Power distributed service communication analysis technical index selection method and related device

Abstract

The invention belongs to the field of power systems, and discloses a power distributed service communication analysis technical index selection method and related devices.The method comprises the steps that firstly, the initial importance weight of each region is determined based on the technical index value of each terminal in a historical collection period; performing global collaboration on the initial weight through the central processing master station and issuing the initial weight; performing iterative updating and optimization on the weight by each regional management system according to local data; and finally, aggregating the optimal weight of each region by the central processing main station, determining a global optimal importance weight, and realizing accurate selection of communication analysis technical indexes according to the global optimal importance weight. According to the method, the traditional static weight is replaced by the data-driven dynamic weight, the distributed characteristics of the system and the real-time change of the service are fully considered, the accuracy and the adaptability of the technical index weight calculation are remarkably improved, and effective support is provided for resource optimization and performance improvement of the communication network.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of power systems, and relates to a power distributed business communication analysis technical index selection method and related devices. BACKGROUND

[0002] With the intelligent upgrading of new power systems, power businesses have put forward more refined requirements for technical indexes such as time delay, bandwidth and reliability of communication networks. This has made quantitative analysis of communication analysis indexes become the core basis for optimizing network resource allocation and guaranteeing service quality of businesses. Therefore, how to select appropriate technical indexes for power business communication analysis is the key support work for meeting the power business communication analysis.

[0003] At present, the selection of power business communication analysis technical indexes mainly adopts a centralized computing framework and a static weight calculation method. However, the centralized computing framework does not fully consider the distributed deployment characteristics of part of power business management systems, which is not conducive to actual application deployment. At the same time, the static weight calculation method is not suitable for the space-time change characteristics of data distribution in the power business scene, which leads to insufficient calculation accuracy of the importance in quantization, affects the performance of the selection of power distributed business communication analysis technical indexes, and further affects the communication analysis results, and finally affects the optimization of resource allocation of communication networks and the improvement of service quality of businesses. SUMMARY

[0004] The purpose of the application is to overcome the shortcomings of the prior art and provide a power distributed business communication analysis technical index selection method and related devices.

[0005] To achieve the above purpose, the application adopts the following technical solutions: The application provides a power distributed service communication analysis technical index selection method, which is applied to a communication analysis technical index selection system.

[0006] Optionally, the initial regional importance weight of each technical index is determined according to the regional normalized values of each technical index in each collection period, and the initial regional importance weight of each technical index is sent to the central processing master station.

[0007] Optionally, the initial central importance weight of each technical index is obtained according to the initial regional importance weight of each technical index of each regional management system, and the initial central importance weight of each technical index is sent to each regional management system.

[0008] Optionally, the method for obtaining the optimal regional importance weight of each technical index according to the initial central importance weight of each technical index and the regional normalized value of each technical index of each collection cycle comprises: iteratively performing an updating step until a preset iteration termination condition is reached, and taking the current regional importance weight of each technical index as the optimal regional importance weight of each technical index; wherein the updating step comprises: randomly selecting the regional normalized value of each technical index of one collection cycle as updating basis data; calculating an estimated value of each technical index according to the initial central importance weight of each technical index by using a preset inverse function; wherein the preset inverse function is an inverse function of the calculation function of the initial regional importance weight of each technical index; and obtaining the change rate of the regional importance weight based on the updating basis data and the estimated value by the following formula: =

[0009] wherein, is the change rate of the i th iteration, is the updating basis data, is the estimated value of each technical index, is the regional importance weight of the i th iteration, is the current regional importance weight of each technical index. The current regional importance weight of each technical index is updated by the following formula:

[0010] = -

[0011] wherein, is the regional importance weight of the i th iteration, is the preset iteration learning rate.

[0012] ​​In a second aspect, the application provides a method for selecting technical indicators for power distributed service communication analysis, which is applied to a regional management system of a communication analysis technical indicator selection system, the communication analysis technical indicator selection system comprising a central processing master station and a plurality of regional management systems in communication connection with the central processing master station; the method comprising: obtaining and averaging and normalizing technical indicator values of each technical indicator of each terminal in a plurality of previous collection periods to obtain regional normalized values of each technical indicator of each collection period; determining initial regional importance weights of each technical indicator according to the regional normalized values of each technical indicator of each collection period and sending the initial regional importance weights to the central processing master station; wherein the initial regional importance weights of each technical indicator are used to trigger the central processing master station to obtain initial central importance weights of each technical indicator according to the initial regional importance weights of each technical indicator of each regional management system and send the initial central importance weights to each regional management system; obtaining optimal regional importance weights of each technical indicator by using a method based on stochastic gradient descent according to the initial central importance weights of each technical indicator and in combination with the regional normalized values of each technical indicator of each collection period and sending the optimal regional importance weights to the central processing master station; wherein the optimal regional importance weights of each technical indicator are used to trigger the central processing master station to obtain optimal central importance weights of each technical indicator according to the optimal regional importance weights of each technical indicator of each regional management system and select a communication analysis technical indicator from each technical indicator according to the optimal central importance weights of each technical indicator.

[0013] Optionally, the determining of the initial regional importance weights of each technical indicator according to the regional normalized values of each technical indicator of each collection period comprises: randomly selecting the regional normalized values of each technical indicator of one collection period as basic data; and determining the initial regional importance weights of each technical indicator by using a static weight calculation method based on a minimum deviation criterion of technical indicator adaptation degree based on the basic data.

[0014] Optionally, the obtaining of the optimal regional importance weights of each technical indicator by using the method based on stochastic gradient descent according to the initial central importance weights of each technical indicator and in combination with the regional normalized values of each technical indicator of each collection period comprises: iteratively performing an updating step until a preset iteration termination condition is reached, and taking the current regional importance weights of each technical indicator as the optimal regional importance weights of each technical indicator; wherein the updating step comprises: randomly selecting the regional normalized values of each technical indicator of one collection period as updating basis data; and calculating estimated values of each technical indicator by using a preset inverse function based on the initial central importance weights of each technical indicator; wherein the preset inverse function is an inverse function of a calculation function of the initial regional importance weights of each technical indicator; and obtaining the regional importance weights based on the change rates of the updating basis data and the estimated values by the following formula: =

[0015] in, For the first Rate of change in each iteration To update the basis data, These are estimated values ​​for each technical indicator. For the first The regional importance weights in the next iteration.

