Active caching method and system applied to distribution network mutual inductor metering misalignment online identification system and medium

Abstract

This application relates to an active caching method, system, and medium for an online identification system of metering inaccuracies in distribution network instrument transformers. The method includes acquiring historical user data from the online identification system of metering inaccuracies in distribution network instrument transformers; extracting first user historical behavior features, first user historical behavior change trend features, and first user historical social features from the historical user data; performing cross-feature fusion on the user historical behavior features and user historical behavior change trend features to obtain first mixed features, second mixed features, third mixed features, and fourth mixed features; fusing the prediction results of the first user historical social features, first mixed features, second mixed features, third mixed features, and fourth mixed features using LSTM to obtain the final user preference prediction result; and sorting the user preference prediction results to achieve active data caching. This application improves the accuracy of user preference prediction.

Description

Technical Field

[0001] This invention relates to the field of smart grids, and more specifically, to an active caching method, system, and medium for use in an online identification system for metering inaccuracies of distribution network transformers. Background Technology

[0002] According to the JJG1189 "Verification Procedure for Measuring Instrument Transformers", the verification cycle for electromagnetic current and voltage transformers shall not exceed 10 years. Verification equipment is required to conduct full-coverage verification of the transformers. Traditional periodic inspection methods suffer from drawbacks such as a large workload and poor timeliness. To improve the automation and informatization level of on-site verification in power systems and reduce the workload of on-site personnel, it is necessary to build an online identification system for metering inaccuracies of distribution network transformers based on the electricity consumption information collection system, thereby realizing condition monitoring, assessment, and inaccuracy identification of distribution network transformers.

[0003] With the development of computer hardware, high-speed network transmission technology, and big data technology, users have placed higher demands on the performance of online identification systems for metering inaccuracies of distribution network transformers based on electricity information collection systems. These demands include shorter response times, higher concurrency, and faster page rendering frame rates. Adding a caching module to the system architecture and employing efficient caching algorithms is one way to achieve shorter response times and higher concurrency. Currently, most caching mechanisms are passive, triggering caching only after a user request reaches the server. This approach still introduces a certain delay when users retrieve content. When the cache is full, algorithms such as LRU (Least Recently Used), LFU (Least Frequently Used), and FIFO (First-In, First-Out) are used to determine which data will be deleted. However, the inability to predict future resource access can lead to the eviction of resources that truly need to be cached, resulting in a low cache hit rate. Summary of the Invention

[0004] To address the aforementioned deficiencies or improvement needs of existing technologies, this paper provides an active caching method, system, and medium for online identification of metering inaccuracies in distribution network instrument transformers. This method extracts and fuses user behavior characteristics from multiple dimensions, thereby improving the accuracy of user preference prediction.

[0005] To achieve the above objectives, this application provides the following technical solution:

[0006] In a first aspect, embodiments of this application provide an active caching method for an online identification system for metering inaccuracies of distribution network instrument transformers, comprising the following steps:

[0007] Obtain historical user data from the online identification system for metering inaccuracies of instrument transformers used in distribution networks, and extract the historical behavioral characteristics of the first user from the historical user data. First, the characteristics of changes in the historical behavior of users. First user's historical social characteristics ;

[0008] User historical behavior characteristics User historical behavior change trend characteristics Cross-feature fusion is performed to obtain the first hybrid feature. Second hybrid characteristics Third Mixed Characteristics Fourth Mixed Characteristics ;

[0009] Using LSTM to analyze the first user's historical social features First Mixed Feature Second hybrid characteristics Third Mixed Characteristics Fourth Mixed Characteristics The prediction results are fused together to obtain the sixth prediction result, which is the final user preference prediction result.

[0010] Based on the obtained user preference prediction results, the user preference prediction results are sorted to achieve proactive data caching.

[0011] User historical behavior characteristics include the number of times a user browses and the click-through rate;

[0012] User historical behavior change trend characteristics include user historical behavior time change trend characteristics and user historical behavior geographic location change trend characteristics:

[0013] The characteristics of user historical behavior over time include the trends of user behavior changes on a weekly, monthly, and quarterly basis.

[0014] The characteristics of changes in user historical behavior based on geographic location include the trends in user behavior across different geographic locations;

[0015] User history social characteristics include a user's activity on social media and interactions with other users.

