Gridding information management method, device, equipment, medium and program of power distribution network

Abstract

The invention belongs to the technical field of information management, and particularly relates to a power distribution network gridding information management method and device, equipment, a medium and a program. The method comprises the steps of obtaining power distribution network evaluation indexes of a specified region; operating the evaluation model by taking the power distribution network evaluation index of the specified region as input to obtain an evaluation result of the specified region; the evaluation result is obtained by weighting the nonlinearly adjusted interval weight on the basis of the membership parameter of the specified region; wherein the non-linear adjustment interval weight is determined according to a gridding mode and / or a gridding target of the power distribution network. According to the method, differentiation processing is carried out on the index membership degree interval, evaluation requirements of different types of grids can be met, the grid score does not only depend on traditional weighting any more, the influence of the optimal or worst index on the whole can be highlighted, and the evaluation accuracy is improved. The problems of complex membership calculation and low evaluation accuracy under the condition of multi-grid information are effectively solved.

Description

Technical Field

[0001] The present application belongs to the field of information management technology, and in particular, relates to methods, devices, equipment, media and programs for grid information management of distribution networks. Background Art

[0002] At present, power grid companies pay unprecedented attention to the needs of stable power supply, energy efficiency management, precise control, and real-time response. With the development of cities and the growth of electricity demand, traditional distribution network planning can no longer meet the needs. There are many problems, such as complex grid structure, chaotic wiring, chaotic substation layout, high equipment load rate, and cross-supply area. Traditional distribution network planning has defects such as inconsistent standards, large subjective arbitrariness, and chaotic operation and management of the power grid after planning. These problems seriously affect the power supply capacity and reliability of the distribution network, and can no longer meet the basic requirements of power companies to build a modern distribution network. Therefore, power grid companies at home and abroad have begun to study and practice grid construction of distribution networks. Grid construction can not only improve the operating efficiency of the power grid, but also enhance the reliability and flexibility of the power grid to meet the growing power demand. Through grid construction, power grid companies can reasonably plan the layout and grid structure of substations to improve the power supply capacity and anti-interference ability of the power grid. Grid construction also helps to realize the intelligent management of the power grid, and realize real-time monitoring and optimized scheduling of the power grid through advanced information technology and automation equipment.

[0003] The new power system is built on the basis of many new grid structures such as smart grid and energy Internet, and all business calibers have basically realized online, data-based and intelligent. Power supply companies or power stations are important access points for power grid data. Under the management of many grids, there are problems such as large and complex data acquisition, insufficient data connectivity, and insufficient multi-platform integration. It is necessary to explore and establish a fusion management platform solution based on high-capacity, low-power, and portable technical features such as cloud computing that is suitable for the new power system. The construction of a new power system depends on the coordinated interaction of the power generation side, the power grid side, and the load side, and accelerates the construction of the characteristics of power grid perception capabilities, information middle platform, and ubiquitous interconnection. At the same time, it is necessary to deeply realize that after the access of new energy, the power supply will fluctuate greatly and the uncertainty on the load side will increase. The deep integration of information systems and power systems is imperative. Cloud computing is the key supporting technology for solving the computing power and algorithms in traditional fields such as the energy industry. It has been widely used in power grid control, fault handling, expert diagnosis and other fields.

[0004] At the same time, the grid construction and management model of power grid promotion is still in the planning and exploration stage. In order to ensure that the distribution network after grid planning and construction can meet the demand for load growth and achieve high-level operation of various operating parameters, it is necessary to conduct a comprehensive evaluation and analysis on whether the grid construction of distribution network is advancing in an orderly manner and whether the overall service level of the power grid is improved. The comprehensive evaluation of the power grid is limited by the difficulty of integrating cross-business data resources and the difficulty of selecting evaluation indicators to reflect the actual operation status of the distribution network. Power management units at all levels are constantly revising the existing distribution network operation management methods and technical systems to adapt to the rapid changes in the scale and form of the distribution network, but there is currently no unified evaluation method and model. However, the current comprehensive evaluation of distribution networks generally has shortcomings such as being relatively simple, the selection of indicators is not comprehensive, and there is a lack of corresponding feedback mechanism. Summary of the invention

[0005] The embodiments of the present application provide a distribution network grid information management method, device, equipment, medium and program, which can solve the problem of difficult comprehensive management of distribution networks.