[0016] The current regional importance weights of each technical indicator are updated using the following formula: = -

[0017] in, For the first The regional importance weights in the next iteration The preset iterative learning rate.

[0018] In a third aspect, this invention provides a method for selecting technical indicators for communication analysis in distributed power services, applied to a central processing master station of a communication analysis technical indicator selection system. The communication analysis technical indicator selection system includes a central processing master station and several regional management systems communicatively connected to the central processing master station. The method includes: obtaining initial central importance weights for each technical indicator based on the initial regional importance weights of each technical indicator in each regional management system, and sending these weights to each regional management system; wherein the initial regional importance weights of each technical indicator in each regional management system are obtained by: the regional management system acquiring the technical indicator values ​​of each technical indicator from each terminal within the previous several acquisition cycles and averaging and normalizing them to obtain the values ​​for each acquisition cycle. The system calculates the regional normalized values ​​of each technical indicator for each collection period; and determines the initial regional importance weight of each technical indicator based on the regional normalized values ​​of each technical indicator for each collection period, and sends it to the central processing station. The initial central importance weight of each technical indicator is used to trigger the regional management system to obtain the optimal regional importance weight of each technical indicator based on the initial central importance weight of each technical indicator and the regional normalized values ​​of each technical indicator for each collection period, using a stochastic gradient descent method, and sends it to the central processing station. Based on the optimal regional importance weight of each technical indicator in each regional management system, the optimal central importance weight of each technical indicator is obtained, and communication analysis technical indicators are selected from each technical indicator based on the optimal central importance weight of each technical indicator.

[0019] Optionally, obtaining the initial central importance weight of each technical indicator based on the initial regional importance weight of each technical indicator in each regional management system includes: averaging the initial regional importance weights of each technical indicator in each regional management system to obtain the initial central importance weight of each technical indicator; obtaining the optimal central importance weight of each technical indicator based on the optimal regional importance weight of each technical indicator in each regional management system includes: averaging the optimal regional importance weights of each technical indicator in each regional management system to obtain the optimal central importance weight of each technical indicator.

[0020] In a fourth aspect, this invention provides a power distributed service communication analysis technical indicator selection device, applied to a regional management system of a communication analysis technical indicator selection system. The communication analysis technical indicator selection system includes a central processing master station and several regional management systems communicatively connected to the central processing master station. The communication analysis technical indicator selection device includes: a data processing module, used to acquire the technical indicator values ​​of each technical indicator from each terminal within several previous acquisition cycles and perform averaging and normalization to obtain the regional normalized value of each technical indicator for each acquisition cycle; and a local weight initialization module, used to determine the initial regional importance weight of each technical indicator based on the regional normalized value of each technical indicator for each acquisition cycle, and send it to the central processing master station; wherein the initial regional importance weight of each technical indicator is used to trigger... The central processing station obtains the initial central importance weight of each technical indicator based on the initial regional importance weight of each technical indicator in each regional management system, and sends it to each regional management system. The local weight optimization module is used to obtain the optimal regional importance weight of each technical indicator based on the initial central importance weight of each technical indicator and the regional normalized value of each technical indicator in each collection period, using a stochastic gradient descent method, and sends it to the central processing station. The optimal regional importance weight of each technical indicator is used to trigger the central processing station to obtain the optimal central importance weight of each technical indicator based on the optimal regional importance weight of each technical indicator in each regional management system, and to select communication analysis technical indicators from the technical indicators based on the optimal central importance weight of each technical indicator.

[0021] Optionally, determining the initial regional importance weight of each technical indicator based on the regional normalized value of each technical indicator in each collection period includes: randomly selecting the regional normalized value of each technical indicator in one collection period as the basic data; and determining the initial regional importance weight of each technical indicator based on the basic data and the principle of minimizing the adaptation deviation of the technical indicator using a static weight calculation method.

[0022] Optionally, the step of obtaining the optimal regional importance weight of each technical indicator based on its initial central importance weight and the regional normalized value of each technical indicator in each collection period, using a stochastic gradient descent method, includes: iteratively performing update steps until a preset iteration termination condition is reached, and using the current regional importance weight of each technical indicator as its optimal regional importance weight; wherein, the update step includes: randomly selecting the regional normalized value of each technical indicator in one collection period as the update basis data; calculating the estimated value of each technical indicator using a preset inverse function based on its initial central importance weight; wherein, the preset inverse function is the inverse function of the calculation function of the initial regional importance weight of each technical indicator; and obtaining the rate of change of the regional importance weight based on the update basis data and the estimated value using the following formula: =

[0023] in, For the first Rate of change in each iteration To update the basis data, These are estimated values ​​for each technical indicator. For the first The regional importance weights in the next iteration.

[0024] The current regional importance weights of each technical indicator are updated using the following formula: = -

[0025] in, For the first The regional importance weights in the next iteration The preset iterative learning rate.

[0026] In a fifth aspect, the present invention provides a power distributed service communication analysis technical indicator selection device, applied to the central processing master station of a communication analysis technical indicator selection system. The communication analysis technical indicator selection system includes the central processing master station and several regional management systems communicatively connected to the central processing master station. The communication analysis technical indicator selection method includes: a global weight initialization module, used to obtain the initial central importance weight of each technical indicator based on the initial regional importance weight of each technical indicator in each regional management system, and send it to each regional management system; wherein the initial regional importance weight of each technical indicator in each regional management system is obtained by the following formula: the regional management system obtains the technical indicator values ​​of each technical indicator of each terminal in the previous several acquisition cycles and performs averaging and normalization to obtain the values ​​of each acquisition... The system generates regional normalized values ​​for each technical indicator in each collection period; and determines the initial regional importance weight of each technical indicator based on the regional normalized values ​​of each technical indicator in each collection period, and sends it to the central processing station. The initial central importance weight of each technical indicator is used to trigger the regional management system to obtain the optimal regional importance weight of each technical indicator based on the initial central importance weight of each technical indicator and the regional normalized values ​​of each technical indicator in each collection period, using a stochastic gradient descent method, and sends it to the central processing station. The indicator selection module is used to obtain the optimal central importance weight of each technical indicator based on the optimal regional importance weight of each technical indicator in each regional management system, and select communication analysis technical indicators from each technical indicator based on the optimal central importance weight of each technical indicator.

