Electric charge business online management method and system

Abstract

The invention belongs to the technical field of electricity charge management, and discloses an electricity charge business online management method and system. The method comprises the following steps: establishing a digital twin model based on a cloud data center, and constructing an electric charge business data analysis model and an electric energy scheduling model; based on the data transmission base station, a plurality of pieces of real-time electricity charge service data are collected and uploaded to the cloud data center; performing data analysis on a plurality of pieces of real-time electricity charge service data by using an electricity charge service data analysis model and a digital twin model; performing electric energy scheduling on the real-time data analysis result by using an electric energy scheduling model; and based on the cloud data center, a digital twin model is used to visualize a plurality of pieces of real-time electric charge business data, real-time data analysis results and real-time electric energy scheduling strategies of the electric charge business service subareas. According to the method, the problems of high cost investment, low efficiency, lack of information interaction, low practicability, poor intuition and poor data processing capability in the prior art are solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of electricity fee management, and in particular relates to an online management method and system for electricity fee business. Background Art

[0002] With the popularization of electric energy and the development of the scale of power systems, more and more power users are joining and generating exponential electricity bill business data. For different types of power users, such as residential power users and corporate power users, electricity bill business management has become a key development direction of the power industry. However, the existing electricity bill business management technology has the following defects, which restricts the development of the industry:

[0003] 1) Traditional electricity fee business management usually relies on manual collection and statistics, which is costly and inefficient and cannot be applied to large-scale electricity fee business management;

[0004] 2) Traditional electricity bill business data has information island effect in management and lacks information interaction, which leads to the waste of value of electricity bill business data and low practicality;

[0005] 3) Traditional electricity bill business management lacks real-time monitoring, data visualization and data analysis functions, has poor data processing capabilities, and cannot intuitively display detailed information on electricity bill business data, which is not conducive to managers to promptly discover abnormal electricity bill business and solve power scheduling problems. Summary of the invention

[0006] In order to solve the problems of high cost investment, low efficiency, lack of information interaction, low practicality, poor intuitiveness and poor data processing capability in the prior art, the present invention aims to provide an online management method and system for electricity bill business.

[0007] The technical solution adopted by the present invention is:

[0008] An online management method for electricity fee business includes the following steps:

[0009] Based on the cloud data center, a digital twin model is established, and an electricity fee business data analysis model and an electricity dispatch model are constructed;

[0010] Based on the data transmission base station, real-time electricity fee business data sent by several data collection devices in the electricity fee business service zone are collected, and several real-time electricity fee business data are encrypted and uploaded to the cloud data center;

[0011] Based on the cloud data center, the electricity fee business data analysis model and the digital twin model are used to analyze the real-time electricity fee business data of the electricity fee business service partition to obtain real-time data analysis results;

[0012] Based on the cloud data center, the power dispatch model is used to dispatch power based on the real-time data analysis results of the electricity fee business service partition to obtain a real-time power dispatch strategy;

[0013] Based on the cloud data center, the digital twin model is used to visualize several real-time electricity bill business data, real-time data analysis results, and real-time power dispatch strategies of the electricity bill business service partition.

[0014] Furthermore, based on the cloud data center, a digital twin model is established, and an electricity fee business data analysis model and a real-time power dispatch strategy are constructed, including the following steps:

[0015] Based on the cloud data center, digital twin model is constructed using digital twin technology according to the basic information of power network and power users;

[0016] Based on several historical electricity fee business data of different electricity fee business service zones, a deep learning algorithm is used to build an electricity fee business data analysis model and generate several historical data analysis results;

[0017] Based on the analysis results of several historical data, a power dispatch model is constructed using a reinforcement learning algorithm.

[0018] Furthermore, the electricity bill business data analysis model is built based on the GNN-LSTM-Attention-MLP algorithm;

[0019] The electric energy scheduling model is built based on the DQN algorithm, and the electric energy scheduling model is provided with an intelligent agent, an experience replay pool and a deep Q network.

[0020] Furthermore, based on the cloud data center, according to the basic information of the power network and the basic information of the power users, the digital twin technology is used to build a digital twin model, including the following steps:

[0021] Based on the cloud data center, a static model of the power network topology is constructed according to the basic information of the power network, and a dynamic model of the power user's electricity fee business is constructed according to the basic information of the power user;

[0022] Add the dynamic models of the electricity user electricity bill business of several electricity users to the corresponding positions of the static model of the power network topology to obtain a comprehensive model of the digital world;

[0023] The data stream collected by the data acquisition device in the physical world is connected to the corresponding dynamic model of the power user's electricity bill business, and a digital twin database and visualization interface are set up to obtain a digital twin model.

