Power customer service risk early warning and processing method and system

Abstract

The invention discloses an electric power customer service risk early warning and processing method and system, and relates to the field of intelligent power grid and customer relationship management, and the method comprises the steps: determining the range of the influence of power failure on electric power customers according to power failure information; predicting customer appeal types possibly caused by the power failure in combination with historical data on the basis of the range of the influence of the power failure on the power customers; setting a power customer service automatic receiving and processing device for processing based on the prediction of customer demand types; according to the invention, preparation of related power failure services can be made before power failure and incoming call of the user, differentiated services can be carried out according to different types of users, incoming call experience of the user to a real problem point to be solved is guaranteed, and customer service satisfaction is improved; the service pressure of a power grid company can be relieved, and the situation that traditional customer service staff only accepts work order service and does not conduct field processing or reply to field service is solved.

Description

Technical Field

[0001] The present invention relates to the fields of smart grid and customer relationship management, and in particular to a method and system for early warning and processing of electric power customer service risks. Background Art

[0002] With the centralized management of the power customer service system, higher requirements are placed on the service management of power supply enterprises. However, the service process and service awareness of the grassroots frontline are far from meeting the needs of customers, resulting in constant complaints from customers about the quality of power supply services, which seriously affects the image of power supply enterprises and brings huge pressure to grassroots service units. Therefore, it is particularly necessary to rectify and pre-control the links and factors that affect high-quality services and improve the service awareness and ability of grassroots employees. The first key link in improving power customer service is to identify the risks of each service. Only after a full risk assessment can the services promised and provided in the service process satisfy users and complete the basic process of customer service. At present, the power customer service field mainly analyzes customer historical requests and attitude data, "labels" customers, and combines power internal line data and production data to provide a passive query and consulting service. This service method lacks the ability to predict and warn of power customer service risks, making it difficult for users to truly experience the feeling of being served. Because it is a passive service, when encountering centralized service demands, it is difficult for the power service department to fully accept them, resulting in low service efficiency and reduced customer satisfaction. Therefore, it is necessary to form a flexible method and device for early warning and handling of power customer service risks, eliminate potential risks before service risks occur, and quickly handle risks when they occur.

[0003] Existing technologies mainly rely on historical data analysis and passive services, providing inquiry and consulting services by "labeling" customers. Although this approach can meet the basic needs of customers to a certain extent, it has the following disadvantages: it is impossible to effectively predict and warn before service risks occur, making it difficult to take measures in advance to reduce risks during the service process; due to the passive service mode, when encountering centralized service demands, it is difficult for the power service department to respond quickly, resulting in low service efficiency and reduced customer satisfaction; it is impossible to reasonably allocate service resources according to risk levels and customer needs, resulting in waste of resources and uneven service quality. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is: how to predict and warn of power customer service risks in advance, and quickly handle risks when they occur, so as to improve service efficiency and customer satisfaction.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a method for early warning and processing of electric power customer service risks, including:

[0008] Determine the scope of electricity customers affected by the power outage based on the power outage information;

[0009] Based on the scope of power customers affected by the power outage and combined with historical data, the types of customer demands that may be triggered by the power outage are predicted;

[0010] Based on the prediction of customer demand types, an automated receiving and processing device for power customer service is set up to handle the situation.

[0011] As a preferred solution for early warning and handling of power customer service risks, the following are some of the solutions:

[0012] Determining the scope of power customers affected by the power outage based on the power outage information includes:

[0013] Receive power outage information generated within the power grid itself, extract the line information and administrative area map information involved in the power outage information, parse the geographical area involved in the power outage line based on the line information and administrative area map information, and determine the scope of power customers affected by the power outage in combination with power user information.

[0014] As a preferred solution for early warning and handling of power customer service risks, the following are some of the solutions:

[0015] The line information involved in extracting the power outage information includes:

[0016] Use natural language processing methods to extract the line information involved in the power outage information and find out the line names involved in the power outage information.

[0017] As a preferred solution for early warning and handling of power customer service risks, the following are some of the solutions:

[0018] The analyzing the geographical area involved in the power outage line based on the line information and the administrative area map information, and determining the scope of power customers affected by the power outage in combination with the power user information includes:

[0019] Based on the addresses of the administrative areas affected by the power outage and combined with the existing power service customer information, target customer information is selected, and the address information and customer information in the target customer information are included in the power customer service early warning catalog.