[0027] Optionally, obtaining the initial central importance weight of each technical indicator based on the initial regional importance weight of each technical indicator in each regional management system includes: averaging the initial regional importance weights of each technical indicator in each regional management system to obtain the initial central importance weight of each technical indicator; obtaining the optimal central importance weight of each technical indicator based on the optimal regional importance weight of each technical indicator in each regional management system includes: averaging the optimal regional importance weights of each technical indicator in each regional management system to obtain the optimal central importance weight of each technical indicator.

[0028] In a sixth aspect, the present invention provides a power distributed business communication analysis technical indicator selection system, comprising a central processing master station and several regional management systems communicatively connected to the central processing master station; wherein, the regional management systems are equipped with the aforementioned communication analysis technical indicator selection device for the regional management system applied to the communication analysis technical indicator selection system; and the central processing master station is equipped with the aforementioned communication analysis technical indicator selection device for the central processing master station applied to the communication analysis technical indicator selection system.

[0029] In a seventh aspect, the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the power distributed business communication analysis technical indicator selection method described above.

[0030] In an eighth aspect, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the power distributed business communication analysis technical indicator selection method described above.

[0031] Compared with the prior art, the present invention has the following beneficial effects: This invention discloses a method for selecting technical indicators for power distributed business communication analysis. First, based on the technical indicator values ​​of each terminal within several previous data collection cycles, the initial regional importance weights of each technical indicator are determined. Then, a central processing master station globally coordinates and distributes these initial weights to each regional management system. The regional management systems update their regional importance weights based on the regional normalized values ​​of each technical indicator from each data collection cycle. Finally, the central processing master station determines the optimal central importance weight based on the optimal regional importance weights of each technical indicator from each regional management system, and selects the communication analysis technical indicators accordingly. This method uses data-driven training weights instead of traditional static weights and fully considers the characteristics and distributed deployment features of each regional management system. It can dynamically adapt to changes in the requirements of power distributed business communication analysis technical indicators in real time, improving the accuracy of the importance weights of the technical indicators, thereby enhancing the communication analysis effect and providing support for communication network resource optimization and performance improvement. Attached Figure Description

[0032] Figure 1 This is a schematic diagram illustrating a typical application scenario of an embodiment of the present invention.

[0033] Figure 2 This is a flowchart illustrating a communication analysis technical indicator selection method applied to a communication analysis technical indicator selection system, according to an embodiment of the present invention.

[0034] Figure 3 This is a flowchart illustrating a method for selecting communication analysis technical indicators in a regional management system, as described in an embodiment of the present invention, applied to a communication analysis technical indicator selection system.

[0035] Figure 4 This is a flowchart illustrating the communication analysis technical indicator selection method applied to the central processing master station of the communication analysis technical indicator selection system according to an embodiment of the present invention.

[0036] Figure 5This is a structural block diagram of a communication analysis technical indicator selection device for a regional management system applied to a communication analysis technical indicator selection system, according to an embodiment of the present invention.

[0037] Figure 6 This is a structural block diagram of a communication analysis technical indicator selection device applied to a communication analysis technical indicator selection system in an embodiment of the present invention. Detailed Implementation

[0038] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0039] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0040] The present invention will now be described in further detail with reference to the accompanying drawings: See Figure 1 This paper illustrates a typical application scenario of the power distributed service communication analysis technical indicator selection method of the present invention. This typical application scenario includes a central processing master station and several regional management systems. The typical application scenario adopts a distributed architecture, with the central processing master station acting as the core decision-making unit and communicating with several regional management systems. These regional management systems cover different regions and are responsible for monitoring and data collection of terminal devices within their respective regions. The power distributed service can be a specific service such as power load management.

[0041] See Figure 2 In one embodiment of the present invention, a method for selecting technical indicators for communication analysis of distributed power services is provided, which is applied to a communication analysis technical indicator selection system. The communication analysis technical indicator selection system includes a central processing master station and several regional management systems that are communicatively connected to the central processing master station.

[0042] Specifically, the method for selecting technical indicators for distributed power business communication analysis includes the following steps: S11: The regional management system obtains the technical indicator values ​​of each terminal in the previous several collection cycles and performs averaging and normalization to obtain the regional normalized value of each technical indicator in each collection cycle. S12: The regional management system determines the initial regional importance weight of each technical indicator based on the regional normalized value of each technical indicator in each collection cycle, and sends it to the central processing station. S13: The central processing station obtains the initial central importance weight of each technical indicator based on the initial regional importance weight of each technical indicator in each regional management system, and sends it to each regional management system. S14: The regional management system, based on the initial central importance weight of each technical indicator and combined with the regional normalized value of each technical indicator in each collection period, uses a stochastic gradient descent method to obtain the optimal regional importance weight of each technical indicator and sends it to the central processing station. S15: The central processing station obtains the optimal central importance weight for each technical indicator based on the optimal regional importance weight of each technical indicator in each regional management system, and selects communication analysis technical indicators from each technical indicator based on the optimal central importance weight of each technical indicator.

[0043] This invention discloses a method for selecting technical indicators for power distributed business communication analysis. First, based on the technical indicator values ​​of each terminal within several previous data collection cycles, the initial regional importance weights of each technical indicator are determined. Then, a central processing master station globally coordinates and distributes these initial weights to each regional management system. The regional management systems update their regional importance weights based on the regional normalized values ​​of each technical indicator from each data collection cycle. Finally, the central processing master station determines the optimal central importance weight based on the optimal regional importance weights of each technical indicator from each regional management system, and selects the communication analysis technical indicators accordingly. This method uses data-driven training weights instead of traditional static weights and fully considers the characteristics and distributed deployment features of each regional management system. It can dynamically adapt to changes in the requirements of power distributed business communication analysis technical indicators in real time, improving the accuracy of the importance weights of the technical indicators, thereby enhancing the communication analysis effect and providing support for communication network resource optimization and performance improvement.

[0044] For example, the power load management service will be used as an example for illustration.

[0045] The power load management system is defined as having one central processing station and N regional management systems. i ( i=1,…,N) regional management systems contain J power load management terminals , No. i ( i Under the regional management system (i = 1, ..., N), the th... j ( j =1,…,J) power load management terminals are identified as follows: The central processing station of the power load management system is identified as follows: .