[0024] Furthermore, according to a number of historical electricity fee business data of different electricity fee business service partitions, a deep learning algorithm is used to construct an electricity fee business data analysis model, and generate a number of historical data analysis results, including the following steps:

[0025] Collecting a number of historical electricity fee business data of different electricity fee business service partitions, and performing data preprocessing to obtain a number of preprocessed historical electricity fee business data of different electricity fee business service partitions;

[0026] Divide a number of pre-processed historical electricity fee business data of different electricity fee business service partitions into a model training set and a model test set;

[0027] Use deep learning algorithms to build an initial electricity fee business data analysis model, and use digital twin models and model training sets to optimize and train the initial electricity fee business data analysis model to obtain an optimized electricity fee business data analysis model and generate several historical data analysis results;

[0028] Use the digital twin model and model test set to test the optimized electricity fee business data analysis model. If the model accuracy is greater than the accuracy threshold, the final electricity fee business data analysis model is output; otherwise, continue the optimization training;

[0029] The final electricity charge business data analysis model is deployed to the digital twin database of the digital twin model, the data stream of the power user's electricity charge business dynamic model is connected to the input end of the final electricity charge business data analysis model, and the output end of the final electricity charge business data analysis model is connected to the visualization interface.

[0030] Furthermore, based on the analysis results of several historical data, a reinforcement learning algorithm is used to construct an energy dispatch model, which includes the following steps:

[0031] Generate problems based on the power dispatch strategy, set up the simulation environment for the reinforcement learning algorithm, and build the intelligent agent and experience replay pool;

[0032] Define the state space of the reinforcement learning algorithm based on the states involved in the historical data analysis results, and construct the input layer of the deep Q network based on the data structure of the historical data analysis results;

[0033] According to the actions involved in the power dispatch strategy, the action space of the reinforcement learning algorithm is defined, and the output layer of the deep Q network is constructed according to the data structure of the power dispatch strategy;

[0034] According to the input layer and the output layer, several hidden layers are set to construct the corresponding deep Q network, connecting the input layer to the state space and the output layer to the action space;

[0035] Define the reward function of the reinforcement learning algorithm based on the impact of each action in the action space;

[0036] Based on the state space, action space and reward function, the deep Q network and intelligent agent are optimized and trained using some historical data analysis results to build an energy dispatch model and generate some historical energy dispatch experience.

[0037] Several historical power dispatching experiences are stored in the experience replay pool, the power dispatching model is deployed to the digital twin database of the digital twin model, the output end of the electricity bill business data analysis model is connected to the input end of the power dispatching model, and the output end of the power dispatching model is connected to the visualization interface.

[0038] Furthermore, based on the cloud data center, using the electricity fee business data analysis model and the digital twin model, data analysis is performed on a number of real-time electricity fee business data of the electricity fee business service partition to obtain real-time data analysis results, including the following steps:

[0039] Based on the cloud data center, a digital twin database is used to receive several real-time electricity fee business data of the same electricity fee business service partition;

[0040] Preprocessing a number of real-time electricity fee business data to obtain a real-time data matrix, and inputting the real-time data matrix into an electricity fee business data analysis model;

[0041] Use the electricity bill business data analysis model to extract the real-time graph features of the digital twin model and the real-time data features of the real-time data matrix;

[0042] According to the preset attention weight value, the real-time image features and the real-time data features are spliced ​​to obtain a real-time spliced ​​feature sequence;

[0043] According to the real-time splicing feature sequence, data analysis is performed to obtain the real-time data analysis results of the corresponding electricity fee business service partition.

[0044] Furthermore, based on the cloud data center, using the power dispatch model, power dispatch is performed on the real-time data analysis results of the electricity fee business service partition to obtain a real-time power dispatch strategy, including the following steps:

[0045] Based on the cloud data center, the state space of the power dispatch model is updated according to the real-time data analysis results to obtain an updated state space;

[0046] Extracting a number of historical power dispatching experiences from the experience playback pool of the power dispatching model, and updating the action space of the power dispatching model according to the number of historical power dispatching experiences to obtain an updated action space;

[0047] Inputting the updated state space into the input layer of the deep Q network of the electric energy scheduling model, and connecting the updated action space to the output layer of the deep Q network to obtain an updated deep Q network;

[0048] Based on the updated deep Q network, use the agent to control the updated deep Q network and output the Q values ​​of possible actions in the updated action space;

[0049] According to the preset reward function, the Q value is iteratively updated to obtain an updated Q value until the number of iterations reaches the threshold;

[0050] According to the greedy strategy, the possible action with the highest updated Q value in each iteration is used as the execution action of the corresponding state;

[0051] The execution actions corresponding to all states in the updated state space are integrated to obtain the real-time power dispatch strategy for the electricity bill service partition.