[0006] In a first aspect, an embodiment of the present application provides a distribution network grid information management method, characterized by comprising:

[0007] Acquire distribution network evaluation indicators of a specified area; the distribution network evaluation indicators include at least one of a load parameter, a self-healing parameter, an efficiency parameter, and an interaction parameter;

[0008] The distribution network evaluation index of the designated area is used as input to run the evaluation model to obtain the evaluation result of the designated area; the evaluation model is a neural network model obtained by training based on samples and labels; the evaluation result is obtained by weighting the interval weights adjusted nonlinearly based on the membership parameter of the designated area;

[0009] The interval weight of the nonlinear adjustment is determined according to the gridding mode and / or gridding target of the distribution network.

[0010] In a possible implementation of the first aspect, the step of taking the distribution network evaluation index of the designated area as input, running the evaluation model, and obtaining the evaluation result of the designated area includes:

[0011] Inputting the distribution network evaluation index of the designated area into the evaluation model to obtain the membership parameter of the designated area;

[0012] According to a preset weight model, a weight value corresponding to the interval where the membership parameter of the designated area is located is calculated, and an evaluation result of the designated area is obtained according to the weight value and the membership parameter of the designated area;

[0013] The weight model is determined according to the gridding mode and / or gridding target of the distribution network.

[0014] In a possible implementation of the first aspect, the step of taking the distribution network evaluation index of the designated area as input, running the evaluation model, and obtaining the evaluation result of the designated area includes:

[0015] The distribution network evaluation index and adjustment parameters of the designated area are input into the evaluation model to obtain the evaluation result of the designated area; wherein the adjustment parameters include at least one of the distribution network reliability classification of the designated area, the gridding method of the distribution network, and the gridding target of the distribution network.

[0016] In a possible implementation of the first aspect, the evaluation model includes an activation function layer, and the activation function layer satisfies: calculating the continuous quantity membership based on a first activation function, and calculating the discrete quantity membership based on a second activation function; wherein the first activation function is different from the second activation function.

[0017] In a possible implementation manner of the first aspect, the step of obtaining a distribution network evaluation index of a specified area includes:

[0018] A distributed database based on a data storage and computing layer deployed at the edge end processes and caches distribution network grid information in real time; wherein the concurrent read and write parameters of the distributed database are higher than a preset first threshold, and the storage capacity parameter is higher than a preset second parameter;

[0019] Based on the dynamic execution of the scheduling algorithm of the logical control layer deployed in the cloud, the multi-grid tasks are decomposed and allocated to the hardware resources for processing; wherein the multi-grid tasks are executed based on the gridding information of the distribution network;

[0020] Based on the data access layer deployed in the cloud, the overall network software and hardware resources are unified and integrated into a virtual resource pool, and the distribution network evaluation indicators of the specified area are obtained through the virtual resource pool.

[0021] In a possible implementation of the first aspect, the cloud is communicatively connected with the grid information collection system:

[0022] The grid information acquisition system receives multi-channel sensor signals, obtains collected data after signal conditioning and data aggregation, and communicates the collected signals to the cloud platform and the host computer;

[0023] The multi-channel sensor signals are obtained via multiple sensors deployed in each grid.

[0024] Compared with the prior art, the above method has the following beneficial effects:

[0025] Through the adaptive membership calculation method based on neural network and the piecewise progressive weighted calculation comprehensive evaluation model, the index membership intervals are differentiated, which can adapt to the evaluation needs of different types of grids. The grid score no longer relies solely on traditional weighting, and can highlight the overall impact of the best or worst indicator, effectively solving the problems of complex membership calculation and low evaluation accuracy in the case of multi-grid information.

[0026] In a second aspect, an embodiment of the present application provides a distribution network grid information management device, including:

[0027] An acquisition module, used to acquire distribution network evaluation indicators of a specified area; the distribution network evaluation indicators include at least one of a load parameter, a self-healing parameter, an efficiency parameter, and an interactive parameter;

[0028] A network module is used to use the distribution network evaluation index of the designated area as input, run an evaluation model, and obtain an evaluation result of the designated area; the evaluation model is a neural network model obtained by training based on samples and labels; the evaluation result is obtained by weighting the interval weights adjusted nonlinearly based on the membership parameter of the designated area;

[0029] The interval weight of the nonlinear adjustment is determined according to the gridding mode and / or gridding target of the distribution network.

[0030] In a third aspect, an embodiment of the present application provides a terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the distribution network grid information management method described in any one of the first aspects above is implemented.

[0031] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the distribution network grid information management method described in any one of the first aspects above is implemented.

[0032] In a fifth aspect, an embodiment of the present application provides a computer program product. When the computer program product is run on a terminal device, the terminal device executes the distribution network grid information management method described in any one of the first aspects above.

[0033] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0035] Figure 1 It is a flow chart of the distribution network grid information management method provided in the embodiment of the present application.

[0036] Figure 2 It is a structural diagram of the distribution network grid information management device provided in an embodiment of the present application.

