Repair processing method and system based on intelligent customer service, medium and program product

Abstract

The invention provides a repair processing method and system based on intelligent customer service, a medium and a program product, and relates to the technical field of electric digital data processing. The method comprises the following steps: acquiring repair information of a user, determining an associated repair information set, extracting geographic position information and fault phenomenon keywords of associated repair points, and generating a three-level pipe network association map; obtaining time-phased water consumption data in the map in real time, and determining an abnormal water consumption area; acquiring hardware state data of an abnormal area, and determining a fault type in combination with pressure fluctuation and a sound wave spectrum; and according to the fault type, generating an intelligent work order including the fault type, the geographic coordinate, the pipe network level and the maintenance tool accessory list. By implementing the method, the intelligence and precision of repair processing are realized.

Description

Technical Field

[0001] This application relates to the technical field of electronic digital data processing, and particularly to a repair handling method, system, medium and program product based on intelligent customer service. Background Art

[0002] In the scenario of water service repair, the traditional customer service system only generates work orders based on the user's immediate description, and can only achieve the electronic recording of repair information, resulting in the work order content only staying on the surface description. This causes maintenance personnel to conduct on-site inspections one by one, with an inefficient disposal process and a lack of data support, making it difficult to meet the requirements of intelligent water service for data-driven precise operation and maintenance.

[0003] To improve the above defects, the prior art proposes a repair classification method based on keyword matching. By presetting keywords such as "leakage", "low water pressure", and "abnormal water quality", the user's description is automatically classified to generate standardized work orders. This technology realizes the rapid classification of work orders through simple text processing and improves the dispatching efficiency to a certain extent.

[0004] However, the core technical problem of the prior art is that it is essentially a passive response mode, completely relying on the user's subjective description of the fault phenomenon and lacking the ability to deeply analyze big data. After the maintenance personnel arrive at the scene, they need to re-collect data to investigate the real cause, resulting in a low matching degree between the work order and the actual fault cause, wasting human and time resources. Especially in a complex pipe network environment, this dispatching mode without data analysis support is difficult to quickly locate the root cause of the fault and accurately allocate maintenance resources. Summary of the Invention

[0005] This application provides a repair handling method, system, medium and program product based on intelligent customer service, which is used to realize the intelligent management of the entire process of user repair and improve the efficiency of pipe network fault diagnosis and the accuracy of repair scheduling.

[0006] In a first aspect, the present application provides a repair handling method based on an intelligent customer service, which is applied to an intelligent customer service system. The method includes: obtaining user repair information and determining an associated repair information set; extracting geographical location information and fault phenomenon keywords of associated repair points from the associated repair information set through natural language processing algorithms; combining the geographical location information to generate a three-level pipe network association map covering all the associated repair points and extending outward by a set distance. The three-level pipe network association map includes a primary associated pipe, a secondary associated pipe, and a tertiary associated pipe. The primary associated pipe is a pipe directly connected to the repair point. The secondary associated pipe is a branch pipe connected to the primary associated pipe. The tertiary associated pipe is a main pipe connected to the secondary associated pipe and having a pipe diameter greater than or equal to a preset pipe diameter threshold; obtaining sub-period water consumption data of each pipe area within the three-level pipe network association map in real time through intelligent metering devices; determining a water consumption abnormal pipe area in combination with the sub-period water consumption data; obtaining real-time hardware status data of the water consumption abnormal pipe area, where the hardware status data at least includes continuous pressure fluctuation data, valve torque, switch frequency, pressure curve, and acoustic frequency spectrum signal; determining a fault type in combination with the continuous pressure fluctuation data and the acoustic frequency spectrum signal; generating a corresponding intelligent work order according to the fault type, where the work order at least includes the fault type, geographical coordinates and pipe network level of the water consumption abnormal pipe area, and a list of tools and accessories required for repair.

[0007] By adopting the above technical solution, first, the user repair information is obtained and the associated repair information set is determined, and key information is extracted from it by using natural language processing algorithms, which is a preliminary screening and mining of the user repair big data. Generating a three-level pipe network association map can visually present the pipe network structure. Using intelligent metering devices to obtain sub-period water consumption data, combining this data to determine the abnormal area, and then obtaining the hardware status data to determine the fault type and generate a work order. The whole process deeply analyzes multi-source big data, from repair information to pipe network data, water consumption data, and hardware status data, advancing layer by layer, solving the problem of lack of deep analysis ability for big data, and realizing accurate fault location and processing.

[0008] Combined with some embodiments of the first aspect, in some embodiments, in the step of obtaining user repair information and determining an associated repair information set, it specifically includes: extracting time data and community information from the user repair information; extracting fault keywords from the user repair information; if the similarity of the fault keywords reaches a set threshold, and the community information is the same or within a set distance range, and the time data is within a set time window, then the corresponding user repair information is merged and determined as the associated repair information set.

[0009] By adopting the above technical solutions, when obtaining the user's repair information to determine the associated repair information set, time data, community information, and fault keywords are extracted. When the similarity of the fault keywords meets the standard, the community information is similar, and the time is within the set window, the relevant repair information is merged. In this way, the scattered and possibly related repair information can be integrated, avoiding duplicate processing and reducing unnecessary work processes. At the same time, the integrated information is more comprehensive, which can provide richer and more accurate data for subsequent analysis of the geographical location and fault phenomena, improve the accuracy and efficiency of fault analysis, and better grasp the overall fault situation.

[0010] Combined with some embodiments of the first aspect, in some embodiments, the step of generating a three - level pipe network association map covering all the associated repair points and extending outward by a set distance in combination with the geographical location information specifically includes: based on the geographical location information, generating a minimum closed irregular spatial envelope through the minimum bounding polygon algorithm; taking the minimum closed irregular spatial envelope as a reference and extending it outward by a set distance to generate a standard irregular spatial envelope; obtaining all pipe data in the standard irregular spatial envelope, where the pipe data at least includes pipe ID, pipe spatial coordinates, and pipe diameter; constructing a three - level pipe network association map according to the standard irregular spatial envelope and the pipe data.