[0016] User historical behavior characteristics User historical behavior change trend characteristics Cross-feature fusion is performed to obtain the first hybrid feature. Second hybrid characteristics Third Mixed Characteristics Fourth Mixed Characteristics The specific process is as follows:

[0017] First user's historical behavior characteristics First user's historical behavior characteristics Cross-fusion is performed to obtain the first hybrid feature. The specific process is as follows:

[0018] (1)

[0019] The third feature extraction module, Conv3, is used to extract the historical behavior features of the first user. Extraction was performed to obtain the second user's historical behavior features. The seventh feature extraction module, Conv7, was used to analyze the historical behavior trend features of the first user. Extraction was performed to obtain the historical behavior trend features of the second user. ; to include the second user's historical behavior characteristics Second, the characteristics of changes in users' historical behavior. Cross-fusion is performed to obtain the second hybrid feature. The specific process is as follows:

[0020] (2)

[0021] (3)

[0022] (4)

[0023] The fourth feature extraction module, Conv4, is used to extract the historical behavior features of the second user. Extraction was performed to obtain the historical behavioral characteristics of the third user. The eighth feature extraction module, Conv8, was used to analyze the historical behavior trend of the second user. Extraction was performed to obtain the historical behavior trend features of the third user. ; The historical behavioral characteristics of third-party users Third-party user historical behavior change trend characteristics Cross-fusion is performed to obtain a third hybrid feature. The specific process is as follows:

[0024] (5)

[0025] (6)

[0026] (7)

[0027] The fifth feature extraction module, Conv5, is used to extract the historical behavior features of the third user. Extraction was performed to obtain the fourth user's historical behavior features. The ninth feature extraction module, Conv9, was used to analyze the historical behavior trends of the third user. Extraction was performed to obtain the fourth user's historical behavior change trend features. ; to include the fourth user's historical behavior characteristics Fourth, the characteristics of changes in user historical behavior. Cross-fusion is performed to obtain the fourth hybrid feature. The specific process is as follows:

[0028] (8)

[0029] (9)

[0030] (10).

[0031] Using LSTM to analyze the first user's historical social features First Mixed Feature Second hybrid characteristics Third Mixed Characteristics Fourth Mixed Characteristics The prediction results are fused to obtain the sixth prediction result, which is the final user preference prediction result. The specific process is as follows:

[0032] (11)

[0033] (12)

[0034] (13)

[0035] (14)

[0036] (15)

[0037] (16)

[0038] in, Indicates data concatenation. , , , , , This represents the prediction results for user preferences of the first, second, third, fourth, fifth, and sixth users. , , , , , Represents the implicit functions of the first, second, third, fourth, fifth, and sixth LSTM models.

[0039] Specifically, user preference P is the user's preference. Select resources Joint probability:

[0040] (17)

[0041] (18)

[0042] (19)

[0043] in, For users, f represents resources. , T is the set of resources. Indicates users within a historical time period Access resources The number of times.

[0044] Secondly, embodiments of this application provide an active caching system for an online identification system of metering inaccuracies in distribution network instrument transformers, including a memory and a processor. The memory includes a program for an active caching method for an online identification system of metering inaccuracies in distribution network instrument transformers. When the program for an active caching method for an online identification system of metering inaccuracies in distribution network instrument transformers is executed by the processor, it implements the steps of the active caching method for an online identification system of metering inaccuracies in distribution network instrument transformers as described above.

[0045] Thirdly, embodiments of this application provide a computer-readable storage medium storing program code, which, when executed by a processor, implements the steps of an active caching method for an online identification system for metering inaccuracies of distribution network transformers as described above.

[0046] Compared with the prior art, the present invention has significant beneficial effects.

[0047] (1) Multi-dimensional extraction and fusion of user behavior features improves the accuracy of user preference prediction, effectively reduces prediction error, and increases recommendation hit rate.

[0048] (2) Based on supervised learning, the active caching mechanism can dynamically adjust the cache content and strategy according to user preferences, which can greatly improve the cache hit rate, reduce redundant cache occupation, and reduce system resource consumption.

[0049] (3) The fusion of multi-dimensional attribute features can break through the limitations of a single feature, enhance the model's ability to represent complex user behaviors, and improve the generalization and robustness of the algorithm in different scenarios.

[0050] (4) Through accurate user preference prediction and proactive caching optimization, the response time of user requests can be shortened, and the smoothness of business services and user experience can be improved.