[0037] Figure 3 It is a schematic diagram of the structure of the terminal device provided in an embodiment of the present application.

[0038] Figure 4 It is a comprehensive evaluation index system provided in the embodiments of the present application.

[0039] Figure 5 It is a block diagram of the distribution network grid information collection system provided in an embodiment of the present application.

[0040] Figure 6 It is a structural block diagram of the distribution network grid information cloud platform provided in the embodiment of the present application.

[0041] Figure 7 It is a distribution network grid information comprehensive management system provided in the embodiment of the present application.

[0042] Figure 8 It is a block diagram of the comprehensive evaluation system provided in the embodiment of the present application.

[0043] Reference numerals:

[0044] Acquisition module 201;

[0045] Network module 202;

[0046] Terminal device 30;

[0047] Processor 301;

[0048] Memory 302;

[0049] Computer program 303. DETAILED DESCRIPTION

[0050] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0051] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0052] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0053] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0054] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0055] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0056] The present application provides a distribution network grid information management method, including:

[0057] Step 102, obtaining distribution network evaluation indicators of a specified area; the distribution network evaluation indicators include at least one of a load parameter, a self-healing parameter, an efficiency parameter, and an interaction parameter;

[0058] Step 104, using the distribution network evaluation index of the designated area as input, running an evaluation model to obtain an evaluation result of the designated area; the evaluation model is a neural network model obtained through sample and label training; the evaluation result is obtained after weighting the interval weights adjusted nonlinearly based on the membership parameter of the designated area;

[0059] The interval weight of the nonlinear adjustment is determined according to the gridding mode and / or gridding target of the distribution network.

[0060] In an optional implementation of this embodiment, step 102 is used to obtain specific values ​​of distribution network evaluation indicators in a specified area, and the construction of distribution network evaluation indicators can be achieved through the following steps:

[0061] Step 101.1, obtaining preliminary indicators based on the distribution network gridding method and / or gridding target;

[0062] Step 101.2, verifying the rationality of the preliminary indicator based on the information contribution rate theory to obtain a verification indicator;

[0063] Step 101.3, performing feedback inspection according to the dynamic adjustment index system of the distribution network, wherein the index system includes the verification index.

[0064] For example, for step 101.1, due to the complex structure of the grid system of the distribution network, the grid construction evaluation indicators are screened from four aspects: high load, high self-healing, high efficiency, and high interaction, taking into account the development trend of digitalization and new energy of the new power system;

[0065] In particular, there may be correlations among the preliminary evaluation indicators, which will cause some indicators to be calculated repeatedly. Therefore, it is necessary to screen the correlations of the preliminary indicators, eliminate some highly correlated indicators through screening, and highlight the key indicators. According to the information contribution rate theory, the key theories are as follows:

[0066] First, the data of the preliminary indicators are standardized to obtain a standardized matrix, and then the correlation coefficient matrix is ​​constructed; secondly, the correlation coefficient matrix is ​​decomposed by eigenvalue to obtain the eigenvalue and the corresponding eigenvector. The contribution rate y in the indicator is calculated according to the variance contribution rate formula k :

[0067]

[0068] Calculate the indicator loading matrix. According to the principle that the cumulative variance contribution rate is greater than 85%, select the principal components corresponding to the first k eigenvalues ​​to obtain the factor loading b of the i-th indicator on the j-th principal component. ij Matrix, as shown in the formula, based on the absolute value of the principal component loading matrix |b ij |Perform index screening,|b ij | is positively correlated with the principal component, |b ij The larger the | is, the more significant it is to the evaluation result, so it will be retained, otherwise it will be deleted.

[0069]

[0070] Finally, the rationality of the indicators is tested. Generally, the percentage of the information retained by the indicators after screening to the information retained after the preliminary screening is used as the basis for testing the rationality of the indicator system. Let S be the matrix of the indicator data covariance, trS be the trace of the covariance matrix, which represents the sum of the main diagonal elements of the covariance matrix; n is the number of indicators after the final screening, and N is the number of indicators after the preliminary screening. Then the indicator information contribution rate I m for:

[0071]

[0072] It is generally believed that if the indicators after the initial screening of less than 30% can reflect more than 95% of the original information, the indicator system can be considered to be reasonably constructed; and when I m When it is ≥90%, the indicator system is considered to have passed the rationality test.

[0073] In addition, the feedback inspection index is mainly due to the continuous transformation of the distribution network, which leads to changes in the evaluation objectives. As time goes by, it is necessary to pay attention to the network structure of the distribution network, compare whether the previous evaluation is reasonable, and then make a scientific and reasonable analysis to reasonably adjust the index system; secondly, it is necessary to pay attention to the operating status of the distribution network in different environments. Once the environment changes, the evaluation index system must be adjusted.