[0046] The power load management service is defined to include M (values ​​ranging from 4 to 10) technical indicators related to communication. Let the current period be the z-th acquisition cycle (z ranges from greater than K, and cycle length ranges from 10 to 60 minutes). The service acquires the technical indicator acquisition results from the previous K acquisition cycles (values ​​ranging from 500 to 1000), where each power load terminal in the k-th acquisition cycle... The technical indicator values ​​for each NO.m technical indicator were collected at equal sampling time intervals (ranging from 5 to 60 seconds), and the expected value of the above data was calculated as the regional value of the corresponding technical indicator within the collection period, which was then identified as follows. ;according to = Comprising each power load terminal Corresponding technical indicator value vector .

[0047] In a specific example, the power load management system comprises a central processing station and N=10 regional management systems, each containing 25 power load management terminals. The power load management service includes five communication-related technical indicators: latency, transmission bandwidth, abnormal alarm rate, complete transmission rate, and hash check coverage. The technical indicator data is collected over the first K=800 collection periods, where each power load terminal collects its technical indicator values ​​at equal sampling intervals of 10 seconds within each collection period.

[0048] For each data collection cycle, for each regional management system, according to = Calculate the region value vector And a normalization operation is used to obtain the corresponding region normalized value vector. =[ ].

[0049] For example, to facilitate subsequent processing and explanation, for each regional management system, the formula is used... = Assigning collection elements ( p =1,…,K), where, =[ ];according to ={ The training dataset that makes up the current acquisition period .

[0050] In one possible implementation, determining the initial regional importance weight of each technical indicator based on the regional normalized value of each technical indicator in each collection period includes: randomly selecting the regional normalized value of each technical indicator in one collection period as the basic data; and determining the initial regional importance weight of each technical indicator based on the basic data and the principle of minimizing the adaptation deviation of the technical indicator using a static weight calculation method.

[0051] Interpretive, the central processing station of the power load management system is used as the central node of the distributed learning and training network, and each regional management system is used as a sub-node of the distributed learning and training network, thus forming a distributed learning and training network for the importance weights of the technical indicators of power load management business.

[0052] For each regional management system, random selection is made from the training dataset. Select an element from the collection As input data, a static weight calculation method (such as the analytic hierarchy process or entropy weight method) based on the criterion of minimizing the fit deviation of technical indicators is adopted. Calculate the initial regional importance weights of each technical indicator in the regional management system. ,in = The initial regional importance weights of each technical indicator in the regional management system are sent via communication to the central processing station of the power load management system, which is the central node of the distributed learning and training network for importance weights.

[0053] The interpretative, static weight calculation method based on the criterion of minimizing the deviation of technical indicator fit specifically refers to a weight calculation method that uses pre-defined fixed rules to ensure that the calculated weights minimize the deviation between the final evaluation result and the ideal situation. The purpose of this method is to find the most reasonable set of fixed weights so that their allocation most accurately reflects the importance of each technical indicator in actual communication services. Specifically, an optimal technical indicator fit state is first defined manually or set according to requirements. Then, using static mathematical models such as the analytic hierarchy process (AHP) or entropy weight method, a set of weight allocation schemes is repeatedly calculated and searched to minimize the deviation between the comprehensive evaluation value calculated by this set of weights and the preset optimal state.

[0054] In one possible implementation, obtaining the initial central importance weight of each technical indicator based on the initial regional importance weight of each technical indicator in each regional management system includes: averaging the initial regional importance weights of each technical indicator in each regional management system to obtain the initial central importance weight of each technical indicator.

[0055] Explanatory, for each technical indicator, the central processing station follows the formula Calculate the initial central importance weight of the corresponding technical indicators. ; and in accordance with = Composition of global technical indicator importance weight vector initial value Then The communication is sent to each regional management system.

[0056] In one possible implementation, the step of obtaining the optimal regional importance weight for each technical indicator based on its initial central importance weight and the regional normalized value of each technical indicator in each acquisition period, using a stochastic gradient descent method, includes: The update process iteratively continues until a preset iteration termination condition is met, and the current regional importance weight of each technical indicator is taken as the optimal regional importance weight for each technical indicator. The update step includes: randomly selecting the regional normalized value of each technical indicator from a collection period as the update basis data; calculating the estimated value of each technical indicator using a preset inverse function based on its initial central importance weight; wherein the preset inverse function is the inverse function of the function used to calculate the initial regional importance weight of each technical indicator; and obtaining the rate of change of the regional importance weight based on the update basis data and the estimated value using the following formula: =

[0057] in, For the first Rate of change in each iteration To update the basis data, These are estimated values ​​for each technical indicator. For the first The regional importance weights in the next iteration.

[0058] The current regional importance weights of each technical indicator are updated using the following formula: = -

[0059] in, For the first The regional importance weights in the next iteration The preset iterative learning rate.

[0060] Explanatory, the preset iteration termination condition can be that the number of iterations reaches a preset iteration limit or the error between the regional importance weights of two adjacent technical indicators is less than a preset iteration convergence error threshold.

[0061] For example, the upper limit of iteration is set to U (the value can be in the range of K), and the iteration learning rate is... (The value can range from 0.01 to 0.1), and the iterative convergence error threshold is... (The value can be in the range of 0.001 to 0.1).

[0062] For the i The regional management system, for the first s In the next iteration, the region normalized values ​​of each technical indicator for a random acquisition period are selected. As the first i The update of the regional management system is based on data, using And based on the principle of minimizing the deviation of technical indicator fit, a static weight inverse calculation method is used. Calculate the first i Regional management system The estimated value =[ ].

[0063] Then, according to the formula = Calculate the first i The regional importance weights of a regional management system are based on the rate of change between the updated baseline data and the estimated values, specifically the L2 norm of the deviation between the technical indicator values ​​and the estimated values, and the rate of change of the regional importance weights. Finally, according to the formula = - Update # i The regional importance weight of each regional management system.

[0064] If the iterative convergence condition is met: < Conditions or number of iterations s If the iteration upper limit threshold U is exceeded, then according to the formula... = Assign a value and output the first value. i The optimal regional importance weights for each technical indicator of a regional management system. and will Communication is sent to the central node for processing at the master station; otherwise, updates are performed. s = s+1, repeat the update process.

[0065] In one possible implementation, obtaining the optimal central importance weight of each technical indicator based on the optimal regional importance weight of each technical indicator of each regional management system includes: averaging the optimal regional importance weights of each technical indicator of each regional management system to obtain the optimal central importance weight of each technical indicator.