[0011] By adopting the above technical solutions, the minimum bounding polygon algorithm can accurately frame the range of associated repair points. The extended standard envelope can cover the surrounding pipes that may be affected. By obtaining detailed pipe data, the map can be accurately constructed, clearly showing the connection and distribution relationships of pipes at all levels, providing an intuitive and accurate pipe network model for subsequent analysis of pipe water usage and fault location.

[0012] Combined with some embodiments of the first aspect, in some embodiments, the step of determining the area of water - using abnormal pipes in combination with the sub - period water consumption data specifically includes: calculating the deviation rate in real - time in combination with the sub - period water consumption data; if the deviation rate exceeds the preset deviation rate threshold, then determining the pipe area corresponding to the deviation rate as the area of water - using abnormal pipes.

[0013] By adopting the above technical solutions, the sub - period water consumption data can reflect the actual water usage of pipes at different time periods. The deviation rate calculation can quantify the difference between the actual water usage and the normal situation. By setting a reasonable threshold, the area of water - using abnormal pipes can be quickly and accurately identified, providing a clear target for subsequent acquisition of hardware status data and determination of fault types, and improving the pertinence and efficiency of fault troubleshooting.

[0014] In combination with some embodiments of the first aspect, in some embodiments, the step of determining the fault type by combining the continuous pressure fluctuation data and the acoustic wave frequency spectrum signal specifically includes: if the deviation rate is greater than the first deviation rate threshold, and the pressure curve drops by more than the first pressure drop threshold within the set time, and the acoustic wave frequency spectrum is greater than the first frequency within the set continuous time, then determine that the fault type is pipeline rupture; if the deviation rate is less than the second deviation rate threshold, and the amplitude by which the valve torque exceeds the standard value is greater than the first valve torque amplitude, and the switching frequency increases abnormally, then determine that the fault type is pipe network blockage; if the deviation rate is less than the third deviation rate threshold, and the amplitude of the pressure curve is greater than the first pressure curve amplitude, and the proportion of low-frequency components in the acoustic wave frequency spectrum is greater than the first acoustic wave spectrum proportion, then determine that the fault type is valve failure.

[0015] By adopting the above technical solutions, through the technical means of constructing a fault determination rule based on multi-parameter combination conditions (deviation rate + pressure curve + acoustic wave frequency spectrum) and training a threshold model based on historical data, the technical problems in the prior art such as low reliability of single-index diagnosis, dependence on manual experience for fault type identification, and inability to accurately distinguish similar fault characteristics are effectively solved. The effects of establishing a fault feature vector through multi-dimensional data joint analysis, automatically matching historical fault patterns, and accurately identifying specific types such as pipeline rupture, blockage, and valve failure are achieved, and the accuracy and intelligent level of fault diagnosis are improved.

[0016] In combination with some embodiments of the first aspect, in some embodiments, after the step of generating a corresponding intelligent work order according to the fault type, it further includes: obtaining the current time information; if the time information is within the set working hours, then obtaining the ability information and location information of the maintenance personnel; combining the ability information and location information, and determining the maintenance personnel with the closest geographical coordinate distance to the abnormal water use pipeline area and matching ability according to the fault type.

[0017] By adopting the above technical solutions, the location information ensures the priority dispatch of the personnel closest to the fault point, and the ability label ensures the matching of the maintenance skills and the fault type, avoiding the efficiency waste caused by "long-distance dispatch" or "skill mismatch". This process deeply integrates personnel data (location, skills) with fault data, and realizes the dynamic matching of "optimal resources - optimal tasks" through real-time calculation, improving the real-time decision-making ability of big data in resource scheduling.

[0018] In combination with some embodiments of the first aspect, in some embodiments, it further includes: if the time information is outside the set working hours, then obtaining the ability information and home address information of the maintenance personnel; combining the ability information and home address information, and determining the maintenance personnel with the closest geographical coordinate distance to the abnormal water use pipeline area and matching ability according to the fault type.

[0019] By adopting the above technical solution, this strategy solves the problem of missing real-time positioning data during non-working hours. It utilizes the address data stored historically to continue the principle of spatial priority, while maintaining the professionalism of capacity matching, ensuring that the fault response during nights or holidays is not disjointed, and enhancing the adaptability of big data analysis to complex time scenarios and the full-time coverage ability.

[0020] In a second aspect, the present application provides an intelligent customer service system, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, and the computer program code includes computer instructions. The one or more processors call the computer instructions to enable the intelligent customer service system to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0021] In a third aspect, the present application provides a computer-readable storage medium, including instructions, which when running on the intelligent customer service system, enable the intelligent customer service system to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0022] In a fourth aspect, the present application provides a computer program product, which when running on the intelligent customer service system, enables the intelligent customer service system to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0023] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. Due to the technical means of using natural language processing algorithms to extract key data of repair information, constructing a three-level pipe network association map to integrate spatial and pipeline attributes, and performing correlation analysis on multi-source dynamic data (time-sharing water consumption data, hardware status data), the technical problems of fragmented user repair data, insufficient visualization of pipe network structure, low analysis efficiency and poor positioning accuracy caused by relying on manual experience in the prior art are effectively solved. Furthermore, the technical effects of automatically mining fault correlation features from multi-dimensional data, constructing an accurate pipe network topology model, and realizing intelligent determination of fault types through data-driven algorithms are achieved, significantly improving the automation level of repair processing and the fault handling efficiency.