[0074] Combining the above examples, we can establish Figure 4 The grid-based comprehensive evaluation index system of distribution network is shown.

[0075] Further, based on Figure 4 The indicator system shown in the figure establishes an evaluation factor set, which is divided into 4 intermediate layers and 15 evaluation indicators, denoted as U = {u1, u2, u3, u4}, among which high load is denoted as A = {a1, a2, a3, a4}, high self-healing is denoted as B = {b1, b2, b3, b4}, high efficiency is denoted as C = {c1, c2, c3, c4}, and high interaction is denoted as D = {d1, d2, d3}.

[0076] In another optional implementation of this embodiment, the weight part can establish an optimal subjective and objective weight model, that is:

[0077] In view of the multi-level and strong fuzziness of the distribution network, the optimal subjective and objective comprehensive weight based on multiplication synthesis normalization can be used to obtain the optimal solution of the indicator weight coefficient, thereby constructing a combined weight that combines subjective and objective factors. The optimal subjective and objective comprehensive weight calculation formula using multiplication synthesis normalization is:

[0078]

[0079] Among them, W1 is the subjective weight determined by the hierarchical analysis method, W2 is the objective weight determined by the CRITIC method, and W * is the calculated combined weight.

[0080] In another optional implementation of this embodiment, nonlinearly adjusted interval weighting is used in the scoring criteria to provide a scoring process that combines a single indicator scoring standard with a type-based grid scoring standard in view of the differences in grid construction foundations in different regions. This method can more accurately classify and evaluate urban, suburban and rural grids.

[0081] By way of example and not limitation, Figure 4 Based on the indicator system shown, the interval weights of nonlinear adjustment can be set based on the following table.

[0082]

[0083]

[0084] According to the above embodiment, in yet another embodiment, two branching schemes in a parallel relationship are provided as follows.

[0085] In the first branch scheme, the step of taking the distribution network evaluation index of the designated area as input, running the evaluation model, and obtaining the evaluation result of the designated area includes:

[0086] Inputting the distribution network evaluation index of the designated area into the evaluation model to obtain the membership parameter of the designated area;

[0087] According to a preset weight model, a weight value corresponding to the interval where the membership parameter of the designated area is located is calculated, and an evaluation result of the designated area is obtained according to the weight value and the membership parameter of the designated area;

[0088] The weight model is determined according to the gridding mode and / or gridding target of the distribution network.

[0089] In this branch solution, the network is responsible for outputting the membership degree. According to the established evaluation index system, the network output of this solution corresponds to the membership degree of 15 indicators, and the interval weights of nonlinear adjustment are calculated by post-processing. This solution is conducive to reducing the difficulty of network training, and the logic of nonlinear adjustment interval weight calculation is clearer through post-processing.

[0090] In the second branch scheme, the step of taking the distribution network evaluation index of the designated area as input, running the evaluation model, and obtaining the evaluation result of the designated area includes:

[0091] The distribution network evaluation index and adjustment parameters of the designated area are input into the evaluation model to obtain the evaluation result of the designated area; wherein the adjustment parameters include at least one of the distribution network reliability classification of the designated area, the gridding method of the distribution network, and the gridding target of the distribution network.

[0092] In this branching scheme, the network is responsible for outputting the membership degree after the nonlinear interval weight adjustment. This scheme uses the black-box network model to directly output the results, which is more convenient and efficient in practical applications.

[0093] In the first branch scheme, the network model constructs a training set through a large amount of indicator data. The core purpose of this training set is to perform adaptive calculations on the membership of each evaluation indicator, and to automatically adjust the membership results according to actual data and specific circumstances, effectively solving the problems of insufficient versatility and poor flexibility in membership calculations in traditional methods. Through adaptive calculations, the role and importance of each indicator in the comprehensive evaluation can be more accurately reflected.

[0094] In the second branch scheme, the calculated membership degree and the scoring model are first processed and combined with the established evaluation criteria to establish a comprehensive evaluation model. This model adopts a black box network model approach. Only the indicator data needs to be input, and the network model can directly output the results, further improving the evaluation efficiency and making the results more objective and reliable.

[0095] According to any of the above embodiments, in yet another embodiment:

[0096] The evaluation model includes an activation function layer, which satisfies: calculating continuous quantity membership based on a first activation function, and calculating discrete quantity membership based on a second activation function; wherein the first activation function is different from the second activation function.