[0052] Furthermore, based on the cloud data center, a digital twin model is used to visualize several real-time electricity bill service data, real-time data analysis results, and real-time power dispatch strategies of the electricity bill service partition, including the following steps:

[0053] Based on the cloud data center, extract some historical electricity fee business data of the electricity fee business service partition received by the digital twin database, and add some historical electricity fee business data to the corresponding power user electricity fee business dynamic model;

[0054] Extract the real-time data analysis results generated by the electricity fee business data analysis model. If there are any abnormalities in the real-time data analysis results, generate corresponding real-time electricity fee business abnormality signals, and use a visualization interface to visualize the real-time electricity fee business abnormality signals and real-time data analysis results;

[0055] The real-time power dispatching strategy generated by the power dispatching model is extracted and visualized using a visualization interface.

[0056] An online management system for electricity fee business, used to implement an online management method for electricity fee business, characterized in that: the system includes a cloud data center, a plurality of data transmission base stations and a plurality of data acquisition devices, the cloud data center is respectively connected to the data transmission base stations of a plurality of electricity fee business service zones, and each data transmission base station is respectively connected to the data acquisition devices in the corresponding electricity fee business service zone;

[0057] The cloud data center is provided with an initialization unit, a data analysis unit, a power dispatching unit and a visualization unit which are connected in sequence.

[0058] The beneficial effects of the present invention are:

[0059] The present invention discloses an online management method and system for electricity charge business, which adopts a data acquisition device to perform automatic data acquisition, a data transmission base station to perform partition management and data transmission, and a cloud data center to perform online management, thereby avoiding the use of manual methods, reducing cost investment, and improving management efficiency, and is suitable for large-scale electricity charge business management; unified data management is performed in the cloud data center, thereby avoiding the existence of information island effect, strengthening information interaction, and improving the value utilization and practicality of data; an electricity charge business data analysis model is used to perform real-time monitoring, data processing and intelligent analysis on a number of real-time electricity charge business data of the electricity charge business service partition, which can efficiently and accurately discover abnormal electricity charge business, locate and predict it, improve data processing capabilities, improve the degree of intelligence, and can use an electric energy dispatching model to dispatch electric energy according to the real-time data analysis results, and guide the solution of electric energy dispatching problems; the digital twin model can intuitively display real-time data, analysis results and electric energy dispatching strategies, thereby improving convenience and reducing management difficulty.

[0060] Other beneficial effects of the present invention will be further described in the specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a flowchart of the online management method of electricity fee business in the present invention.

[0062] Figure 2 It is a structural block diagram of the online management system for electricity fee business in the present invention. DETAILED DESCRIPTION

[0063] The present invention will be further explained below in conjunction with the accompanying drawings and specific embodiments.

[0064] Embodiment 1:

[0065] like Figure 1 As shown, this embodiment provides an online management method for electricity fee services, including the following steps:

[0066] S1: Based on the cloud data center, a digital twin model is established, and an electricity fee business data analysis model and an electric energy dispatch model are constructed, including the following steps:

[0067] S1-1: Based on the cloud data center, according to the basic information of the power network and the basic information of power users, the digital twin technology is used to build a digital twin model, including the following steps:

[0068] S1-1-1: Based on the cloud data center and according to the basic information of the power network, a static model of the power network topology is constructed, and based on the basic information of the power users, a dynamic model of the power user electricity fee business is constructed;

[0069] S1-1-2: Add the dynamic models of the electricity user electricity fee business of several electricity users to the corresponding positions of the static model of the power network topology to obtain a comprehensive model of the digital world;

[0070] S1-1-3: Connect the data stream collected by the data collection device in the physical world to the corresponding dynamic model of the power user's electricity fee business, and set up a digital twin database and a visualization interface to obtain a digital twin model;

[0071] S1-2: Based on several historical electricity fee business data of different electricity fee business service partitions, a deep learning algorithm is used to build an electricity fee business data analysis model and generate several historical data analysis results, including the following steps:

[0072] S1-2-1: collecting a number of historical electricity fee business data of different electricity fee business service partitions, and performing data preprocessing to obtain a number of preprocessed historical electricity fee business data of different electricity fee business service partitions;

[0073] S1-2-2: Divide a number of pre-processed historical electricity fee business data of different electricity fee business service partitions into a model training set and a model test set;

[0074] S1-2-3: Use deep learning algorithms to build an initial electricity fee business data analysis model, and use the digital twin model and model training set to optimize the initial electricity fee business data analysis model to obtain an optimized electricity fee business data analysis model and generate several historical data analysis results;

[0075] The electricity fee business data analysis model is constructed based on the GNN-LSTM-Attention-MLP algorithm, and the electricity fee business data analysis model includes a graph feature extraction module constructed based on the Graph Neural Networks (GNN) algorithm, a data feature extraction module constructed based on the Long Short-Term Memory (LSTM) algorithm, an attention weight module constructed based on the Attention mechanism, and a data analysis module constructed based on the Multilayer Perceptron (MLP) algorithm. The graph feature extraction module and the data feature extraction module are both connected to the attention weight module, and the attention weight module is connected to the data analysis module;