[0024] 2. By adopting technical means such as the minimum circumscribed polygon algorithm to generate a spatial envelope, setting a distance for outward extension to analyze the scope, and constructing a three-level pipe network hierarchy relationship based on pipe IDs and pipe diameters, the technical problems in the prior art of fuzzy pipe network analysis scope, unclear pipe association levels, and fragmentation of spatial data and physical attributes are effectively solved. Furthermore, the technical effects of converting the geographical distribution of repair points into a computable pipe network topology model, highlighting the key role of main pipes through hierarchical division, and providing an accurate spatial index for subsequent data monitoring are achieved, making it possible for visual analysis of the pipe network structure and rapid positioning of the fault area.

[0025] 3. By adopting technical means such as constructing a fault determination rule using a multi-parameter combination condition (deviation rate + pressure curve + acoustic wave spectrum) and training a threshold model based on historical data, the technical problems in the prior art of low reliability of single-index diagnosis, dependence on manual experience for fault type identification, and inability to accurately distinguish similar fault characteristics are effectively solved. Furthermore, the technical effects of establishing a fault feature vector through multi-dimensional data joint analysis, automatically matching historical fault patterns, and accurately identifying specific types such as pipe rupture / clogging / valve failure are achieved, improving the accuracy and intelligent level of fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a flowchart of a repair handling method based on an intelligent customer service in an embodiment of the present application; Figure 2 is another flowchart of a repair handling method based on an intelligent customer service in an embodiment of the present application; Figure 3 is a schematic structural diagram of an entity device of an intelligent customer service system in an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification and appended claims of the present application, the singular forms "a", "an", "the", "above", "said", "this" are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in the present application refers to and includes any or all possible combinations of one or more of the listed items.

[0028] Hereinafter, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of the present application, unless otherwise stated, the meaning of "a plurality" is two or more.

[0029] For ease of understanding, the following is a description of the process of the method provided by this implementation. Figure 1 , which is a flow chart of a repair processing method based on intelligent customer service in an embodiment of the present application.

[0030] S101, obtaining user repair information and determining a related repair information set; Among them, user repair information refers to the fault reporting data submitted by users through the intelligent customer service system, which refers to information in various forms such as text descriptions, voice records, pictures or videos, which are used to indicate the pipe network problems encountered by users, such as "low water pressure in Building 1 of a certain community" and "there is a sound of water leakage downstairs in Unit 3". The associated repair information set refers to a group of repair information with spatiotemporal correlation and similar fault characteristics. By analyzing the correlation of dimensions such as time, location, and fault type, scattered individual repair information is aggregated into a statistically significant set. For example, the "water leakage" incident reported multiple times in the same community within 24 hours can be regarded as an associated repair information set.

[0031] This step is executed when the user initiates a repair request through the intelligent customer service system, and the application scenario is the initial data collection and preprocessing stage of water pipe network failure. Specifically, the intelligent customer service system first obtains the repair information submitted by the user in real time through multiple channels such as online chat windows, voice calls, and APP repair portals, and performs preliminary cleaning on each piece of information to remove duplicate submissions or data with incorrect formats. Then it enters the association analysis phase. The system builds a three-dimensional screening model based on the time sliding window (such as the past 1 hour, 12 hours, etc.), geographic fences (such as cells, 500 meters in radius, etc.) and fault keywords (such as "leakage", "blockage", "abnormal water pressure", etc.). In the time dimension, the repair information submitted in the same time period is included in the candidate set; in the geographical dimension, the address described by the user is converted into longitude and latitude coordinates through the address resolution algorithm, and the spatial distance is calculated to determine whether it falls within the preset geographical range; in the fault dimension, the keyword matching algorithm is used to calculate the semantic similarity between the repair content and the preset fault type. When a repair report meets the time window, geographical overlap, and fault keyword similarity exceeds a threshold (such as 80%), the system merges it with the existing association set to form a complete set of associated repair report information. This process can effectively filter out isolated and invalid data, focus on high-frequency fault areas, and provide a structured data set for subsequent pipe network analysis.

[0032] In some embodiments, determining the associated repair information set can be achieved in various ways: Optionally, the system first extracts the timestamp, community name or address coordinates, and fault keywords (such as matching roots like "leak", "block", "pressure" through regular expressions) from each repair information, then establishes a three-dimensional index table of time-geography-keywords, performs hash bucketing on all repair information, and the information within the same bucket is automatically marked as a potential associated set. Finally, the DBSCAN density clustering algorithm is used to calculate the density of the data in the bucket, and the samples that are density-reachable are clustered into an associated set. For example, if there are 5 repair information within 3 hours in a certain community, all of which contain the keyword "leakage" and the address coordinates are less than 200 meters apart, the system clusters them into a leakage associated set. Optionally, an association determination method based on a rule engine is adopted, and multiple determination rules are preset, such as "the same community + the same fault type + ≥3 repairs within 24 hours" automatically triggers the creation of an associated set, or "adjacent communities + similar fault keywords + time interval ≤1 hour" is regarded as a cross-regional associated set. For example, in the adjacent Community A and Community B, 3 reports of "low water pressure" and 2 reports of "small water flow" are respectively reported between 9 am and 10 am. The system determines through semantic analysis that "low water pressure" and "small water flow" belong to the same type of fault, and they are close in time and space, and are merged into a water pressure anomaly associated set. It can be understood that a machine learning-based association prediction model can also be adopted. The classifier (such as random forest, neural network, etc.) is trained through historical associated set data, and the time, geography, and text features of the new repair information are input, and the probability value of it belonging to a certain existing associated set is output. When the probability exceeds the threshold, it is automatically merged, which is not limited here.