[0066] Explanatory, the central processing station follows the formula Calculate the optimal central importance weight for each technical indicator. .

[0067] In one possible implementation, the step of selecting communication analysis technical indicators from various technical indicators includes: selecting the technical indicator corresponding to the optimal central importance weight with the largest number of preset values ​​as the communication analysis technical indicator, based on the optimal central importance weight of each technical indicator.

[0068] For example, the optimal central importance weight for each technical indicator The calculation operation is performed by sorting from largest to smallest. Based on the sorting results, the top M / 2 technical indicators are selected as communication analysis technical indicators.

[0069] Explanatoryly, the method for selecting technical indicators for power distributed business communication analysis of the present invention uses the collection cycle as a node for looping. When the next collection cycle arrives, the collection cycle number is incremented by 1 and the above process is repeated.

[0070] See Figure 3 In one embodiment of the present invention, a method for selecting technical indicators for communication analysis of distributed power services is provided, which is applied to the regional management system of the communication analysis technical indicator selection system.

[0071] The method for selecting technical indicators for power distributed service communication analysis in this embodiment includes the following steps: S21: Obtain the technical indicator values ​​of each terminal in the previous several collection cycles, and perform averaging and normalization to obtain the regional normalized value of each technical indicator in each collection cycle.

[0072] S22: Based on the regional normalized values ​​of each technical indicator in each collection cycle, determine the initial regional importance weight of each technical indicator and send it to the central processing station; wherein, the initial regional importance weight of each technical indicator is used to trigger the central processing station to obtain the initial central importance weight of each technical indicator based on the initial regional importance weight of each technical indicator in each regional management system and send it to each regional management system. S23: Based on the initial central importance weight of each technical indicator and the regional normalized value of each technical indicator in each collection period, the optimal regional importance weight of each technical indicator is obtained by using a stochastic gradient descent method and sent to the central processing master station; wherein, the optimal regional importance weight of each technical indicator is used to trigger the central processing master station to obtain the optimal central importance weight of each technical indicator based on the optimal regional importance weight of each technical indicator in each regional management system, and to select communication analysis technical indicators from each technical indicator based on the optimal central importance weight of each technical indicator.

[0073] In one possible implementation, determining the initial regional importance weight of each technical indicator based on the regional normalized value of each technical indicator in each collection period includes: randomly selecting the regional normalized value of each technical indicator in one collection period as the basic data; and determining the initial regional importance weight of each technical indicator based on the basic data and the principle of minimizing the adaptation deviation of the technical indicator using a static weight calculation method.

[0074] In one possible implementation, the step of obtaining the optimal regional importance weight for each technical indicator based on its initial central importance weight and the regional normalized value of each technical indicator in each acquisition period, using a stochastic gradient descent method, includes: The update process iteratively continues until a preset iteration termination condition is met, and the current regional importance weight of each technical indicator is taken as the optimal regional importance weight for each technical indicator. The update step includes: randomly selecting the regional normalized value of each technical indicator from a collection period as the update basis data; calculating the estimated value of each technical indicator using a preset inverse function based on its initial central importance weight; wherein the preset inverse function is the inverse function of the function used to calculate the initial regional importance weight of each technical indicator; and obtaining the rate of change of the regional importance weight based on the update basis data and the estimated value using the following formula: =

[0075] in, For the first Rate of change in each iteration To update the basis data, These are estimated values ​​for each technical indicator. For the first The regional importance weights in the next iteration.

[0076] The current regional importance weights of each technical indicator are updated using the following formula: = -

[0077] in, For the first The regional importance weights in the next iteration The preset iterative learning rate.

[0078] See Figure 4 In one embodiment of the present invention, a method for selecting technical indicators for communication analysis in power distributed services is provided, which is applied to the central processing master station of the communication analysis technical indicator selection system.

[0079] The method for selecting technical indicators for power distributed service communication analysis in this embodiment includes the following steps: S31: Based on the initial regional importance weights of each technical indicator in each regional management system, the initial central importance weights of each technical indicator are obtained and sent to each regional management system. Specifically, the initial regional importance weights of each technical indicator in each regional management system are obtained using the following formula: The regional management system obtains the technical indicator values ​​of each terminal in the previous several collection cycles, averages and normalizes them to obtain the regional normalized values ​​of each technical indicator in each collection cycle; and based on the regional normalized values ​​of each technical indicator in each collection cycle, determines the initial regional importance weights of each technical indicator and sends them to the central processing station. The initial central importance weights of each technical indicator are used to trigger the regional management system to obtain the optimal regional importance weights of each technical indicator based on the initial central importance weights of each technical indicator, combined with the regional normalized values ​​of each technical indicator in each collection cycle, using a stochastic gradient descent method, and then sends them to the central processing station.

[0080] S32: Based on the optimal regional importance weight of each technical indicator in each regional management system, obtain the optimal central importance weight of each technical indicator, and select communication analysis technical indicators from each technical indicator based on the optimal central importance weight of each technical indicator.

[0081] In one possible implementation, obtaining the initial central importance weight of each technical indicator based on the initial regional importance weight of each technical indicator in each regional management system includes: averaging the initial regional importance weights of each technical indicator in each regional management system to obtain the initial central importance weight of each technical indicator.

[0082] In one possible implementation, obtaining the optimal central importance weight of each technical indicator based on the optimal regional importance weight of each technical indicator of each regional management system includes: averaging the optimal regional importance weights of each technical indicator of each regional management system to obtain the optimal central importance weight of each technical indicator.

[0083] The following are embodiments of the apparatus of the present invention, which can be used to execute embodiments of the method of the present invention. For details not disclosed in the apparatus embodiments, please refer to the embodiments of the method of the present invention.

[0084] See Figure 5 In another embodiment of the present invention, a power distributed business communication analysis technical indicator selection device is provided, which is applied to the regional management system of the communication analysis technical indicator selection system. It can be used to implement the above-mentioned power distributed business communication analysis technical indicator selection method applied to the regional management system of the communication analysis technical indicator selection system. Specifically, the power distributed business communication analysis technical indicator selection device includes a data processing module, a local weight initialization module, and a local weight optimization module.