[0051] (5) This technical solution can improve service quality while reducing the load and redundant calculations on the backend server, thereby reducing the overall operating cost of the system. Attached Figure Description

[0052] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 A flowchart provided for an embodiment of the present invention;

[0054] Figure 2 This is a comparison chart of hit rates in embodiments of the present invention;

[0055] Figure 3 This is a comparison chart of response times in embodiments of the present invention;

[0056] Figure 4 This is a comparison chart of resource utilization rates in an embodiment of the present invention. Detailed Implementation

[0057] The technical solutions of the embodiments of this application will now be described with reference to the accompanying drawings. It should be noted that similar reference numerals and letters in the following drawings indicate similar items; therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0058] The terms “comprising,” “including,” or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase “comprising one…” does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0059] The terms “first,” “second,” etc., are used only to distinguish one entity or operation from another, and should not be construed as indicating or implying relative importance, nor as requiring or implying any such actual relationship or order between these entities or operations.

[0060] like Figure 1As shown in the embodiment of the present invention, the present invention provides an active caching method for an online identification system for metering inaccuracies of power distribution transformers, comprising the following steps:

[0061] Obtain historical user data from the online identification system for metering inaccuracies of instrument transformers used in distribution networks, and extract the historical behavioral characteristics of the first user from the historical user data. First, the characteristics of changes in the historical behavior of users. First user's historical social characteristics ;

[0062] User historical behavior characteristics User historical behavior change trend characteristics Cross-feature fusion is performed to obtain the first hybrid feature. Second hybrid characteristics Third Mixed Characteristics Fourth Mixed Characteristics ;

[0063] Using LSTM to analyze the first user's historical social features First Mixed Feature Second hybrid characteristics Third Mixed Characteristics Fourth Mixed Characteristics The prediction results are fused together to obtain the sixth prediction result, which is the final user preference prediction result.

[0064] Based on the obtained user preference prediction results, the user preference prediction results are sorted to achieve proactive data caching.

[0065] To verify the superiority of the present invention, the following experiments were conducted using traditional random caching strategies and LRU (Least Recently Used) caching strategies as control groups:

[0066] Table 1 Core Evaluation Indicators

[0067] index Calculation method Cache hit rate Cache hits / Total requests × 100% Average request response time Sum of all request response times / Total number of requests (in milliseconds) System resource utilization Cache usage memory / total allocated memory × 100% (reflects the ability to control redundant cache)

[0068] from Figure 2 , 3 As can be seen from section 4, the experimental conclusions are analyzed as follows:

[0069] (1) Cache hit rate: The experimental group improved by 16.1% compared with LRU and by 109.9% compared with random cache, proving that multi-dimensional feature fusion can accurately capture user preferences, and supervised models can effectively predict cache hit probability, greatly improving the hit effect;

[0070] (2) Request response time: The experimental group shortened the response time by 64.5% compared to LRU and by 86.1% compared to random cache. The high hit rate reduced the number of queries to the backend server, which directly reduced the response latency.

[0071] (3) Resource utilization rate: The experimental group reduced the resource utilization rate by 21.0% compared with LRU and by 33.4% compared with random cache, indicating that the active caching strategy only caches content with high hit probability, effectively reducing redundant caching and reducing system resource consumption;

[0072] In summary, the "multi-dimensional feature fusion + supervised learning" active caching strategy in this invention patent is significantly better than traditional caching strategies in terms of hit rate, response speed, and resource utilization, verifying the effectiveness of the technical solution of this invention.

[0073] This application provides an active caching system for an online identification system of metering inaccuracies in distribution network instrument transformers, including a memory and a processor. The memory includes a program for an active caching method for an online identification system of metering inaccuracies in distribution network instrument transformers. When the program for an active caching method for an online identification system of metering inaccuracies in distribution network instrument transformers is executed by the processor, it implements the steps of the active caching method for an online identification system of metering inaccuracies in distribution network instrument transformers as described above.

[0074] This application provides a computer-readable storage medium storing program code. When the program code is executed by a processor, it implements the steps of an active caching method for an online identification system for metering inaccuracies of distribution network transformers as described above.

[0075] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0076] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0077] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0078] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0079] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0080] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0081] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0082] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. An active caching method for an online identification system of metering inaccuracies in distribution network instrument transformers, characterized in that, Includes the following steps: Obtain historical user data from the online identification system for metering inaccuracies of instrument transformers used in distribution networks, and extract the historical behavioral characteristics of the first user from the historical user data. First, the characteristics of changes in the historical behavior of users. First user's historical social characteristics ; User historical behavior characteristics User historical behavior change trend characteristics Cross-feature fusion is performed to obtain the first hybrid feature. Second hybrid characteristics Third Mixed Characteristics Fourth Mixed Characteristics ; Using LSTM to analyze the first user's historical social features First Mixed Feature Second hybrid characteristics Third Mixed Characteristics Fourth Mixed Characteristics The prediction results are fused together to obtain the sixth prediction result, which is the final user preference prediction result. Based on the obtained user preference prediction results, the user preference prediction results are sorted to achieve proactive data caching.