[0097] More specifically, this embodiment learns the nonlinear mapping relationship between the index value and the membership value by training the neural network model, and can automatically adjust the membership function according to the index data to adapt to different data distributions and fuzzy sets. Moreover, through a large number of training samples, the neural network can capture the potential mode of the membership, thereby enhancing the adaptive ability of the membership calculation, which is very suitable for the membership calculation of a large amount of complex data of gridded indicators. The membership calculation based on the neural network is divided into two parts: forward propagation and membership calculation.

[0098] (1) Forward Propagation

[0099] Forward propagation is completed by layer-by-layer calculation from the input layer to the output layer. The output of each layer is a linear combination processed by the activation function. Assume that the neural network contains L layers, and the output of each layer is represented by h (l) Represented, then the output of the lth layer can be expressed as:

[0100] h (l) =σ(W (l) h (l-1) +b (l) ) (2)

[0101] Where: W (l) represents the weight matrix of the lth layer; b (l) represents the bias vector of the lth layer; σ represents the activation function; h (l-1) is the output of the previous layer.

[0102] (2) Membership calculation (i.e. activation function layer)

[0103] If the indicator membership value is a continuous quantity (a real number between [0, 1]), the Sigmoid function is used for calculation:

[0104]

[0105] Where z is the linear combination result of the output layer neurons.

[0106] If the indicator membership value is a discrete classification result (the probability of belonging to a certain category), the Softmax function is used for calculation:

[0107]

[0108] where z i is the linear combination result of each category, A i (x) represents the membership of input x to category i.

[0109] In an optional implementation, line operation index data is usually measured as a continuous quantity, including standard wiring rate, line overload rate, N-1 pass rate, etc.; while index data such as electric vehicle charging piles and substation outgoing line intervals are usually measured as discrete quantities.

[0110] (3) Loss Function

[0111] The loss function used when training a neural network (such as cross entropy or mean square error) is used to measure the difference between the predicted membership and the actual membership. For continuous membership prediction, the following formula is often used for calculation:

[0112]

[0113] Where N is the number of samples; A i is the true membership; is the membership predicted by the neural network; the weights and biases of the neural network are adjusted through the back propagation algorithm to minimize the loss function and thus optimize the membership calculation model.

[0114] On this basis, if the first branch scheme mentioned above is adopted, the segmented progressive weighted comprehensive score calculation model can be constructed as follows:

[0115] In order to more flexibly reflect the relationship between weight and membership, the membership is divided into intervals, and different weight enhancement factors are assigned to the membership in different intervals. i ·A i It reflects the basic influence of the weight and membership of each indicator; using 1+β·g(A i ) introduces nonlinear adjustment, according to the membership degree A i The interval provides different weight enhancement factors; by adjusting β, the amplitude of progressive enhancement can be controlled, which is suitable for different application scenarios. A larger β will make the index with high membership more significantly affected, while a smaller β is close to the traditional weighted average formula. The specific calculation formula is as follows:

[0116]

[0117] Among them, S is the comprehensive score;

[0118] W i is the weight of the i-th indicator;

[0119] A i is the membership degree of the i-th indicator;

[0120] β is the adjustment factor, which can control the progressive amplitude;

[0121] g(A i) is a progressive function that determines the improvement in different intervals, as an example but not a limitation:

[0122]

[0123] This model is suitable for scenarios where extra attention needs to be paid to indicators with higher membership. In the grid evaluation system, indicators with high membership (such as more than 0.9) mean better grid construction effects, so they are given greater weight in the comprehensive score. By differentiating the different membership intervals, it provides great flexibility, and by adjusting the parameter β, it can adapt to the evaluation needs of different types of grids.

[0124] According to the above embodiment, in yet another embodiment:

[0125] The step of obtaining the distribution network evaluation index of the specified area includes:

[0126] A distributed database based on a data storage and computing layer deployed at the edge end processes and caches distribution network grid information in real time; wherein the concurrent read and write parameters of the distributed database are higher than a preset first threshold, and the storage capacity parameter is higher than a preset second parameter;

[0127] Based on the dynamic execution of the scheduling algorithm of the logical control layer deployed in the cloud, the multi-grid tasks are decomposed and allocated to the hardware resources for processing; wherein the multi-grid tasks are executed based on the gridding information of the distribution network;

[0128] Based on the data access layer deployed in the cloud, the overall network software and hardware resources are unified and integrated into a virtual resource pool, and the distribution network evaluation indicators of the specified area are obtained through the virtual resource pool.

[0129] The cloud is connected to the grid information collection system:

[0130] The grid information acquisition system receives multi-channel sensor signals, obtains collected data after signal conditioning and data aggregation, and communicates the collected signals to the cloud platform and the host computer;

[0131] The multi-channel sensor signals are obtained via multiple sensors deployed in each grid.