[0085] The data processing module is used to acquire the technical indicator values ​​of each terminal in the previous several collection cycles, and then average and normalize them to obtain the regional normalized value of each technical indicator in each collection cycle. The local weight initialization module is used to determine the initial regional importance weight of each technical indicator based on the regional normalized value of each technical indicator in each collection cycle, and send it to the central processing master station. The initial regional importance weight of each technical indicator is used to trigger the central processing master station to obtain the initial central importance weight of each technical indicator based on the initial regional importance weight of each technical indicator in each regional management system, and then send it to each regional management system. The domain management system includes a local weight optimization module. This module uses the initial central importance weight of each technical indicator, combined with the regional normalized value of each technical indicator in each collection period, to obtain the optimal regional importance weight of each technical indicator using a stochastic gradient descent method, and then sends this result to the central processing station. The optimal regional importance weight of each technical indicator triggers the central processing station to obtain the optimal central importance weight of each technical indicator based on the optimal regional importance weight of each technical indicator in the regional management system. Based on the optimal central importance weight of each technical indicator, the central processing station then selects communication analysis technical indicators from among the technical indicators.

[0086] In one possible implementation, determining the initial regional importance weight of each technical indicator based on the regional normalized value of each technical indicator in each collection period includes: randomly selecting the regional normalized value of each technical indicator in one collection period as the basic data; and determining the initial regional importance weight of each technical indicator based on the basic data and the principle of minimizing the adaptation deviation of the technical indicator using a static weight calculation method.

[0087] In one possible implementation, the step of obtaining the optimal regional importance weight for each technical indicator based on its initial central importance weight and the regional normalized value of each technical indicator in each acquisition period, using a stochastic gradient descent method, includes: The update process iteratively continues until a preset iteration termination condition is met, and the current regional importance weight of each technical indicator is taken as the optimal regional importance weight for each technical indicator. The update step includes: randomly selecting the regional normalized value of each technical indicator from a collection period as the update basis data; calculating the estimated value of each technical indicator using a preset inverse function based on its initial central importance weight; wherein the preset inverse function is the inverse function of the function used to calculate the initial regional importance weight of each technical indicator; and obtaining the rate of change of the regional importance weight based on the update basis data and the estimated value using the following formula: =

[0088] in, For the first Rate of change in each iteration To update the basis data, These are estimated values ​​for each technical indicator. For the first The regional importance weights in the next iteration.

[0089] The current regional importance weights of each technical indicator are updated using the following formula: = -

[0090] in, For the first The regional importance weights in the next iteration The preset iterative learning rate.

[0091] See Figure 6 In another embodiment of the present invention, a power distributed business communication analysis technical indicator selection device is provided, which is applied to the central processing master station of the communication analysis technical indicator selection system. It can be used to implement the above-mentioned power distributed business communication analysis technical indicator selection method applied to the central processing master station of the communication analysis technical indicator selection system. Specifically, the power distributed business communication analysis technical indicator selection device includes a global weight initialization module and an indicator selection module.

[0092] The global weight initialization module is used to obtain the initial central importance weight of each technical indicator based on the initial regional importance weight of each technical indicator in each regional management system, and then send it to each regional management system. The initial regional importance weight of each technical indicator in each regional management system is obtained by the following formula: the regional management system obtains the technical indicator values ​​of each terminal in the previous several collection periods, averages and normalizes them to obtain the regional normalized value of each technical indicator in each collection period; and determines the initial regional importance weight of each technical indicator based on the regional normalized value of each technical indicator in each collection period. The data is then sent to the central processing station. The initial central importance weight of each technical indicator is used to trigger the regional management system to obtain the optimal regional importance weight of each technical indicator based on the initial central importance weight of each technical indicator and the regional normalized value of each technical indicator in each collection period, using a stochastic gradient descent method, and then send it to the central processing station. The indicator selection module is used to obtain the optimal central importance weight of each technical indicator based on the optimal regional importance weight of each technical indicator in each regional management system, and to select communication analysis technical indicators from each technical indicator based on the optimal central importance weight of each technical indicator.

[0093] In one possible implementation, obtaining the initial central importance weight of each technical indicator based on the initial regional importance weight of each technical indicator in each regional management system includes: averaging the initial regional importance weights of each technical indicator in each regional management system to obtain the initial central importance weight of each technical indicator.

[0094] In one possible implementation, obtaining the optimal central importance weight of each technical indicator based on the optimal regional importance weight of each technical indicator of each regional management system includes: averaging the optimal regional importance weights of each technical indicator of each regional management system to obtain the optimal central importance weight of each technical indicator.

[0095] In another embodiment of the present invention, a power distributed business communication analysis technical indicator selection system is provided, including a central processing master station and several regional management systems communicatively connected to the central processing master station; wherein, the regional management systems are equipped with the aforementioned communication analysis technical indicator selection device for the central processing master station applied to the communication analysis technical indicator selection system; and the central processing master station is equipped with the aforementioned communication analysis technical indicator selection device for the central processing master station applied to the communication analysis technical indicator selection system.

[0096] All relevant content of each step involved in the aforementioned embodiment of the power distributed business communication analysis technical indicator selection method can be referenced to the functional description of the corresponding functional module of the power distributed business communication analysis technical indicator selection device in the embodiment of the present invention, and will not be repeated here.

[0097] The module division in this embodiment of the invention is illustrative and represents only one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional modules in the various embodiments of the invention can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0098] In another embodiment of the present invention, a computer device is provided, comprising a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions from the computer storage medium to achieve corresponding method flows or corresponding functions. The processor described in this embodiment of the present invention can be used for the selection of technical indicators for power distributed business communication analysis.

[0099] In another embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and extended storage media supported by the computer device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, the storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps related to the selection of power distributed service communication analysis technical indicators in the above embodiments.

[0100] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention 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.

[0101] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. 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 illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0102] 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 boxesFigure 1 The function specified in one or more boxes.

[0103] 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.

[0104] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for selecting technical indicators in power distributed service communication analysis, characterized in that, The system is applied to a communication analysis technical indicator selection system, which includes a central processing station and several regional management systems that are communicatively connected to the central processing station. The method for selecting communication analysis technical indicators includes: The regional management system obtains the technical indicator values ​​of each terminal in the previous several collection periods and performs averaging and normalization to obtain the regional normalized values ​​of each technical indicator in each collection period. The regional management system determines the initial regional importance weight of each technical indicator based on the regional normalized value of each technical indicator in each collection period, and sends it to the central processing station. The central processing station obtains the initial central importance weight of each technical indicator based on the initial regional importance weight of each technical indicator in each regional management system, and sends it to each regional management system. The regional management system, based on the initial central importance weight of each technical indicator and the regional normalized value of each technical indicator in each collection period, uses a stochastic gradient descent method to obtain the optimal regional importance weight of each technical indicator and sends it to the central processing station. The central processing station obtains the optimal central importance weight for each technical indicator based on the optimal regional importance weight of each technical indicator in each regional management system, and selects communication analysis technical indicators from the technical indicators based on the optimal central importance weight of each technical indicator.