[0076] The graph feature extraction module is used to extract the graph features of the digital twin model, wherein the graph features include the basic user information features of the dynamic model of the electricity user's electricity fee business as a node in the digital twin model as the node features, and the power network topology information features of the static model of the power network topology as the edge features between the nodes, and finally constitute the features of the digital twin model;

[0077] The data feature extraction module is used to extract the data features contained in the real-time electricity fee business data, and to characterize the electricity fee information and other electricity consumption behavior information of the power user;

[0078] The attention weight module is set with attention weight values, which performs weighted splicing of graph features and data features, thus improving the ability to characterize abnormal situations in electricity bill services;

[0079] The data analysis module is a fully connected network. It analyzes the splicing features, predicts data labels, and realizes data analysis. The data analysis results include not only the abnormal electricity bill business prediction results of power users, but also the regional power consumption prediction results of the electricity bill business service partition, power utilization prediction results, load balancing prediction results, future power consumption prediction results and other information. It not only realizes the location of abnormal power users, but also realizes the prediction of power demand for power network.

[0080] S1-2-4: Use the digital twin model and the model test set to test the optimized electricity fee business data analysis model. If the model accuracy is greater than the accuracy threshold, the final electricity fee business data analysis model is output; otherwise, the optimization training continues;

[0081] S1-2-5: deploy the final electricity fee business data analysis model to the digital twin database of the digital twin model, connect the data stream of the power user electricity fee business dynamic model to the input end of the final electricity fee business data analysis model, and connect the output end of the final electricity fee business data analysis model to the visualization interface;

[0082] S1-3: Based on the analysis results of several historical data, a reinforcement learning algorithm is used to build an energy dispatch model, including the following steps:

[0083] S1-3-1: Generate problems based on the power dispatch strategy, set up the simulation environment of the reinforcement learning algorithm, and build the intelligent agent and experience replay pool;

[0084] The power dispatch model is constructed based on the Deep Q Network (DQN) algorithm, and the power dispatch model is provided with an intelligent agent, an experience replay pool and a Deep Q Network;

[0085] S1-3-2: Define the state space of the reinforcement learning algorithm based on the states involved in the historical data analysis results, and construct the input layer of the deep Q network based on the data structure of the historical data analysis results;

[0086] The states in the state space mainly include abnormal electricity bill services, abnormal power users, abnormal locations, regional power consumption forecasts at the next moment, power utilization forecasts, load balancing forecasts, and future power consumption forecasts in the electricity bill service partition;

[0087] S1-3-3: Define the action space of the reinforcement learning algorithm according to the actions involved in the power scheduling strategy, and construct the output layer of the deep Q network according to the data structure of the power scheduling strategy;

[0088] The actions in the action space mainly include the power generation decision of the power generation equipment at the next moment, the charging and discharging decision of the energy storage system, the adjustment decision of the transmission equipment, and the power allocation decision of each electricity service zone;

[0089] S1-3-4: According to the input layer and the output layer, set several hidden layers, build the corresponding deep Q network, connect the input layer to the state space, and connect the output layer to the action space;

[0090] S1-3-5: Define the reward function of the reinforcement learning algorithm based on the impact of each action in the action space;

[0091] S1-3-6: Based on the state space, action space and reward function, use some historical data analysis results to optimize the training of the deep Q network and the agent, build an energy dispatch model, and generate some historical energy dispatch experience;

[0092] S1-3-7: Store some historical power dispatching experiences in the experience playback pool, deploy the power dispatching model to the digital twin database of the digital twin model, connect the output end of the electricity fee business data analysis model to the input end of the power dispatching model, and connect the output end of the power dispatching model to the visualization interface;

[0093] Also includes:

[0094] S1-4: Based on a trusted institution, key generation and digital identity authentication are performed for all data transmission base stations, including the following steps:

[0095] S1-4-1: Initialize the key based on the trusted institution to obtain the public parameters, master key and initial key;

[0096] The formula is:

[0097]

[0098] Where GP is the public parameter; MSK is the master key; PK is the initial key; a is the integer domain Z p Random numbers; H1, H2, H3, H4, H5, H6, H u All are target hash functions; g, g1, g a are all random numbers that are generators of the cyclic group G; e(g,g) a is a bilinear map of random numbers g;

[0099] S1-4-2: Use an asymmetric encryption algorithm to generate a public-private key pair for the data transmission base station, store the private key in the public-private key pair in the corresponding data transmission base station, and store the public key in the public-private key pair in the cloud data center;

[0100] The formula is:

[0101]