[0033] S102. Extract the geographical location information and fault phenomenon keywords of the associated repair points in the associated repair information set through natural language processing algorithms; Among them, the natural language processing algorithm represents a set of technologies for processing, understanding, and generating human language, which refers to the method of analyzing the grammar, semantics, and pragmatics of natural language through computer programs, such as word segmentation, named entity recognition, keyword extraction, etc. The associated repair point refers to the physical location corresponding to each specific repair event in the associated repair information set, which is the geographical coordinates (such as longitude and latitude) obtained through address parsing or user positioning, used to identify the specific location where the fault occurs. For example, the coordinate point corresponding to the green belt on the east side of Building 1 in a certain community. The geographical location information refers to the spatial location data related to the associated repair point, including detailed address, administrative division, geographical coordinates, etc., used to represent the spatial range where the fault occurs. For example, "Building 1, XX Community, XX Street, XX District, XX City" and its corresponding longitude and latitude coordinates. The fault phenomenon keyword refers to the words or phrases that can characterize the fault type and characteristics, which are extracted from the repair content through text analysis, such as "leakage", "urgent water flow", "abnormal pipeline noise", etc., and are used for subsequent fault classification and pipeline network analysis.

[0034] This step is executed after the associated repair information set is constructed, and the application scenario is the preprocessing stage of extracting key feature data from unstructured text data. Specifically, the intelligent customer service system first performs sentence splitting and word segmentation on each repair content in the associated repair information set, and uses a word segmentation tool to split the natural language text into the smallest semantic units (such as words). Then it enters the named entity recognition (NER) link, and uses a pre-trained model to recognize geographical location entities (such as community names, road names, house numbers, etc.) and fault phenomenon entities (such as "leakage", "blockage", "water pressure", etc.) in the text. For the extraction of geographical location information, the system first identifies address keywords, such as "XX Community" and "No. XX, XX Road", and then converts the text address into an accurate longitude and latitude coordinate through an address encoding interface (such as Baidu Map API), and records the address confidence level (such as the matching accuracy is at the house number level, building level or community level). For the extraction of fault phenomenon keywords, a preset algorithm is used to calculate the importance score of words, and the words with scores higher than the threshold are selected as keywords. At the same time, normalization processing is performed through a thesaurus. For example, "the water pipe is leaking" and "the pipeline is seeping water" are uniformly mapped to the keyword "leakage". In addition, the system also identifies the modifiers (such as "serious", "continuous", "sudden", etc.) in the text for subsequent fault severity assessment. The whole process realizes the automatic conversion from the user's natural language description to structured geographical coordinates and fault labels, providing basic data support for generating a pipe network association map and fault diagnosis.

[0035] In some embodiments, the extraction of geographical location information and fault phenomenon keywords can be achieved in various ways: Optionally, a rule template-based extraction method is adopted, and address templates (such as "[administrative region][street][community][building number][unit number]") and fault templates (such as "[fault type]+[degree word]+[location]") are predefined, and regular expressions are used to match the strings in the text that conform to the template pattern.

[0036] S103. Combine the geographical location information to generate a three-level pipe network association map covering all the associated repair points and extending outward by a set distance. The three-level pipe network association map includes primary associated pipes, secondary associated pipes, and tertiary associated pipes. The primary associated pipes are the pipes directly connected to the repair points, the secondary associated pipes are the branch pipes connected to the primary associated pipes, and the tertiary associated pipes are the main pipes connected to the secondary associated pipes and with a pipe diameter greater than or equal to a preset pipe diameter threshold; Among them, the geographical location information represents spatial data such as the geographical coordinates and detailed addresses of the associated repair points extracted in step S102, and is used to locate the specific location where the fault occurs. The three-level pipe network association map refers to a multi-level pipe network topology model constructed based on geographical information and pipe network structure, which graphically displays the spatial association relationship between the repair point and each level of pipeline, including the hierarchical structure and connection relationship of primary, secondary, and tertiary pipelines. The primary associated pipeline refers to the pipeline directly physically connected to the repair point, usually the branch pipe in the user's home or building, with a smaller pipe diameter (such as DN20 - DN50). For example, the branch pipe connected to the faucet in the resident's home. The secondary associated pipeline refers to the branch pipeline connected to the primary pipeline, usually the sub-pipe between buildings or within the community, with a medium pipe diameter (such as DN80 - DN150). For example, the water supply branch pipe connecting each building in the community. The tertiary associated pipeline refers to the main pipeline connected to the secondary pipeline, usually the main water supply pipe under the municipal road, with a larger pipe diameter (≥ preset threshold, such as DN200), which undertakes the main water conveyance function. For example, the DN300 water supply main pipe under the urban main road. The set distance represents the spatial range extended outward with the associated repair point as the center, and is used to determine the analysis range of the map. For example, when the set distance is 200 meters, it means that the map covers all pipelines within a radius of 200 meters centered on the repair point.

[0037] This step is executed after obtaining the geographical location information and the keywords of the fault phenomenon. The application scenario is the stage of constructing a pipe network topology model to support subsequent water usage data monitoring and fault location. Specifically, the intelligent customer service system first generates a minimum closed irregular spatial envelope that contains all reported repair points based on the geographical coordinates of the associated repair points using the minimum bounding polygon algorithm. This envelope tightly wraps all reported repair points to form the initial analysis scope. Then, taking this minimum envelope as the benchmark, it evenly expands outward at a set distance (such as 100 meters, 200 meters, etc.) to generate a standard irregular spatial envelope to ensure that it covers the pipe areas that may be affected around the reported repair points. Next, the system retrieves all pipe data within the standard envelope range from the pipe network geographic information system (GIS), including attribute information such as pipe ID, spatial coordinates, pipe diameter, material, and burial depth. According to the pipe hierarchy division rules, the pipes directly connected to the reported repair points are marked as primary associated pipes, the branch pipes connected to the primary pipes are marked as secondary associated pipes, and the pipes connected to the secondary pipes and with a pipe diameter ≥ the preset threshold are marked as tertiary associated pipes. Finally, using a graph database or a GIS visualization tool, the pipes at all levels and their connection relationships are displayed in the form of a graph. Nodes represent pipe nodes (such as valves, water meter wells), edges represent pipe connection relationships, and different levels of pipes are distinguished by different colors or line types (such as primary pipes are blue dotted lines, secondary pipes are green solid lines, and tertiary pipes are red thick lines). The location of the reported repair points and the fault keywords are marked in the graph to form an intuitive three-level pipe network association graph. This graph not only shows the spatial distribution of the reported repair points but also reveals their physical connection relationships with the pipe networks at all levels, providing a spatial index for subsequent monitoring of water consumption data in different time periods and fault diagnosis.