[0020] As a preferred solution for early warning and handling of power customer service risks, the following are some of the solutions:

[0021] The types of customer demands include:

[0022] Users want to know the cause of the power outage and when it will be restored; users who have had faults in the past want to know whether the old fault has reappeared; special industries and special personnel may suffer economic losses due to prolonged power outages and need to be prepared for emergency power use.

[0023] As a preferred solution for early warning and handling of power customer service risks, the following are some of the solutions:

[0024] The method of setting up an automatic receiving and processing device for power customer service based on the prediction of the type of customer demand includes:

[0025] The automatic receiving and processing device for electric power customer service includes three parts: information receiving part, information processing part and information reply part.

[0026] As a preferred solution for early warning and handling of power customer service risks, the following are some of the solutions:

[0027] The method of setting an automatic receiving and processing device for power customer service based on the prediction of the type of customer demand also includes:

[0028] After receiving the user's incoming call information, the information is processed in the information processing part, and the voice is converted into text; the user's address information and intention information are analyzed; the user's complete incoming call information is reorganized; the address elements and phone number elements of the reorganized information are extracted and split, and the comparison is performed; after the comparison is completed, the user's call demand classification is determined, and the reply materials are prepared and specific information is edited according to the classification to reply the calling customer.

[0029] In a second aspect, an embodiment of the present invention provides a power customer service risk warning and processing system, including:

[0030] A power outage impact range determination module is used to determine the range of power customers affected by the power outage based on the power outage information;

[0031] The customer demand type prediction module is used to predict the type of customer demands that may be caused by the power outage based on the scope of power customers affected by the power outage and combined with historical data;

[0032] The electric power customer service automation processing module is used to set up the electric power customer service automation receiving and processing device for processing based on the prediction of the customer demand type.

[0033] In a third aspect, an embodiment of the present invention provides a computing device, including:

[0034] Memory and processor;

[0035] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the one or more programs are executed by the one or more processors, the one or more processors implement the electric power customer service risk warning and processing method as described in any embodiment of the present invention.

[0036] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the electric power customer service risk warning and processing method.

[0037] Beneficial effects of the present invention: The present invention can prepare for related power outage services before the user calls due to the power outage, provide differentiated services according to different types of users, ensure that the user calls to experience the real problem to be solved, and improve customer service satisfaction. At the same time, the device provided by the present invention can be completed using a simple telephone and a stand-alone computer, and a flexible and mobile 24-hour customer service staff can be formed at the end power supply station of the power service, reducing the service pressure of the power grid company and solving the problem that traditional customer service only accepts work order services, does not handle on-site, and responds to on-site services. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0039] Figure 1 It is an overall flow chart of the risk warning and processing method for electric power customer service according to the present invention;

[0040] Figure 2 It is a flow chart of determining the power outage impact range of the power customer service risk warning and processing method of the present invention;

[0041] Figure 3 It is a schematic diagram of customer demand types of the electric power customer service risk warning and processing method of the present invention;

[0042] Figure 4 It is a processing flow chart of an electric power customer service automated receiving and processing device of the electric power customer service risk early warning and processing method described in the present invention. DETAILED DESCRIPTION

[0043] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0044] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0045] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0046] Example 1

[0047] Reference Figure 1 , which is the first embodiment of the present invention, and provides a method for early warning and processing of risk of electric power customer service, comprising:

[0048] S1: Determine the scope of electricity customers affected by the power outage based on the power outage information;

[0049] S2: Based on the scope of power customers affected by the power outage and combined with historical data, predict the types of customer demands that may be triggered by the power outage;

[0050] S3: Based on the prediction of customer demand types, an automated receiving and processing device for power customer service is set up for processing.

[0051] It should be noted that through steps S1-S3, a complete closed loop from power outage information acquisition to customer demand prediction and then to automated processing can be achieved. This process not only improves the response speed and processing efficiency of power customer service, but also responds to possible customer demands in advance through data analysis and prediction, thereby improving customer satisfaction and enterprise operating efficiency.

[0052] Example 2

[0053] Reference Figure 1-Figure 4 , which is an embodiment of the present invention, provides a method for early warning and processing of power customer service risks based on the previous embodiment, including:

[0054] In the embodiment of the present application, determining the scope of power customers affected by the power outage according to the power outage information in the above step S1 includes:

[0055] Receive power outage information generated within the power grid itself, extract the line information and administrative area map information involved in the power outage information, parse the geographical area involved in the power outage line based on the line information and administrative area map information, and determine the scope of power customers affected by the power outage in combination with power user information.