2. The active caching method for an online identification system for metering inaccuracies of distribution network instrument transformers according to claim 1, characterized in that, User historical behavior characteristics include the number of times a user browses and the click-through rate; User historical behavior change trend characteristics include user historical behavior time change trend characteristics and user historical behavior geographic location change trend characteristics: The characteristics of user historical behavior over time include the trends of user behavior changes on a weekly, monthly, and quarterly basis. The characteristics of changes in user historical behavior based on geographic location include the trends in user behavior across different geographic locations; User history social characteristics include a user's activity on social media and interactions with other users.

3. The active buffering method for an online identification system for metering inaccuracies of distribution network instrument transformers according to claim 1, characterized in that, User historical behavior characteristics User historical behavior change trend characteristics Cross-feature fusion is performed to obtain the first hybrid feature. Second hybrid characteristics Third Mixed Characteristics Fourth Mixed Characteristics The specific process is as follows: First user's historical behavior characteristics First user's historical behavior characteristics Cross-fusion is performed to obtain the first hybrid feature. The specific process is as follows: （1） The third feature extraction module, Conv3, is used to extract the historical behavior features of the first user. Extraction was performed to obtain the second user's historical behavior features. The seventh feature extraction module, Conv7, was used to analyze the historical behavior trend features of the first user. Extraction was performed to obtain the historical behavior trend features of the second user. ; Second user historical behavior characteristics Second, the characteristics of changes in users' historical behavior. Cross-fusion is performed to obtain the second hybrid feature. The specific process is as follows: （2） （3） （4） The fourth feature extraction module, Conv4, is used to extract the historical behavior features of the second user. Extraction was performed to obtain the historical behavioral characteristics of the third user. The eighth feature extraction module, Conv8, was used to analyze the historical behavior trend of the second user. Extraction was performed to obtain the historical behavior trend features of the third user. ; Third-party user historical behavior characteristics Third-party user historical behavior change trend characteristics Cross-fusion is performed to obtain a third hybrid feature. The specific process is as follows: （5） （6） （7） The fifth feature extraction module, Conv5, is used to extract the historical behavior features of the third user. Extraction was performed to obtain the fourth user's historical behavior features. The ninth feature extraction module, Conv9, was used to analyze the historical behavior trends of the third user. Extraction was performed to obtain the fourth user's historical behavior change trend features. ; Fourth user historical behavior characteristics Fourth, the characteristics of changes in user historical behavior. Cross-fusion is performed to obtain the fourth hybrid feature. The specific process is as follows: （8） （9） （10）。 4. The active buffering method for an online identification system for metering inaccuracies of distribution network instrument transformers according to claim 1, characterized in that, Using LSTM to analyze the first user's historical social features First Mixed Feature Second hybrid characteristics Third Mixed Characteristics Fourth Mixed Characteristics The prediction results are fused to obtain the sixth prediction result, which is the final user preference prediction result. The specific process is as follows: （11） （12） （13） （14） （15） （16） in, Indicates data concatenation. , , , , , This represents the prediction results for user preferences of the first, second, third, fourth, fifth, and sixth users. , , , , , Represents the implicit functions of the first, second, third, fourth, fifth, and sixth LSTM models. Specifically, user preference P is the user's preference. Select resources Joint probability: (17) (18) (19) in, For users, f represents resources. , T is the set of resources. Indicates users within a historical time period Access resources The number of times.

5. An active buffering system for online identification of metering inaccuracies in distribution network instrument transformers, characterized in that, The device includes a memory and a processor. The memory includes a program for an active caching method applied to an online identification system for metering inaccuracies of distribution network instrument transformers. When the program for the active caching method applied to an online identification system for metering inaccuracies of distribution network instrument transformers is executed by the processor, it implements the steps of the active caching method for an online identification system for metering inaccuracies of distribution network instrument transformers as described in any one of claims 1 to 4.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program code, which, when executed by a processor, implements the steps of an active caching method for an online identification system for metering inaccuracies of power distribution transformers as described in any one of claims 1 to 4.

Images