[0102] In the formula, SK u is the private key of the data transmission base station u; b is the integer domain Z p Random number; L u , K' u is the private key parameter of the data transmission base station u; H3 is the target hash function of the public parameter GP; u is the data transmission base station indicator; MSK is the master key; PK is the initial key; PK u g is the public key of the data transmission base station u; b , g a , g ab is the random number of the generator of the cyclic group G; V u To transmit the attribute information of the data base station u;

[0103] S1-4-3: Use digital identity authentication technology to generate signature information for each data transmission base station, and store the signature information in the corresponding data transmission base station;

[0104] The formula is:

[0105]

[0106] Where k′ is a random number; K u KID is the registration parameter of the data transmission base station u; u KID is the registration ID of the data transmission base station u; u and the corresponding K u The signature information {K u ,KID u}; H1 is the target hash function; ID u The device ID of the data transmission base station u; is the prime order; P is the base point of the prime domain; mod(*) is the remainder function;

[0107] S2: Based on the data transmission base station, real-time electricity fee service data sent by several data collection devices in the electricity fee service zone are collected, and several real-time electricity fee service data are encrypted and uploaded to the cloud data center, including the following steps:

[0108] S2-1: Based on the data transmission base station, receive the real-time electricity fee service data sent by several data collection devices in the electricity fee service zone, and obtain the real-time electricity fee service data set;

[0109] S2-2: Encrypt the real-time electricity fee service data set according to the private key to obtain the encrypted real-time electricity fee service data set;

[0110] The formula is:

[0111] M u1 =E(SK u1 ,m u1 )

[0112] Where M u1 is the encrypted real-time electricity fee business data set; E(*) is the asymmetric encryption function; m u1 It is a real-time electricity bill business data set; SK u1 is the private key of the data transmission base station u1; u1 is the data transmission base station indicator;

[0113] S2-3: According to the signature information, the encrypted real-time electricity fee service data set is signed to obtain the corresponding real-time signature data, and the encrypted real-time electricity fee service data set and the real-time signature data are uploaded to the cloud data center;

[0114] The formula is:

[0115]

[0116] In the formula, r' is a random number; is the prime order; P is the base point of the prime domain; H2 is the target hash function; K u1 For the signature information {K u1 ,KID u1} Registration parameters of data transmission base station u1; KID u1 The ID is the registration ID of the data transmission base station u1 in the signature information; u1 is the entity ID of the data transmission base station u1; the real-time signature data is {ID u1 ,M u1 ,γ'={K u1 ,R u1 ,B u1}}; R u1 ,B u1 ,γ' are the signature parameters of data transmission base station u1;

[0117] S3: Based on the cloud data center, using the electricity fee business data analysis model and the digital twin model, data analysis is performed on several real-time electricity fee business data of the electricity fee business service partition to obtain real-time data analysis results, including the following steps:

[0118] S3-1: Based on the cloud data center, using the digital twin database, receiving several real-time electricity fee business data of the same electricity fee business service partition, including the following steps:

[0119] S3-1: Based on the cloud data center, the real-time signature data uploaded by the data transmission base station is verified;

[0120] The formula is:

[0121] β u1 B u1 P=β u1 H2(R u1 ,ID u1 ,K u1 )R u1 +β u1 K u1 +β u1 H1(ID u1 ,K u1 )PK u1

[0122] If the left formula is equal to the right formula, the signature verification is successful;

[0123] S3-2: After the signature verification is passed, the encrypted real-time electricity fee service data set is decrypted using the corresponding public key to obtain a number of decrypted real-time electricity fee service data;

[0124] The formula is:

[0125] m' u1 =E - (PK u1 ,M u1 )

[0126] In the formula, m' u1 The decrypted real-time electricity fee business data set includes several decrypted real-time electricity fee business data; E - (*) is an asymmetric decryption function; M u1 It is the encrypted real-time electricity fee business data set; PK u1 The public key of the data transmission base station u1;

[0127] S3-2: pre-processing the decrypted real-time electricity fee business data to obtain a real-time data matrix, and inputting the real-time data matrix into the electricity fee business data analysis model;

[0128] Preprocessing includes data cleaning, normalization, and matrix conversion. While eliminating duplicate and erroneous data, it also performs standardized conversion, eliminates magnitude differences and noise, and converts the data into a matrix format that can be recognized by subsequent models.