[0056] Specifically, Figure 2 As shown, the NLP natural language processing method is used to extract the line information involved in the power outage information and find out the line names involved in the power outage information.

[0057] For example, on November 20, 2024, due to overload, the Haifu 1# transformer tripped, and the Haifu 1 horizontal line, Longji line, provincial government line, and Longkun line were out of power. Find the words with "line" at the end of the power outage information to form the "line" that affects it. Then query the coordinate point location of the line in the line database according to the "line" name. Finally, use the coordinate point location combined with GIS map information to query the administrative area address involved and determine the administrative area address involved in the power outage line.

[0058] In another possible implementation, the line information involved in the power outage information can be extracted by combining a hybrid method of a rule base and machine learning to find out the line names involved in the power outage information:

[0059] Create a rule base:

[0060] Keyword matching: Define keywords related to power lines, such as "line", "road", "substation", etc., and preliminarily locate possible line names through keyword matching.

[0061] Regular expression: Design a regular expression to match common line naming formats, such as "XX line", "XX branch", etc.

[0062] Machine Learning Model Training:

[0063] Feature extraction: Extract text features from power outage information, such as word frequency, N-gram, part-of-speech tagging, etc.

[0064] Label annotation: Manually annotate a batch of power outage information samples and clearly indicate which words are line names.

[0065] Model selection: Choose a machine learning model suitable for named entity recognition (NER), such as conditional random field (CRF), LSTM+CRF, etc.

[0066] Hybrid method application:

[0067] Preliminary screening: Use keyword matching and regular expressions in the rule base to perform preliminary screening of power outage information and extract possible line names.

[0068] Model verification: The initially screened route names are input into the trained machine learning model to further confirm and correct the accuracy of the route names.

[0069] Result integration: Integrate the route names after model verification with the screening results of the rule base to obtain the final route information list.

[0070] For example, the following power outage information is provided:

[0071] "Due to equipment maintenance, the Haifu East Line, Haifu West Line and Longhua Line of Haifu 2# Substation will be shut down on December 5, 2024, and the power restoration time is expected to be 3 pm on the same day."

[0072] Rule base application:

[0073] Keyword matching: extract words containing "line", such as "Haifu East Line", "Haifu West Line", and "Longhua Line".

[0074] Regular expression: Use regular expression to match the pattern of "XX line" to further confirm the above words.

[0075] Machine Learning Model Validation:

[0076] Input text: "Due to equipment maintenance, the Haifu East Line, Haifu West Line and Longhua Line of Haifu 2# Substation will be shut down on December 5, 2024. The estimated time of restoration is 3:00 p.m. on the same day."

[0077] Model output: Confirm that "Haifu East Line", "Haifu West Line" and "Longhua Line" are line names, and may correct or identify line names not covered by the rule base.

[0078] Results integration:

[0079] The final route list: "Haifu East Line", "Haifu West Line", and "Longhua Line".

[0080] Furthermore, based on the addresses of the administrative areas affected by the power outage and combined with the existing electricity service customer information, the existing customer information is circled, and the address information and customer information in the information is entered into the electricity customer service warning directory.

[0081] In the embodiment of the present application, in the above step S2, based on the scope of power customers affected by the power outage and in combination with historical data, the type of customer demands that may be caused by the power outage is predicted, including:

[0082] like Figure 3As shown, the types of customer demands include: users want to know the cause of power outage and the time for power restoration; users who have had faults in the past want to know whether the old fault has recurred; special industries and special personnel will suffer economic losses due to prolonged power outages and need to be prepared for emergency power use (such as the breeding industry, medical and health industry, food processing industry, etc.).

[0083] It should be noted that after predicting the types of customer demands that may be caused by this power outage, different types of demands can be received and processed automatically in the subsequent automatic receiving device.

[0084] In another possible implementation, how to classify the types of customer demands can be implemented by using a variety of methods and technical means to improve the accuracy and efficiency of classification. The following is a detailed implementation:

[0085] 1. Data preparation and preprocessing

[0086] 1.1 Data Collection:

[0087] Historical power outage data: Collect relevant information of previous power outage events, including power outage time, power outage scope, power outage cause, etc.

[0088] Customer feedback data: Collect customer feedback information from previous power outages, including customer demands, feedback time, solutions, etc.

[0089] 1.2 Data preprocessing:

[0090] Text cleaning: remove irrelevant symbols and stop words, and unify the format.

[0091] Feature extraction: Extract features such as keywords, phrases, and sentiment from customer feedback.