[0038] In some embodiments, generating the three-level pipe network association graph can be achieved in various ways: Optionally, a rule-based hierarchy division method is adopted. First, the connection relationship between the reported repair points and the pipes is determined through spatial topology analysis (such as judging whether the reported repair points are on a certain pipe through point-line spatial intersection analysis), and the directly connected pipes are set as primary; then, starting from the end nodes of the primary pipes, the downstream pipes connected to them are searched, and those with a pipe diameter less than the preset threshold are set as secondary, and those greater than or equal to the threshold are set as tertiary; this process is repeated until all pipes are traversed.

[0039] S104. Real-time obtain the water consumption data in different time periods of each pipe area within the three-level pipe network association graph through intelligent metering devices; Among them, the intelligent metering device refers to an Internet of Things device installed in the pipe network with data collection, transmission, and processing functions, such as ultrasonic flow meters, electromagnetic flow meters, intelligent water meters, etc., which are used to monitor the water flow data in the pipeline in real time. For example, the electromagnetic flow meter installed on a certain DN200 main pipe can collect flow data in real time. The water consumption data by time period refers to the water consumption data of each pipe area statistically calculated according to a preset time interval (such as 15 minutes, 1 hour, etc.), which is used to reflect the water usage patterns and abnormal fluctuations during different time periods. For example, the water consumption by time period of a branch pipe in a certain community from 7 to 9 am (peak water usage) is 50 cubic meters per hour, and at night it is 5 cubic meters per hour. The pipe area refers to the spatial range of different levels of pipes divided according to the three-level pipe network association map. For example, the building branch pipe area corresponding to the primary associated pipe, the community sub-main pipe area corresponding to the secondary associated pipe, and the municipal main pipe area corresponding to the tertiary associated pipe.

[0040] The timing of this step is after the generation of the three-level pipe network association map, and the application scenario is the data collection stage for real-time monitoring of the water usage status of the pipe network. Specifically, the intelligent customer service system first locates the intelligent metering devices (such as flow meters, water meters) installed on each pipe according to the pipe ID and spatial coordinates in the three-level pipe network association map, and establishes a real-time data connection through Internet of Things communication protocols such as LoRa and NB-IoT. Then, according to the preset time granularity (such as 15 minutes as a time period), it periodically collects the flow data of each device, and calculates the water consumption of the corresponding pipe area in combination with the pipe diameter and length. For pipe areas without directly installed metering devices (such as secondary branch pipes), the system indirectly calculates through the difference between the upstream main pipe flow data and the downstream branch pipe flow data. For example, if the main pipe flow is 100 cubic meters per hour and the total known downstream branch pipe flow is 80 cubic meters per hour, then the water consumption of the intermediate branch pipe area is 20 cubic meters per hour. The collected data is denoised (such as removing out-of-range data and correcting sensor errors), and then stored in the time series database according to the pipe area and time stamp to form a water consumption dataset by time period. This dataset is not only used to monitor the current water usage situation in real time, but also can be used as historical data for training the water usage pattern model to provide a benchmark for anomaly detection.

[0041] S105. Determine the pipe areas with abnormal water usage in combination with the water consumption data by time period; Among them, the water consumption data by time period represents the water consumption statistical data of each pipe area at different time periods obtained through step S104, which is used to reflect the normal water usage patterns and abnormal fluctuations. The pipe area with abnormal water usage refers to the pipe area where the water consumption deviates from the normal fluctuation range, including situations where the water consumption suddenly surges, drops sharply, or continuously exceeds / falls below the threshold. For example, if the water consumption of a main pipe area during non-peak hours increases by 30% compared with the same period in history, it may indicate a pipe leak or illegal water use.

[0042] This step is executed after the collection of water consumption data in time periods is completed, and the application scenario is the stage of pipeline network anomaly detection based on historical data and real-time data. Specifically, the intelligent customer service system first establishes a water consumption baseline model for each pipeline area. By analyzing historical time-period data (such as the past 30 days), it calculates the average water consumption, standard deviation, and fluctuation range for each time period, and determines the threshold range of normal water consumption (such as the average value ± 2 times the standard deviation). For the time-period data collected in real time, the system calculates the deviation rate between the current water consumption and the baseline model for each time period. The formula is: Deviation rate = (Real-time water consumption - Average water consumption) / Average water consumption × 100%. When the deviation rate exceeds the preset threshold (such as ± 15%), an anomaly warning is triggered, and the corresponding pipeline area is marked as a potential anomaly area. To avoid misjudgment, the system also conducts multi-dimensional verification: ① Time dimension verification, checking whether the anomaly occurs continuously in multiple time periods (such as the deviation rate exceeding the threshold in 2 consecutive time periods); ② Space dimension verification, analyzing whether the water consumption data of adjacent pipeline areas is abnormally synchronized (such as when the main pipeline is abnormal, checking whether the downstream branch pipelines show synchronous flow anomalies); ③ Business logic verification, adjusting the threshold in combination with external factors such as holidays and weather (such as increasing the water consumption peak threshold by 10% during high-temperature periods in summer). After multi-layer verification, the pipeline areas with abnormal water consumption are finally determined and highlighted in red in the three-level pipeline network association map, providing a clear target for subsequent fault diagnosis.