[0129] S3-3: Use the electricity bill business data analysis model, use the graph feature extraction module to extract the real-time graph features of the digital twin model, and use the data feature extraction module to extract the real-time data features of the real-time data matrix;

[0130] S3-4: According to the attention weight value preset by the attention weight module, the real-time image feature and the real-time data feature are spliced ​​to obtain a real-time spliced ​​feature sequence;

[0131] S3-5: Use the data analysis module to perform data analysis based on the real-time splicing feature sequence to obtain the real-time data analysis results of the corresponding electricity fee business service partition;

[0132] S4: Based on the cloud data center, using the power dispatch model, the real-time data analysis results of the electricity fee business service partition are used to perform power dispatch, and a real-time power dispatch strategy is obtained, including the following steps:

[0133] S4-1: Based on the cloud data center, according to the real-time data analysis results, the state space of the power dispatch model is updated to obtain the updated state space S'=[s'1,...,s' i" ,...,s' I ], where s' i" is the updated i'th state value, i' is the state indicator, and I' is the total number of state space dimensions;

[0134] S4-2: extract some historical power dispatching experience from the experience playback pool of the power dispatching model, and update the action space of the power dispatching model according to some historical power dispatching experience, and obtain the updated action space A'=[a'1,...,a' j" ,...,a' I ], where a' j" is the updated j-th "action value", j" is the action indicator, and I is the total number of action space dimensions;

[0135] S4-3: Input the updated state space into the input layer of the deep Q network of the electric energy scheduling model, and connect the updated action space to the output layer of the deep Q network to obtain an updated deep Q network;

[0136] S4-4: Based on the updated deep Q network, use the agent to control the updated deep Q network and output the Q values ​​of possible actions in the updated action space;

[0137] S4-5: Iteratively update the Q value according to the preset reward function to obtain an updated Q value until the number of iterations reaches the threshold;

[0138] The formula is:

[0139] Q(s' p' ,a'p' )=(1-α")·Q(s p' ,a p' )+α"·(R(s p' ,a p' ,s' p' )+γ·Q max (s p' ,a p' ))

[0140] In the formula, Q(s' p' ,a' p' ) is the updated state value s' p' and the updated action value a' p' The corresponding updated Q value; Q(s p' ,a p' ) is the state value s p' and action value a p' The corresponding predicted Q value; α" is the learning rate; Q max (s p' ,a p' ) is the highest predicted Q value; p' is the comprehensive indicator; γ is the update parameter;

[0141] S4-6: According to the greedy strategy, the possible action with the highest updated Q value in each iteration is used as the execution action of the corresponding state;

[0142] S4-7: Integrate the execution actions corresponding to all states in the updated state space to obtain the real-time power dispatch strategy of the electricity fee business service partition;

[0143] S5: Based on the cloud data center, the digital twin model is used to visualize several real-time electricity bill business data, real-time data analysis results, and real-time power dispatch strategies of the electricity bill business service partition, including the following steps:

[0144] S5-1: Based on the cloud data center, extract a number of historical electricity fee business data of the electricity fee business service partition received by the digital twin database, and add the several historical electricity fee business data to the corresponding power user electricity fee business dynamic model;

[0145] S5-2: extracting the real-time data analysis results generated by the electricity fee business data analysis model. If the real-time data analysis results are abnormal, generating a corresponding real-time electricity fee business abnormal signal, and using a visualization interface to visualize the real-time electricity fee business abnormal signal and the real-time data analysis results;

[0146] S5-3: Extract the real-time power scheduling strategy generated by the power scheduling model, and visualize the real-time power scheduling strategy using a visualization interface.

[0147] Embodiment 2:

[0148] like Figure 2 As shown, this embodiment provides an online management system for electricity fee business, which is used to implement an online management method for electricity fee business, and is characterized in that: the system includes a cloud data center, a plurality of data transmission base stations and a plurality of data acquisition devices, the cloud data center is respectively connected to the data transmission base stations of a plurality of electricity fee business service partitions, and each data transmission base station is respectively connected to the data acquisition devices in the corresponding electricity fee business service partitions;

[0149] A data collection device is used to collect real-time electricity bill service data of power users in the electricity bill service area and send it to the corresponding data transmission base station;

[0150] A data transmission base station is used to collect real-time electricity fee business data sent by a number of data collection devices in the electricity fee business service zone, and encrypt and upload the real-time electricity fee business data to the cloud data center;

[0151] The cloud data center is provided with an initialization unit, a data analysis unit, a power dispatching unit and a visualization unit which are connected in sequence;

[0152] Initialization unit, used to establish a digital twin model, and build an electricity fee business data analysis model and an electric energy scheduling model;

[0153] A data analysis unit, used to use an electricity fee business data analysis model and a digital twin model to perform data analysis on a number of real-time electricity fee business data of an electricity fee business service partition, and obtain real-time data analysis results;

[0154] An electric energy dispatching unit, used to use an electric energy dispatching model to perform electric energy dispatching on the real-time data analysis results of the electricity fee business service partition to obtain a real-time electric energy dispatching strategy;

[0155] The visualization unit is used to visualize a number of real-time electricity charge business data, real-time data analysis results, and real-time electric energy dispatching strategies of the electricity charge business service partition using a digital twin model.

