, customer satisfaction and economic benefits of the power system.

6. A method for diversified evaluation of power systems based on log fusion as claimed in claim 1, characterized in that: The evaluation of the application performance of the power grid operation and maintenance management software and the dispatch decision support system includes: Extracting the application indicator data from the fusion log, the application indicator data including system operation status, response time, error rate, user experience satisfaction data, user behavior pattern and system abnormality information; Formulate a functional test plan based on the system operation status, response time and error rate to verify whether the system meets the preset functional requirements, thereby evaluating the system functionality of the power system; Based on the user experience satisfaction data, the user satisfaction is quantified using a Likert scale, and the average user satisfaction score and user satisfaction distribution are calculated, thereby evaluating the user experience satisfaction of the power system; According to the user behavior pattern and system abnormality information, based on the preset key business process metrics, calculate the business process efficiency value of the power system; Combined with the system functionality, user experience satisfaction and business process efficiency value of the power system, the application performance of the power grid operation and maintenance management software and the dispatching decision support system is evaluated.

7. A method for diversified evaluation of power systems based on log fusion as claimed in claim 1, characterized in that: The method of evaluating the data processing efficiency of the power grid comprises: Extracting the data interaction indicator data from the fusion log; Verifying the integrity and accuracy of the data interaction indicator data through preset verification rules, and then evaluating the data processing stability of the power system; According to the difference between the timestamps corresponding to the respective data in the data interaction indicator data, the real-time performance of data processing of the power system is evaluated; Performing anomaly detection on the data interaction indicator data according to a preset security audit tool, thereby evaluating the data processing security of the power system; The data processing efficiency of the power grid is evaluated by combining the data processing stability, data processing real-time performance and data processing security of the power system.

8. The method for diversified evaluation of power system based on log fusion according to claim 1, characterized in that: The evaluation of the basic operation and support capabilities of the power grid security system includes: Extracting the basic software and hardware indicator data from the fusion log, wherein the basic software and hardware indicator data includes historical operation data of software and hardware, hardware resource usage data, and maintenance record data; Detecting potential faults and performance bottlenecks of the power system based on the historical operation data, and then evaluating the availability of the power system; According to the hardware resource usage data, evaluating resource utilization of the power system through data analysis and visual display; Analyzing the fault location, fault frequency and maintenance time according to the maintenance record data, and then evaluating the maintenance efficiency of the power system; Combined with the availability, resource utilization and maintenance efficiency of the power system, the basic operation and support capabilities of the power grid security system are evaluated.

9. The method for diversified evaluation of power system based on log fusion according to claim 1, characterized in that: The assessment of the construction quality, maintenance status and upgrade capabilities of power facilities includes: Extracting the infrastructure indicator data from the fusion log, the infrastructure indicator data including operation status data of the substation, voltage data of the substation, current data of the substation, power data of the substation, load data of the transmission line, temperature data of the transmission line, and fault status information of the transmission line; Analyze the infrastructure indicator data based on big data technology to identify the operation modes and trends of substations, transmission lines and data centers; Based on GIS technology, the spatial location information of power grid facilities is combined with the infrastructure indicator data to determine the coverage of the power grid, the distribution of facilities and the potential impact area; Based on sensors and IoT technology, obtain real-time status data of power facilities; Predicting the remaining life, maintenance requirements and failure probability of each device of the power facility based on the real-time status data and the infrastructure indicator data through a preset prediction model; The construction quality, maintenance status and upgrade capabilities of power facilities are evaluated based on the operating modes and trends of the substations, transmission lines and data centers, the coverage of the power grid, the distribution of facilities and potential impact areas, the remaining life, maintenance requirements and failure probability of each equipment in the power facilities.

10. A power system diversified assessment system based on log fusion, characterized in that: Includes acquisition module, log fusion module and evaluation module; The acquisition module is used to acquire logs corresponding to various indicator data in the power system, wherein the indicator data includes business indicator data, application indicator data, data interaction indicator data, basic software and hardware indicator data, and infrastructure indicator data; The log fusion module is used to perform log fusion on each of the logs through a preset log fusion technology to obtain a fused log; The evaluation module is used to perform diversified evaluation and analysis on the power system according to various indicator data in the fusion log to obtain diversified evaluation results. The diversified evaluation and analysis includes evaluating the economy and reliability of the power grid operation strategy, evaluating the application performance of the power grid operation and maintenance management software and the dispatching decision support system, evaluating the data processing efficiency of the power grid, evaluating the basic operation and support capabilities of the power grid security system, and evaluating the construction quality, maintenance status and upgrade capabilities of power facilities.

Images