[0043] In some embodiments, determining the pipeline areas with abnormal water consumption can be achieved in various ways: Optionally, a threshold detection method based on statistics is adopted. First, a normality test is performed on the historical time-period data of each pipeline area. If it conforms to a normal distribution, the 3σ principle (average value ± 3 times the standard deviation) is used to set the anomaly threshold; if it does not conform to a normal distribution, the quantile method (such as the 95% quantile) is used to set the upper threshold and the 5% quantile is used to set the lower threshold. For example, the historical hourly water consumption data of a branch pipeline in a certain community shows a skewed distribution, with the 95% quantile being 60 cubic meters per hour and the 5% quantile being 10 cubic meters per hour. Then, the time periods when the real-time water consumption exceeds 60 cubic meters or is lower than 10 cubic meters are determined as abnormal. Optionally, an LSTM anomaly detection model in machine learning is adopted. By training the model to learn the feature representation of normal water consumption patterns, an anomaly score is calculated for the real-time data, and when the score exceeds the threshold, it is determined as abnormal. For example, an LSTM model is used to train the time-period water consumption sequence of a certain main pipeline. The model predicts that the water consumption at the next moment is 80 cubic meters per hour, while the actual collected value is 120 cubic meters per hour. The anomaly score exceeds the threshold, and it is determined that the water consumption in this area is abnormal. It can be understood that dynamic thresholds can also be set in combination with expert experience. For example, during a fire drill, the water consumption threshold in the relevant area is temporarily increased to avoid false alarms, which is not limited here.

[0044] S106. Obtain the real-time hardware status data of the area of the abnormal water-using pipeline. The hardware status data at least includes continuous pressure fluctuation data, valve torque, switch frequency, pressure curve, and acoustic frequency spectrum signal; Among them, the real-time hardware status data represents the device operation parameters collected in real time by sensors installed in the area of the abnormal water-using pipeline, and is used to reflect the physical states of the pipeline and its accessory devices (such as valves and water pumps). The continuous pressure fluctuation data refers to the water pressure change data in the pipeline continuously collected by a pressure sensor, recorded in the form of a time series (such as 1 pressure value per second), and is used to detect abnormal pressure fluctuations (such as a sudden pressure drop may indicate a pipeline rupture). The valve torque refers to the torque value required during the opening and closing process of the valve, measured by a torque sensor, and is used to determine whether the valve is stuck or damaged (such as a sudden increase in torque may indicate that there is a foreign object blocking inside the valve body). The switch frequency refers to the number of times the valve is opened and closed within a unit time, recorded by a counter, and is used to monitor whether the valve operation frequency is abnormal (such as frequent opening and closing may cause wear of the valve flap). The pressure curve refers to a continuous curve plotted with time as the horizontal axis and pressure value as the vertical axis, and is used to visually display the pressure change trend (such as a continuous pressure drop may indicate a leak). The acoustic frequency spectrum signal refers to the acoustic signal generated by the fluid or mechanical vibration in the pipeline collected by an acoustic sensor, and the frequency distribution data obtained after Fourier transform, and is used to analyze the vibration characteristics (such as high-frequency acoustic waves may indicate a pipeline leak, and low-frequency acoustic waves may indicate a valve failure).

[0045] S107. Determine the fault type by combining the continuous pressure fluctuation data and the acoustic frequency spectrum signal; Among them, the continuous pressure fluctuation data represents the continuous sequence data of the water pressure in the pipeline changing with time collected in real time by a pressure sensor, and is used to reflect the dynamic change characteristics of the pressure. The acoustic frequency spectrum signal refers to the frequency distribution data of the vibration signal in the pipeline collected by an acoustic sensor and processed by Fourier transform, and is used to characterize the fluid state in the pipeline or the operation characteristics of mechanical components. Different fault types will correspond to specific frequency components (such as a pipeline rupture is often accompanied by high-frequency acoustic waves, and valve failures are mostly manifested as low-frequency vibrations). The fault type refers to the specific pipe network fault category determined according to the characteristics of pressure and acoustic data. In this application, it mainly includes three types: pipeline rupture, pipe network blockage, and valve failure. Each type corresponds to a unique multi-parameter combination feature.

[0046] This step is executed after obtaining the real-time hardware status data of the abnormal water-using pipeline area, and the application scenario is the intelligent fault type identification stage based on multi-parameter data analysis. Specifically, the intelligent customer service system first preprocesses the continuous pressure fluctuation data, removes high-frequency noise through moving average filtering, and calculates characteristic parameters such as the slope (i.e., the pressure drop rate), amplitude, and duration of the pressure curve. At the same time, it conducts frequency-domain analysis on the acoustic wave spectrum signal, extracts the amplitudes and energy ratios of the main frequency and harmonic components, and identifies the characteristic frequency intervals (such as the high-frequency band of 2000 - 4000 Hz and the low-frequency band of 50 - 200 Hz). Then, according to the preset fault determination rule base, it jointly matches the pressure and acoustic wave characteristics: Pipe rupture: When the pressure curve drops by more than the first pressure drop threshold (such as 0.1 MPa) within a set time (such as 5 minutes), and the deviation rate of the continuous pressure fluctuation data is greater than the first deviation rate threshold (such as 20%), and at the same time, the high-frequency components (>1500 Hz) in the acoustic wave spectrum are greater than the first frequency threshold (such as 50 dB) within a continuous time (such as 3 minutes), it is determined as a pipe rupture. High-frequency acoustic waves are generated by the impact of water jets on the pipe wall, and the sudden pressure drop reflects the pressure loss caused by fluid leakage.