[0092] Data labeling: Manually label historical customer feedback data to clarify the type of demand to which each feedback belongs.

[0093] 2. Classification model selection and training

[0094] 2.1 Feature Engineering:

[0095] Term Frequency-Inverse Document Frequency (TF-IDF): used to extract important features from text.

[0096] Word Embedding: Such as Word2Vec, GloVe, etc., convert text into vector representation.

[0097] Sentiment analysis: Identify the emotional tendencies in customer feedback, such as positive, neutral, and negative.

[0098] 2.2 Model selection:

[0099] Supervised learning models: such as support vector machine (SVM), random forest, gradient boosted tree (GBDT), etc.

[0100] Deep learning models: such as convolutional neural networks (CNN), recurrent neural networks (RNN), long short-term memory networks (LSTM), etc.

[0101] 2.3 Model training:

[0102] Dataset division: Divide the labeled dataset into training set, validation set and test set.

[0103] Model training: Use the training set to train the model and adjust hyperparameters to optimize model performance.

[0104] Model evaluation: Use the validation set and test set to evaluate the classification accuracy, precision, recall, and F1 value of the model.

[0105] 3. Prediction and application

[0106] 3.1 Real-time prediction:

[0107] Data input: Input relevant information and historical data of the current power outage event into the trained model.

[0108] Prediction results: The model outputs the types of customer demands that may be caused by this power outage.

[0109] 3.2 Automated processing:

[0110] Application of classification results: Automatically assign to the corresponding processing flow based on the predicted customer demand type.

[0111] Automated responses: Pre-prepared answer templates to automatically respond to customers based on different types of requests.

[0112] Emergency processing: For the demands of special industries and special personnel, such as the breeding industry, medical and health industry, food processing industry, etc., priority will be given to processing and emergency electricity preparation plans will be provided.

[0113] For example, the current power outage event is as follows:

[0114] "On December 5, 2024, the Haifu 1# transformer tripped, and power was cut off in the areas of Haifu 1 Heng Line, Longji Line, Shengfu Line, and Longkun Line."

[0115] Data preparation and preprocessing:

[0116] Text cleaning: remove irrelevant symbols and unify the format.

[0117] Feature extraction: Extract keywords "Haifu 1# transformer", "trip", "Haifu 1 horizontal line", "Longji line", "Provincial government line", and "Longkun line".

[0118] Data annotation: Combine historical data to annotate the types of customer demands in similar power outage incidents.

[0119] Classification model selection and training:

[0120] Feature engineering: Use TF-IDF to extract text features and use Word2Vec to generate word vectors.

[0121] Model selection: Select the SVM model for training.

[0122] Model training: Use the training set data to train the SVM model and adjust the hyperparameters to optimize performance.

[0123] Model evaluation: Use the validation set and test set to evaluate the classification accuracy of the model.

[0124] Prediction and Application:

[0125] Data input: Input the information of the current power outage event into the trained SVM model.

[0126] Prediction results: Predict possible customer demand types, such as "users want to know the cause of the power outage and the time of power restoration", "users who have had faults in the past want to know whether the old fault has recurred", and "users in special industries need emergency power preparations".

[0127] Automated processing: Automatically receive and process customer calls based on prediction results, and provide targeted responses and services.

[0128] In the embodiment of the present application, in the above step S3, based on the prediction of the customer demand type, setting the power customer service automated receiving and processing device for processing includes:

[0129] like Figure 4 As shown, the automatic receiving and processing device for electric power customer service includes three parts: information receiving part, information processing part and information reply part;

[0130] After receiving the user's incoming call information, the first step is to process the information in the information processing device. The specific action method is:

[0131] To convert speech into text, we mainly use the more mature speech recognition conversion components from the outside world.

[0132] The term frequency-inverse proximity algorithm (TF-IDF) is used for semantic understanding to analyze the user's location information and intention information.

[0133] Then, based on the user's language address and the user's caller ID information, the user's complete caller information is reorganized.

[0134] The address elements and phone number elements are extracted and split from the reorganized information, and compared with the information generated in the first and second stages. The comparison process is implemented using the Levenshtein Distance algorithm.

[0135] After the comparison is completed, the user's call demand classification is determined, and the response materials are prepared and specific information is edited according to the classification, and the caller is answered through text-to-speech. At the same time, special classifications are formed into short message content and sent to the customer service personnel's mobile phone via SMS.

[0136] In another possible implementation, based on the recorded call records, after the power outage ends, the service experience is actively compiled and the user feedback figures are counted to complete the overall process of risk warning and handling of overall power customer service.