[0132] In an optional implementation, the architecture provided in this embodiment can be specifically designed as a distribution network grid information collection system based on STM32 and FPGA, in conjunction with a cloud-edge-end distribution network grid information management system based on a cloud platform.

[0133] 1. Design of distribution network grid information collection system based on STM32 and FPGA

[0134] This implementation method is based on a modular design concept. Combined with the advantages of distributed deployment of the cloud platform, the information acquisition circuit, signal conditioning circuit, analog-to-digital conversion circuit, and communication circuit of the grid information acquisition system are modularly designed. It has the ability of regional coverage, real-time acquisition, and remote monitoring and scheduling, improves the efficiency of multi-source information acquisition and processing, and can effectively improve the reliability of the system. At the same time, the system has strong maintainability and later scalability. The main control circuit of the system is composed of STM32 and FPGA. The rich peripheral interfaces of STM32 can support multiple ways of combined communication with the host computer; and FPGA can realize the multi-interface expansion and multi-source data acquisition and processing of the system due to its good flexibility. The system design uses STM32 and FPGA as the main circuits, which can realize data acquisition, transmission, processing, interaction based on the combination of 4G communication and CAN bus communication, as well as communication between STM32 and the host computer. The software design of STM32 and FPGA can meet the design requirements of the system for multiple types of distribution network grids. The flexible expansion performance of FPGA can be used to realize multiple CAN bus data acquisition channels, and the data between different acquisition channels are independent of each other and do not affect each other, so that the construction of regional distribution network grids can achieve comprehensive coverage of distributed monitoring. Figure 5 shown.

[0135] 2. Cloud-edge-end distribution network grid information management system based on cloud platform

[0136] In the context of the grid division and management model of the distribution network, the operating data of the new power system has expanded dramatically, and traditional data processing systems can no longer meet the demand. The huge volume, diverse types and low value density characteristics of power data require us to adopt more scientific processing methods to accurately extract key information. By integrating cloud platform technology into the existing grid data processing system, efficient and distributed data processing can be achieved, with more emphasis on data analysis and mining, and more accurate power demand and load forecasting. The architecture of the cloud computing platform proposed in the present invention is a data storage and operation layer, a logic control layer, a data access layer and a data application layer. The cloud platform structure block diagram is shown in the following figure. Figure 6As shown in the figure: In the data storage and operation layer at the edge, by establishing a MongoDB database, it is responsible for the real-time processing and storage of multi-type grid data, and realizes the distributed caching of grid information. The MongoDB database has the advantages of high concurrent reading and writing, massive information storage and high reliability expansion, which is completely consistent with the complex information scenarios under multiple grids. Using this database for distributed caching can not only ensure the integrity of the original data information, but also solve the communication load problem caused by uploading a large amount of data from multiple grids to the cloud; in the cloud, the logical control layer decomposes and allocates the complex tasks of multiple grids to hardware resources for processing through the dynamic execution of the scheduling algorithm; the core of the data access layer is virtualization technology. With the support of virtualization technology, the cloud platform can integrate the overall software and hardware resources of the network into a virtual resource pool to ensure the access and stable processing of upstream and downstream data; and the data application layer uses deep learning algorithm models to predict grid load, deeply explore the potential value of data and conduct a comprehensive evaluation of terminal grids, so as to realize refined management and scheduling decisions of grids. At the same time, the MySQL database on the cloud and the MongoDB database on the edge are used to jointly store multi-grid information, establish a typical library for classified grid construction, and complete the storage, backup and management of massive real-time data.

[0137] The overall application of the present application will be described below based on various implementation schemes.

[0138] More specifically, the present application relates to the technical field of distribution network grid information collection and management system, and in particular to a distribution network grid data fusion processing and comprehensive evaluation method based on adaptive membership and piecewise progressive weighted model.

[0139] like Figure 7 As shown, in general, some embodiments of the present application relate to a regional distribution network grid information collection management technology and a comprehensive evaluation method, which takes the distribution network grid construction as the research object, and from the perspective of the effectiveness of the new power system distribution network grid construction, adopts multi-source information fusion technology to realize the collection of grid power information, and uploads real-time power information to the cloud. Through the superiority of distributed computing on the cloud platform, massive power information is stored and calculated, and the cloud big data analysis model is used to deeply mine the fused data. Finally, the grid information is comprehensively evaluated in the visualization terminal, and the real-time information of the distribution network classification grid is mastered to achieve the purpose of maximizing the efficiency of the distribution network grid management.

[0140] Based on the architecture of the collection layer, communication layer, data layer and computing layer of cloud computing technology, a modular and integrated design approach is adopted to further improve the efficiency of power grid management.