[0047] Pipe network blockage: When the amplitude of the pressure curve is small but it remains in a low-pressure state continuously, the deviation rate is less than the second deviation rate threshold (such as -10%), the amplitude by which the valve torque exceeds the standard value is greater than the first valve torque amplitude (such as 30%), and the switching frequency increases abnormally (such as twice the daily average switching times), combined with the fact that there are fewer high-frequency components and a slight increase in low-frequency noise in the acoustic wave spectrum, it is determined as a pipe network blockage. Blockage leads to an increase in water flow resistance, the valve needs a greater torque to open, and the increase in switching frequency may be due to frequent attempts to clear the blockage.

[0048] Valve failure: When the pressure curve shows periodic fluctuations (such as pressure oscillations caused by pump start and stop), the amplitude is greater than the first pressure curve amplitude (such as 0.05 MPa), the proportion of low-frequency components (<200 Hz) in the acoustic wave spectrum is greater than the first acoustic wave spectrum proportion (such as 40%), and the deviation rate is less than the third deviation rate threshold (such as ±5%), it is determined as a valve failure. Low-frequency vibrations are mostly caused by wear, jamming, or poor sealing of valve components, and the pressure fluctuations are related to the valve opening and closing cycle.

[0049] The system makes a comprehensive judgment on multiple parameters through a fuzzy logic algorithm, allowing a certain degree of parameter tolerance (such as ±5% of the pressure drop threshold) to avoid misjudgment caused by single-parameter fluctuations. For example, if the pressure in a certain pipeline area drops by 0.09 MPa (close to the 0.1 MPa threshold), but the high-frequency energy of the acoustic wave continuously exceeds 55 dB, and the deviation rate reaches 22%, it is still determined as a pipe rupture.

[0050] S108. Generate a corresponding intelligent work order according to the fault type. The work order shall at least include the fault type, the geographical coordinates of the abnormal water - using pipeline area and the pipe network level, and the list of tools and accessories required for maintenance.

[0051] Among them, the fault type represents the specific fault category determined through step S107 (such as pipeline rupture, pipe network blockage, valve failure), which is used to guide the maintenance personnel to select a targeted disposal plan. The geographical coordinates refer to the precise location of the abnormal water - using pipeline area on the map (such as longitude and latitude), which are obtained through address resolution or GPS positioning to ensure that the maintenance personnel can quickly reach the site. The pipe network level refers to the level (primary, secondary, tertiary) of the faulty pipeline in the three - level pipe network association map, reflecting the importance of the pipeline and the maintenance priority (for example, the failure of the tertiary main pipeline needs to be processed first). The list of tools and accessories required for maintenance is a list of tools and spare parts pre - configured according to the fault type. For example, for pipeline rupture, welding equipment, leak - repair clamps and pipes of corresponding diameters need to be carried, and for valve failure, wrenches, sealing rings and spare valves need to be prepared.

[0052] The timing of this step is after the fault type is determined, and the application scenario is the execution stage of maintenance task generation and resource scheduling. Specifically, the intelligent customer service system first calls the corresponding work order template according to the fault type and automatically fills in the following information: Fault type: Directly reference the determination result of S107, such as "pipeline rupture".

[0053] Geographical coordinates and pipe network level: Extract the central coordinates (such as 116.4810°E, 39.9219°N) of the abnormal water - using area and the level label (such as "tertiary main pipeline") from the three - level pipe network association map, and mark the fault point in the map.

[0054] List of maintenance tools and accessories: Associate the fault type with tools and accessories through a rule mapping table. For example: Tools corresponding to pipeline rupture: Pipe cutter, electric welding machine, pressure test pump; Accessories: DN200 steel pipe, leak - repair glue, sealing ring.

[0055] Tools corresponding to pipe network blockage: Pipe dredger, endoscope; Accessories: dredging spring, filter screen.

[0056] Tools corresponding to valve failure: Socket wrench, torque wrench; Accessories: DN150 valve, sealing packing.

[0057] In addition, the work order can also automatically generate a link to the 3D pipe network model of the fault site, historical maintenance records (if any) and safety operation guidelines for the reference of maintenance personnel. After the work order is generated, the system automatically assigns it to the nearest and skill - matched personnel according to the current time (working hours / non - working hours) and the status of maintenance personnel (on duty / on leave), and notifies the recipient through text messages, APP push, etc.

[0058] In the embodiments of the present application, by adopting technical means such as using natural language processing algorithms to extract key data of repair information, constructing a three-level pipe network association map to integrate spatial and pipeline attributes, and performing correlation analysis on multi-source dynamic data (time-segmented water consumption data, hardware status data), the technical problems in the prior art of fragmented user repair data, insufficient visualization of pipe network structure, low analysis efficiency and poor positioning accuracy due to relying on manual experience for fault diagnosis are effectively solved. Furthermore, the technical effects of automatically mining fault correlation features from multi-dimensional data, constructing an accurate pipe network topology model, and realizing intelligent determination of fault types through data-driven algorithms are achieved, significantly improving the automation level of repair processing and the efficiency of fault handling.

[0059] In some embodiments, after the intelligent work order processing is completed, a series of quality inspection operations will be performed to comprehensively and objectively measure the performance of the intelligent customer service and take corresponding measures according to the evaluation results. First, the voice information of the user is obtained. After the user completes the repair and the work order processing is over, the intelligent customer service system will automatically collect the voice records during the interaction between the user and the intelligent customer service. These voice information contain the direct feedback of the user on the repair processing, which may cover various aspects such as the evaluation of the fault solution situation and the feeling about the customer service communication method. Then, the system uses automatic speech recognition technology to convert the voice information into text information. The core principle of this technology is to analyze and process the voice signal through acoustic models, language models, etc., and convert the sound signal into a text form that the computer can understand and process. For example, if the user says "Your repair speed is too slow. I've been waiting for a long time and no one has come to handle it", the speech recognition technology will accurately convert it into the corresponding text, providing a basis for subsequent analysis.