[0156] The present invention discloses an online management method and system for electricity charge business, which adopts a data acquisition device to perform automatic data acquisition, a data transmission base station to perform partition management and data transmission, and a cloud data center to perform online management, thereby avoiding the use of manual methods, reducing cost investment, and improving management efficiency, and is suitable for large-scale electricity charge business management; unified data management is performed in the cloud data center, thereby avoiding the existence of information island effect, strengthening information interaction, and improving the value utilization and practicality of data; an electricity charge business data analysis model is used to perform real-time monitoring, data processing and intelligent analysis on a number of real-time electricity charge business data of the electricity charge business service partition, which can efficiently and accurately discover abnormal electricity charge business, locate and predict it, improve data processing capabilities, improve the degree of intelligence, and can use an electric energy dispatching model to dispatch electric energy according to the real-time data analysis results, and guide the solution of electric energy dispatching problems; the digital twin model can intuitively display real-time data, analysis results and electric energy dispatching strategies, thereby improving convenience and reducing management difficulty.

[0157] The present invention is not limited to the above optional implementations, and anyone can derive other various forms of products under the enlightenment of the present invention. The above specific implementations should not be understood as limiting the scope of protection of the present invention. The scope of protection of the present invention should be based on the definition in the claims, and the description can be used to interpret the claims.

Claims

1. An online management method for electricity fee business, characterized by: The steps include: Based on the cloud data center, a digital twin model is established, and an electricity fee business data analysis model and an electricity dispatch model are constructed; Based on the data transmission base station, real-time electricity fee business data sent by several data collection devices in the electricity fee business service zone are collected, and several real-time electricity fee business data are encrypted and uploaded to the cloud data center; Based on the cloud data center, the electricity fee business data analysis model and the digital twin model are used to analyze the real-time electricity fee business data of the electricity fee business service partition to obtain real-time data analysis results; Based on the cloud data center, the power dispatch model is used to dispatch power based on the real-time data analysis results of the electricity fee business service partition to obtain a real-time power dispatch strategy; Based on the cloud data center, the digital twin model is used to visualize several real-time electricity bill business data, real-time data analysis results, and real-time power dispatch strategies of the electricity bill business service partition.

2. The method for online management of electricity bill services according to claim 1, characterized in that: Based on the cloud data center, a digital twin model is established, and an electricity fee business data analysis model and a real-time power dispatch strategy are constructed, including the following steps: Based on the cloud data center, digital twin model is constructed using digital twin technology according to the basic information of power network and power users; Based on several historical electricity fee business data of different electricity fee business service zones, a deep learning algorithm is used to build an electricity fee business data analysis model and generate several historical data analysis results; Based on the analysis results of several historical data, a power dispatch model is constructed using a reinforcement learning algorithm.

3. The method for online management of electricity fee services according to claim 2, characterized in that: The electricity fee business data analysis model is constructed based on the GNN-LSTM-Attention-MLP algorithm; The electric energy dispatching model is constructed based on the DQN algorithm, and the electric energy dispatching model is provided with an intelligent agent, an experience replay pool and a deep Q network.

4. The method for online management of electricity fee services according to claim 3, characterized in that: Based on the cloud data center, according to the basic information of the power network and the basic information of power users, the digital twin technology is used to build a digital twin model, including the following steps: Based on the cloud data center, a static model of the power network topology is constructed according to the basic information of the power network, and a dynamic model of the power user's electricity fee business is constructed according to the basic information of the power user; Add the dynamic models of the electricity user electricity bill business of several electricity users to the corresponding positions of the static model of the power network topology to obtain a comprehensive model of the digital world; The data stream collected by the data acquisition device in the physical world is connected to the corresponding dynamic model of the power user's electricity bill business, and a digital twin database and visualization interface are set up to obtain a digital twin model.

5. The method for online management of electricity fee services according to claim 4, characterized in that: Based on several historical electricity fee business data of different electricity fee business service partitions, a deep learning algorithm is used to build an electricity fee business data analysis model and generate several historical data analysis results, including the following steps: Collecting a number of historical electricity fee business data of different electricity fee business service partitions, and performing data preprocessing to obtain a number of preprocessed historical electricity fee business data of different electricity fee business service partitions; Divide a number of pre-processed historical electricity fee business data of different electricity fee business service partitions into a model training set and a model test set; Use deep learning algorithms to build an initial electricity fee business data analysis model, and use digital twin models and model training sets to optimize and train the initial electricity fee business data analysis model to obtain an optimized electricity fee business data analysis model and generate several historical data analysis results; Use the digital twin model and model test set to test the optimized electricity fee business data analysis model. If the model accuracy is greater than the accuracy threshold, the final electricity fee business data analysis model is output; otherwise, continue the optimization training; The final electricity charge business data analysis model is deployed to the digital twin database of the digital twin model, the data stream of the power user's electricity charge business dynamic model is connected to the input end of the final electricity charge business data analysis model, and the output end of the final electricity charge business data analysis model is connected to the visualization interface.