[0141] In the comprehensive evaluation of visualization terminals, by introducing neural network algorithms, an adaptive membership calculation method based on neural networks and a piecewise progressive weighted comprehensive evaluation model are proposed. The index membership intervals are treated differently, which can adapt to the evaluation needs of different types of grids. The grid score no longer relies solely on traditional weighting, and can highlight the overall impact of the best or worst indicators, effectively solving the problems of complex membership calculation and low evaluation accuracy in the case of multi-grid information.

[0142] Combining the multi-dimensional data processing and analysis capabilities of the cloud platform database, a classified grid problem library and a reserve project library are established, which can accurately monitor and manage the regional distribution network grid information under the new power system in real time. The real-time update of grid information on the cloud establishes the interconnection between grid units, further improving the efficiency and benefits of distribution network grid management and construction.

[0143] It is worth noting that the comprehensive evaluation model for grid-based construction of distribution network based on adaptive membership and piecewise progressive weighted model provided in some embodiments of the present application is combined with Figure 8 ,The specific description of the model is as follows.

[0144] According to the reliability level of different power supply areas of the distribution network, a classification grid evaluation standard is formulated. Based on the subjective and objective optimal combination weights, an adaptive membership calculation method based on a neural network and a piecewise progressive weighted calculation comprehensive evaluation model are provided. Differentiated processing is performed on the index membership interval, which can adapt to the evaluation needs of different types of grids, so that the grid score no longer relies solely on traditional weighting, and can highlight the impact of the best or worst index on the overall situation, effectively solving the problem of complex membership calculation and low evaluation accuracy in the case of multi-grid information. Combined with the grid management platform, it can achieve precise control of multi-grid construction in the region and better solve the priority problem in grid construction.

[0145] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0146] Corresponding to the distribution network grid information management method described in the above embodiment, Figure 2 A structural block diagram of a distribution network grid information management device provided in an embodiment of the present application is shown. For ease of explanation, only the parts related to the embodiment of the present application are shown.

[0147] Reference Figure 2 , the device comprises:

[0148] An acquisition module 201 is used to acquire distribution network evaluation indicators in a specified area; the distribution network evaluation indicators include at least one of a load parameter, a self-healing parameter, an efficiency parameter, and an interactive parameter;

[0149] The network module 202 is used to use the distribution network evaluation index of the designated area as input, run the evaluation model, and obtain the evaluation result of the designated area; the evaluation model is a neural network model obtained by training based on samples and labels; the evaluation result is obtained by weighting the interval weights adjusted nonlinearly based on the membership parameter of the designated area;

[0150] The interval weight of the nonlinear adjustment is determined according to the gridding mode and / or gridding target of the distribution network.

[0151] In an optional implementation, the network module 202 includes:

[0152] A membership module, used for inputting the distribution network evaluation index of the designated area into the evaluation model to obtain the membership parameter of the designated area;

[0153] A weight module, used to calculate the weight value corresponding to the interval where the membership parameter of the specified area is located according to a preset weight model, and obtain the evaluation result of the specified area according to the weight value and the membership parameter of the specified area;

[0154] The weight model is determined according to the gridding mode and / or gridding target of the distribution network.

[0155] In another optional embodiment, the network module 202 includes:

[0156] An evaluation module is used to input the distribution network evaluation indicators and adjustment parameters of the specified area into the evaluation model to obtain the evaluation results of the specified area; wherein the adjustment parameters include at least one of the distribution network reliability classification of the specified area, the gridding method of the distribution network, and the gridding target of the distribution network.

[0157] Optionally, the evaluation model includes an activation function layer, and the activation function layer satisfies: calculating the continuous quantity membership based on a first activation function, and calculating the discrete quantity membership based on a second activation function; wherein the first activation function is different from the second activation function.

[0158] Optionally, the acquisition module 201 includes:

[0159] A database unit, used for real-time processing and caching of distribution network grid information based on a distributed database deployed at the data storage operation layer of the edge end; wherein the concurrent read and write parameter of the distributed database is higher than a preset first threshold, and the storage capacity parameter is higher than a preset second parameter;

[0160] A scheduling unit, configured to decompose and allocate multi-grid tasks to hardware resources for processing based on the dynamic execution of a scheduling algorithm of a logical control layer deployed in the cloud; wherein the multi-grid tasks are executed based on the gridding information of the distribution network;

[0161] The integration unit is used to integrate the overall network software and hardware resources into a virtual resource pool based on the data access layer deployed in the cloud, and obtain the distribution network evaluation index of the specified area through the virtual resource pool.