2. The method for selecting technical indicators for power distributed service communication analysis according to claim 1, characterized in that, The determination of the initial regional importance weight of each technical indicator based on the regional normalized value of each technical indicator in each collection period includes: The regional normalized values ​​of each technical indicator for a random collection period are used as the basic data. Based on the basic data and the principle of minimizing the deviation of technical indicator fit, the initial regional importance weight of each technical indicator is determined by the static weight calculation method.

3. The method for selecting technical indicators for power distributed service communication analysis according to claim 1, characterized in that, The step of obtaining the initial central importance weight of each technical indicator based on the initial regional importance weight of each technical indicator in each regional management system includes: averaging the initial regional importance weights of each technical indicator in each regional management system to obtain the initial central importance weight of each technical indicator. The step of obtaining the optimal central importance weight of each technical indicator based on the optimal regional importance weight of each technical indicator in each regional management system includes: averaging the optimal regional importance weights of each technical indicator in each regional management system to obtain the optimal central importance weight of each technical indicator.

4. The method for selecting technical indicators for power distributed service communication analysis according to claim 1, characterized in that, The process of obtaining the optimal regional importance weights for each technical indicator based on its initial central importance weights and the regional normalized values ​​of each technical indicator in each acquisition period, using a stochastic gradient descent method, includes: The update process is iterated until a preset iteration termination condition is met, and the current regional importance weight of each technical indicator is taken as the optimal regional importance weight for each technical indicator; the update steps include: The regional normalized values ​​of each technical indicator in a random collection period are used as the basis for updating the data. Based on the initial central importance weights of each technical indicator, the estimated values ​​of each technical indicator are calculated using a preset inverse function; where the preset inverse function is the inverse function of the calculation function for the initial regional importance weights of each technical indicator; and the rate of change of the regional importance weights based on the updated data and the estimated values ​​is obtained using the following formula: = in, For the first Rate of change in each iteration To update the basis data, These are estimated values ​​for each technical indicator. For the first The regional importance weights in the next iteration; The current regional importance weights of each technical indicator are updated using the following formula: = - in, For the first The regional importance weights in the next iteration The preset iterative learning rate.

5. A method for selecting technical indicators for power distributed service communication analysis, characterized in that, A regional management system applied to a communication analysis technical indicator selection system, wherein the communication analysis technical indicator selection system includes a central processing station and several regional management systems that are communicatively connected to the central processing station. The method for selecting communication analysis technical indicators includes: The technical indicator values ​​of each terminal in the previous several collection periods are obtained and averaged and normalized to obtain the regional normalized values ​​of each technical indicator in each collection period. Based on the regional normalized values ​​of each technical indicator in each collection period, the initial regional importance weight of each technical indicator is determined and sent to the central processing station; wherein, the initial regional importance weight of each technical indicator is used to trigger the central processing station to obtain the initial central importance weight of each technical indicator based on the initial regional importance weight of each technical indicator in each regional management system, and send it to each regional management system. Based on the initial central importance weights of each technical indicator and the regional normalized values ​​of each technical indicator in each collection period, a stochastic gradient descent-based method is used to obtain the optimal regional importance weights of each technical indicator and send them to the central processing station. The optimal regional importance weights of each technical indicator are used to trigger the central processing station to obtain the optimal central importance weights of each technical indicator based on the optimal regional importance weights of each technical indicator in each regional management system, and to select communication analysis technical indicators from the technical indicators based on the optimal central importance weights of each technical indicator.

6. The method for selecting technical indicators for power distributed service communication analysis according to claim 5, characterized in that, The determination of the initial regional importance weight of each technical indicator based on the regional normalized value of each technical indicator in each collection period includes: The regional normalized values ​​of each technical indicator for a random collection period are used as the basic data. Based on the basic data and the principle of minimizing the deviation of technical indicator fit, the initial regional importance weight of each technical indicator is determined by the static weight calculation method.

7. The method for selecting technical indicators for power distributed service communication analysis according to claim 5, characterized in that, The process of obtaining the optimal regional importance weights for each technical indicator based on its initial central importance weights and the regional normalized values ​​of each technical indicator in each acquisition period, using a stochastic gradient descent method, includes: The update process is iterated until a preset iteration termination condition is met, and the current regional importance weight of each technical indicator is taken as the optimal regional importance weight for each technical indicator; the update steps include: The regional normalized values ​​of each technical indicator in a random collection period are used as the basis for updating the data. Based on the initial central importance weights of each technical indicator, the estimated values ​​of each technical indicator are calculated using a preset inverse function; where the preset inverse function is the inverse function of the calculation function for the initial regional importance weights of each technical indicator; and the rate of change of the regional importance weights based on the updated data and the estimated values ​​is obtained using the following formula: = in, For the first Rate of change in each iteration To update the basis data, These are estimated values ​​for each technical indicator. For the first The regional importance weights in the next iteration; The current regional importance weights of each technical indicator are updated using the following formula: = - in, For the first The regional importance weights in the next iteration The preset iterative learning rate.

8. A method for selecting technical indicators for power distributed service communication analysis, characterized in that, A central processing master station is applied to a communication analysis technical indicator selection system, the communication analysis technical indicator selection system includes a central processing master station and several regional management systems that are communicatively connected to the central processing master station. The method for selecting communication analysis technical indicators includes: Based on the initial regional importance weights of each technical indicator in each regional management system, the initial central importance weights of each technical indicator are obtained and sent to each regional management system. The initial regional importance weights of each technical indicator in each regional management system are obtained through the following formula: The regional management system obtains the technical indicator values ​​of each terminal in the previous several collection cycles and performs averaging and normalization to obtain the regional normalized value of each technical indicator in each collection cycle; and determines the initial regional importance weight of each technical indicator based on the regional normalized value of each technical indicator in each collection cycle, and sends it to the central processing station; the initial central importance weight of each technical indicator is used to trigger the regional management system to obtain the optimal regional importance weight of each technical indicator based on the initial central importance weight of each technical indicator, combined with the regional normalized value of each technical indicator in each collection cycle, using a stochastic gradient descent method, and sends it to the central processing station. Based on the optimal regional importance weight of each technical indicator in each regional management system, the optimal central importance weight of each technical indicator is obtained, and communication analysis technical indicators are selected from each technical indicator based on the optimal central importance weight of each technical indicator.