6. The method for online management of electricity fee services according to claim 4, characterized in that: Based on the analysis results of several historical data, a power dispatch model is constructed using a reinforcement learning algorithm, including the following steps: Generate problems based on the power dispatch strategy, set up the simulation environment for the reinforcement learning algorithm, and build the intelligent agent and experience replay pool; Define the state space of the reinforcement learning algorithm based on the states involved in the historical data analysis results, and construct the input layer of the deep Q network based on the data structure of the historical data analysis results; According to the actions involved in the power dispatch strategy, the action space of the reinforcement learning algorithm is defined, and the output layer of the deep Q network is constructed according to the data structure of the power dispatch strategy; According to the input layer and the output layer, several hidden layers are set to construct the corresponding deep Q network, connecting the input layer to the state space and the output layer to the action space; Define the reward function of the reinforcement learning algorithm based on the impact of each action in the action space; Based on the state space, action space and reward function, the deep Q network and intelligent agent are optimized and trained using some historical data analysis results to build an energy dispatch model and generate some historical energy dispatch experience. Several historical power dispatching experiences are stored in the experience replay pool, the power dispatching model is deployed to the digital twin database of the digital twin model, the output end of the electricity bill business data analysis model is connected to the input end of the power dispatching model, and the output end of the power dispatching model is connected to the visualization interface.

7. The method for online management of electricity fee services according to claim 5, characterized in that: Based on the cloud data center, the electricity fee business data analysis model and the digital twin model are used to analyze some real-time electricity fee business data of the electricity fee business service partition to obtain real-time data analysis results, including the following steps: Based on the cloud data center, a digital twin database is used to receive several real-time electricity fee business data of the same electricity fee business service partition; Preprocessing a number of real-time electricity fee business data to obtain a real-time data matrix, and inputting the real-time data matrix into an electricity fee business data analysis model; Use the electricity bill business data analysis model to extract the real-time graph features of the digital twin model and the real-time data features of the real-time data matrix; According to the preset attention weight value, the real-time image features and the real-time data features are spliced ​​to obtain a real-time spliced ​​feature sequence; According to the real-time splicing feature sequence, data analysis is performed to obtain the real-time data analysis results of the corresponding electricity fee business service partition.

8. The method for online management of electricity fee services according to claim 6, characterized in that: Based on the cloud data center, the power dispatch model is used to dispatch power based on the real-time data analysis results of the electricity fee business service partition to obtain a real-time power dispatch strategy, including the following steps: Based on the cloud data center, the state space of the power dispatch model is updated according to the real-time data analysis results to obtain an updated state space; Extracting a number of historical power dispatching experiences from the experience playback pool of the power dispatching model, and updating the action space of the power dispatching model according to the number of historical power dispatching experiences to obtain an updated action space; Inputting the updated state space into the input layer of the deep Q network of the electric energy scheduling model, and connecting the updated action space to the output layer of the deep Q network to obtain an updated deep Q network; Based on the updated deep Q network, use the agent to control the updated deep Q network and output the Q values ​​of possible actions in the updated action space; According to the preset reward function, the Q value is iteratively updated to obtain an updated Q value until the number of iterations reaches the threshold; According to the greedy strategy, the possible action with the highest updated Q value in each iteration is used as the execution action of the corresponding state; The execution actions corresponding to all states in the updated state space are integrated to obtain the real-time power dispatch strategy for the electricity bill service partition.

9. The method for online management of electricity fee services according to claim 4, characterized in that: Based on the cloud data center, the digital twin model is used to visualize several real-time electricity bill business data, real-time data analysis results, and real-time power dispatch strategies of the electricity bill business service partition, including the following steps: Based on the cloud data center, extract some historical electricity fee business data of the electricity fee business service partition received by the digital twin database, and add some historical electricity fee business data to the corresponding power user electricity fee business dynamic model; Extract the real-time data analysis results generated by the electricity fee business data analysis model. If there are any abnormalities in the real-time data analysis results, generate corresponding real-time electricity fee business abnormality signals, and use a visualization interface to visualize the real-time electricity fee business abnormality signals and real-time data analysis results; The real-time power dispatching strategy generated by the power dispatching model is extracted and visualized using a visualization interface.

10. An online management system for electricity fee services, used to implement the online management method for electricity fee services as claimed in any one of claims 1 to 9, characterized in that: The system includes a cloud data center, several data transmission base stations and several data collection devices. The cloud data center is respectively connected to the data transmission base stations of several electricity fee business service zones, and each of the data transmission base stations is respectively connected to the data collection devices in the corresponding electricity fee business service zones. The cloud data center is provided with an initialization unit, a data analysis unit, a power dispatching unit and a visualization unit which are connected in sequence.

Images