[0162] More specifically, the cloud is connected in communication with the grid information collection system:

[0163] The grid information acquisition system receives multi-channel sensor signals, obtains collected data after signal conditioning and data aggregation, and communicates the collected signals to the cloud platform and the host computer;

[0164] The multi-channel sensor signals are obtained via multiple sensors deployed in each grid.

[0165] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0166] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0167] The present application also provides a terminal device, such as Figure 3As shown, the terminal device 30 includes: at least one processor 301, a memory 302, and a computer program 303 stored in the memory and executable on the at least one processor, and the processor implements the steps in any of the above-mentioned method embodiments when executing the computer program.

[0168] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0169] An embodiment of the present application provides a computer program product. When the computer program product runs on a mobile terminal, the mobile terminal can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0170] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the camera / terminal device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), electric carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0171] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0172] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0173] In the embodiments provided in the present application, it should be understood that the disclosed devices / network equipment and methods can be implemented in other ways. For example, the device / network equipment embodiments described above are merely schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0174] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0175] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A distribution network grid information management method, characterized in that: include: Acquire distribution network evaluation indicators of a specified area; the distribution network evaluation indicators include at least one of a load parameter, a self-healing parameter, an efficiency parameter, and an interaction parameter; The distribution network evaluation index of the designated area is used as input to run the evaluation model to obtain the evaluation result of the designated area; the evaluation model is a neural network model obtained by training based on samples and labels; the evaluation result is obtained by weighting the interval weights adjusted nonlinearly based on the membership parameter of the designated area; The interval weight of the nonlinear adjustment is determined according to the gridding mode and / or gridding target of the distribution network.

2. The distribution network grid information management method according to claim 1, characterized in that: The step of taking the distribution network evaluation index of the designated area as input, running the evaluation model, and obtaining the evaluation result of the designated area comprises: Inputting the distribution network evaluation index of the designated area into the evaluation model to obtain the membership parameter of the designated area; According to a preset weight model, a weight value corresponding to the interval where the membership parameter of the designated area is located is calculated, and an evaluation result of the designated area is obtained according to the weight value and the membership parameter of the designated area; The weight model is determined according to the gridding mode and / or gridding target of the distribution network.

3. The distribution network grid information management method according to claim 1, characterized in that: The step of taking the distribution network evaluation index of the designated area as input, running the evaluation model, and obtaining the evaluation result of the designated area comprises: The distribution network evaluation index and adjustment parameters of the designated area are input into the evaluation model to obtain the evaluation result of the designated area; wherein the adjustment parameters include at least one of the distribution network reliability classification of the designated area, the gridding method of the distribution network, and the gridding target of the distribution network.

4. The distribution network grid information management method according to any one of claims 1 to 3, characterized in that: The evaluation model includes an activation function layer, which satisfies: calculating continuous quantity membership based on a first activation function, and calculating discrete quantity membership based on a second activation function; wherein the first activation function is different from the second activation function.

5. The distribution network grid information management method according to any one of claims 1 to 3, characterized in that: The step of obtaining the distribution network evaluation index of the specified area includes: A distributed database based on a data storage and operation layer deployed at the edge end processes and caches distribution network grid information in real time; wherein the concurrent read and write parameters of the distributed database are higher than a preset first threshold, and the storage capacity parameter is higher than a preset second parameter; Based on the dynamic execution of the scheduling algorithm of the logical control layer deployed in the cloud, the multi-grid tasks are decomposed and allocated to the hardware resources for processing; wherein the multi-grid tasks are executed based on the gridding information of the distribution network; Based on the data access layer deployed in the cloud, the overall network software and hardware resources are unified and integrated into a virtual resource pool, and the distribution network evaluation indicators of the specified area are obtained through the virtual resource pool.

6. The distribution network grid information management method according to claim 5, characterized in that: The cloud is connected to the grid information collection system: The grid information acquisition system receives multi-channel sensor signals, obtains collected data after signal conditioning and data aggregation, and communicates the collected signals to the cloud platform and the host computer; The multi-channel sensor signals are obtained via multiple sensors deployed in each grid.

7. A distribution network grid information management device, characterized in that: include: An acquisition module, used to acquire distribution network evaluation indicators of a specified area; the distribution network evaluation indicators include at least one of a load parameter, a self-healing parameter, an efficiency parameter, and an interactive parameter; A network module, used to run an evaluation model using the distribution network evaluation index of the designated area as input to obtain an evaluation result of the designated area; The evaluation model is a neural network model obtained through sample and label training; the evaluation result is obtained after weighting by nonlinearly adjusted interval weights based on the membership parameter of the designated area; The interval weight of the nonlinear adjustment is determined according to the gridding mode and / or gridding target of the distribution network.

8. A terminal 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, the method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

10. A computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images