9. The method for selecting technical indicators for power distributed service communication analysis according to claim 8, characterized in that, The step of obtaining the initial central importance weight of each technical indicator based on the initial regional importance weight of each technical indicator in each regional management system includes: averaging the initial regional importance weights of each technical indicator in each regional management system to obtain the initial central importance weight of each technical indicator. The step of obtaining the optimal central importance weight of each technical indicator based on the optimal regional importance weight of each technical indicator in each regional management system includes: averaging the optimal regional importance weights of each technical indicator in each regional management system to obtain the optimal central importance weight of each technical indicator.

10. A device for selecting technical indicators for power distributed service communication analysis, characterized in that, A regional management system applied to a communication analysis technical indicator selection system, wherein the communication analysis technical indicator selection system includes a central processing station and several regional management systems that are communicatively connected to the central processing station. The communication analysis technical indicator selection device includes: The data processing module is used to obtain the technical indicator values ​​of each terminal in the previous several collection cycles and to average and normalize them to obtain the regional normalized values ​​of each technical indicator in each collection cycle. The local weight initialization module is used to determine the initial regional importance weight of each technical indicator based on the regional normalized value of each technical indicator in each collection period, and send it to the central processing master station; wherein, the initial regional importance weight of each technical indicator is used to trigger the central processing master station to obtain the initial central importance weight of each technical indicator based on the initial regional importance weight of each technical indicator in each regional management system, and send it to each regional management system. The local weight optimization module is used to obtain the optimal regional importance weight of each technical indicator based on the initial central importance weight of each technical indicator and the regional normalized value of each technical indicator in each collection period, using a stochastic gradient descent method, and then send it to the central processing master station. The optimal regional importance weight of each technical indicator is used to trigger the central processing master station to obtain the optimal central importance weight of each technical indicator based on the optimal regional importance weight of each technical indicator in each regional management system, and to select communication analysis technical indicators from the technical indicators based on the optimal central importance weight of each technical indicator.

11. The power distributed service communication analysis technical indicator selection device according to claim 10, characterized in that, The determination of the initial regional importance weight of each technical indicator based on the regional normalized value of each technical indicator in each collection period includes: The regional normalized values ​​of each technical indicator for a random collection period are used as the basic data. Based on the basic data and the principle of minimizing the deviation of technical indicator fit, the initial regional importance weight of each technical indicator is determined by the static weight calculation method.

12. The power distributed service communication analysis technical indicator selection device according to claim 10, characterized in that, The process of obtaining the optimal regional importance weights for each technical indicator based on its initial central importance weights and the regional normalized values ​​of each technical indicator in each acquisition period, using a stochastic gradient descent method, includes: The update process is iterated until a preset iteration termination condition is met, and the current regional importance weight of each technical indicator is taken as the optimal regional importance weight for each technical indicator; the update steps include: The regional normalized values ​​of each technical indicator in a random collection period are used as the basis for updating the data. Based on the initial central importance weights of each technical indicator, the estimated values ​​of each technical indicator are calculated using a preset inverse function; where the preset inverse function is the inverse function of the calculation function for the initial regional importance weights of each technical indicator; and the rate of change of the regional importance weights based on the updated data and the estimated values ​​is obtained using the following formula: = in, For the first Rate of change in each iteration To update the basis data, These are estimated values ​​for each technical indicator. For the first The regional importance weights in the next iteration; The current regional importance weights of each technical indicator are updated using the following formula: = - in, For the first The regional importance weights in the next iteration The preset iterative learning rate.

13. A device for selecting technical indicators for power distributed service communication analysis, characterized in that, A central processing master station is applied to a communication analysis technical indicator selection system, the communication analysis technical indicator selection system includes a central processing master station and several regional management systems that are communicatively connected to the central processing master station. The method for selecting communication analysis technical indicators includes: The global weight initialization module is used to obtain the initial central importance weight of each technical indicator based on the initial regional importance weight of each technical indicator in each regional management system, and send it to each regional management system. The initial regional importance weights of each technical indicator in each regional management system are obtained through the following formula: The regional management system obtains the technical indicator values ​​of each terminal in the previous several collection cycles and performs averaging and normalization to obtain the regional normalized value of each technical indicator in each collection cycle; and determines the initial regional importance weight of each technical indicator based on the regional normalized value of each technical indicator in each collection cycle, and sends it to the central processing station; the initial central importance weight of each technical indicator is used to trigger the regional management system to obtain the optimal regional importance weight of each technical indicator based on the initial central importance weight of each technical indicator, combined with the regional normalized value of each technical indicator in each collection cycle, using a stochastic gradient descent method, and sends it to the central processing station. The indicator selection module is used to obtain the optimal central importance weight of each technical indicator based on the optimal regional importance weight of each technical indicator in each regional management system, and to select communication analysis technical indicators from each technical indicator based on the optimal central importance weight of each technical indicator.

14. The power distributed service communication analysis technical indicator selection device according to claim 13, characterized in that, The step of obtaining the initial central importance weight of each technical indicator based on the initial regional importance weight of each technical indicator in each regional management system includes: averaging the initial regional importance weights of each technical indicator in each regional management system to obtain the initial central importance weight of each technical indicator. The step of obtaining the optimal central importance weight of each technical indicator based on the optimal regional importance weight of each technical indicator in each regional management system includes: averaging the optimal regional importance weights of each technical indicator in each regional management system to obtain the optimal central importance weight of each technical indicator.

15. A power distributed service communication analysis technical indicator selection system, characterized in that, The system includes a central processing station and several regional management systems that are communicatively connected to the central processing station; wherein, the regional management systems are equipped with the communication analysis technical indicator selection device as described in any one of claims 10 to 12; and the central processing station is equipped with the communication analysis technical indicator selection device as described in claim 13 or 14.

16. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the power distributed service communication analysis technical indicator selection method as described in any one of claims 1 to 9.

17. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the power distributed business communication analysis technical indicator selection method as described in any one of claims 1 to 9.

Images