[0137] Example 3

[0138] The above is a schematic scheme of the power customer service risk warning and processing method of this embodiment. It should be noted that the technical scheme of the power customer service risk warning and processing system and the technical scheme of the power customer service risk warning and processing method above belong to the same concept. For the details not described in detail in the technical scheme of the power customer service risk warning and processing system in this embodiment, please refer to the description of the technical scheme of the power customer service risk warning and processing method above.

[0139] This embodiment also provides a system based on the risk early warning and processing method of power customer service, including:

[0140] A power outage impact range determination module is used to determine the range of power customers affected by the power outage based on the power outage information;

[0141] The customer demand type prediction module is used to predict the type of customer demands that may be caused by the power outage based on the scope of power customers affected by the power outage and combined with historical data;

[0142] The electric power customer service automation processing module is used to set up the electric power customer service automation receiving and processing device for processing based on the prediction of the customer demand type.

[0143] This embodiment also provides a computing device, which is applicable to the situation of risk warning and processing method for power customer service, including:

[0144] Memory and processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the power customer service risk warning and processing method proposed in the above embodiment.

[0145] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the power customer service risk warning and processing method proposed in the above embodiment is implemented.

[0146] The storage medium proposed in this embodiment and the power customer service risk warning and processing method proposed in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be found in the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0147] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for early warning and processing of power customer service risks, characterized in that: include: Determine the scope of electricity customers affected by the power outage based on the power outage information; Based on the scope of power customers affected by the power outage and combined with historical data, the types of customer demands that may be triggered by the power outage are predicted; Based on the prediction of customer demand types, an automated receiving and processing device for power customer service is set up to handle the situation.

2. The power customer service risk warning and processing method according to claim 1, characterized in that: Determining the scope of power customers affected by the power outage based on the power outage information includes: Receive power outage information generated within the power grid itself, extract the line information and administrative area map information involved in the power outage information, parse the geographical area involved in the power outage line based on the line information and administrative area map information, and determine the scope of power customers affected by the power outage in combination with power user information.

3. The method for early warning and processing of power customer service risks according to claim 2, characterized in that: The line information involved in extracting the power outage information includes: Use natural language processing methods to extract the line information involved in the power outage information and find out the line names involved in the power outage information.

4. The power customer service risk warning and processing method according to claim 3, characterized in that: The analyzing the geographical area involved in the power outage line based on the line information and the administrative area map information, and determining the scope of power customers affected by the power outage in combination with the power user information includes: Based on the addresses of the administrative areas affected by the power outage and combined with the existing power service customer information, target customer information is selected, and the address information and customer information in the target customer information are included in the power customer service early warning catalog.

5. The method for early warning and processing of electric power customer service risks according to claim 4, characterized in that: The types of customer demands include: Users want to know the cause of the power outage and when it will be restored; users who have had faults in the past want to know whether the old fault has reappeared; special industries and special personnel may suffer economic losses due to prolonged power outages and need to be prepared for emergency power use.

6. The method for early warning and processing of electric power customer service risks according to claim 5, characterized in that: The method of setting up an automatic receiving and processing device for power customer service based on the prediction of the type of customer demand includes: The automatic receiving and processing device for electric power customer service includes three parts: information receiving part, information processing part and information reply part.

7. The method for early warning and processing of electric power customer service risks according to claim 6, characterized in that: The method of setting an automatic receiving and processing device for power customer service based on the prediction of the type of customer demand also includes: After receiving the user's incoming call information, the information is processed in the information processing part, and the voice is converted into text; the user's address information and intention information are analyzed; the user's complete incoming call information is reorganized; the address elements and phone number elements of the reorganized information are extracted and split, and the comparison is performed; after the comparison is completed, the user's call demand classification is determined, and the reply materials are prepared and specific information is edited according to the classification to reply the calling customer.

8. A system using the power customer service risk warning and processing method according to any one of claims 1 to 7, characterized in that: include: A power outage impact range determination module is used to determine the range of power customers affected by the power outage based on the power outage information; The customer demand type prediction module is used to predict the type of customer demands that may be caused by the power outage based on the scope of power customers affected by the power outage and combined with historical data; The electric power customer service automation processing module is used to set up the electric power customer service automation receiving and processing device for processing based on the prediction of the customer demand type.

9. A computing device comprising: Memory and processor; The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the electric power customer service risk warning and processing method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the method for risk warning and processing of electric power customer service as described in any one of claims 1 to 7.

Images