[0060] Then, the converted text information is input into the customer service quality inspection large model. This model is constructed through deep learning and uses multiple sensitive vocabulary sets and emotion vocabulary sets with quality inspection score annotations during prior training. The sensitive vocabulary set contains words that may imply problems with the service, such as "complaint", "terrible", etc.; the emotion vocabulary set covers various words expressing emotions, like "happy", "angry", etc. By learning a large amount of such annotated data, the model can analyze the input text information, judge the performance of the intelligent customer service during the repair process from multiple dimensions such as vocabulary usage and sentence structure, and then give a preliminary quality inspection score. For example, if negative sensitive words appear multiple times in the text information, the model will give a relatively low preliminary quality inspection score.

[0061] At the same time, the system will also combine voice information and use the emotion recognition model to determine the user's emotional information about the intelligent customer service handling of the repair report. The emotion recognition model is also based on deep learning and is obtained by training multiple voice information sets with emotional information annotations. It can recognize emotional features in speech, such as voice intonation, speech speed, volume changes, etc., to judge the user's emotional state, and can at least recognize three emotions: anger, satisfaction, and neutrality. Assuming that the user's tone is excited, the speed is fast, and the volume is loud, the emotion recognition model is likely to determine that the emotion is angry; if the voice is stable and the tone is relaxed, it may be determined to be satisfied or neutral.

[0062] Next, the preliminary quality inspection score is adjusted according to the results of the emotion recognition model. If the emotional information is anger, it means that the user is extremely dissatisfied with the service of the intelligent customer service. At this time, the preliminary quality inspection score will be deducted according to the preset deduction value to determine the final quality inspection score. For example, if the preliminary quality inspection score is 80 points and the preset deduction value is 20 points, then the final quality inspection score will become 60 points. If the emotional information is satisfied, it means that the user approves of the service. On the basis of the preliminary quality inspection score, the preset additional points will be added to determine the final quality inspection score. For example, if the preliminary score is 80 points and the preset additional points are 10 points, the final score will be 90 points. If the emotional information is neutral, it means that the user has no obvious likes or dislikes for the service, and the preliminary quality inspection score will be directly determined as the final quality inspection score.

[0063] Finally, the final quality inspection score will be compared with the set passing score threshold. If the final quality inspection score is lower than the set passing score threshold, it means that the service quality of the intelligent customer service does not meet the standard, and the system will trigger the intervention of manual customer service. With their professional knowledge and rich experience, manual customer service can communicate with users in a more in-depth and humane way, further understand users' needs and dissatisfaction, solve users' problems in a timely manner, make up for the shortcomings of intelligent customer service, improve user satisfaction, and ensure the high-quality completion of repair processing services.

[0064] After combining the above content, the following is a more detailed description of the process of the method provided by this implementation. Figure 2 , is another flow chart of the repair processing method based on intelligent customer service in an embodiment of the present application.

[0065] S201, based on the geographic location information, generating a minimum closed irregular space envelope by a minimum circumscribed polygon algorithm; Among them, the geographical location information represents the longitude and latitude coordinates or detailed address data of the associated repair points extracted in step S102, which is used to locate the spatial position set where the fault occurs. For example, the coordinate point set corresponding to multiple repair points is {(x1, y1), (x2, y2),..., (xn, yn)}. The minimum circumscribed polygon algorithm refers to an algorithm that generates the smallest closed polygon containing all points by calculating the convex hull or concave hull of a planar point set. The minimum closed irregular spatial envelope refers to an irregular polygon area generated by the algorithm that tightly wraps all associated repair points. Its number of sides and shape are determined by the distribution of the repair points. For example, when the repair points are linearly distributed, a long and narrow polygon is generated, and when they are scattered, an irregular convex polygon is generated.

[0066] The timing of this step is after obtaining the geographical location information of the associated repair points, and the application scenario is the initial definition stage of the pipe network analysis scope. Specifically, the intelligent customer service system first converts each repair point in the associated repair information set into geographical coordinates to form a coordinate point set. Then, it calls the minimum circumscribed polygon algorithm (such as the convex hull algorithm based on the Qhull library) to process the coordinate point set: the algorithm traverses all points, finds the leftmost point as the starting point, and sequentially selects the edge points in the counterclockwise direction to ensure that all repair points are located inside or on the polygon. Finally, a minimum closed polygon composed of a vertex sequence is generated. This envelope line closely follows the distribution edge of the repair points, eliminating redundant space. For example, if 5 repair points in a certain community are L-shaped, the convex hull algorithm will generate a pentagon containing these 5 points, accurately reflecting the spatial boundary of the fault concentration area. The generated polygon is stored in the GeoJSON format, containing attributes such as vertex coordinates, area, and perimeter, providing a benchmark for subsequent expansion of the analysis scope.

[0067] S202: Extend outward by a set distance based on the minimum closed irregular spatial envelope to generate a standard irregular spatial envelope; Among them, the set distance represents the preset spatial expansion range according to the pipe network analysis requirements, which is used to include the pipe areas that may be affected around the repair points in the analysis scope. The standard irregular spatial envelope refers to a polygon area formed by uniformly expanding the set distance on the basis of the minimum envelope. Its shape is similar to the minimum envelope but has a larger range. For example, if the original minimum envelope is a pentagon, it becomes a pentagon with the side lengths extended outward after extension, ensuring coverage of the pipe network around the repair points.

[0068] This step is executed after the generation of the minimum closed irregular space envelope, and the application scenario is the stage of expanding the pipeline network analysis scope to cover potential fault-related areas. Specifically, the intelligent customer service system first performs parallel extrapolation on each edge of the minimum envelope by a set distance, and at the same time performs rounding processing on the vertices of the polygon (to avoid calculation errors caused by acute angles) to generate an expanded polygon. For example, if the original minimum envelope is a square with a side length of 100 meters and the set distance is 50 meters, then the expanded one is a square with a side length of 200 meters, and the area is expanded to 4 times the original. The expanded envelope is realized through spatial topology operations (such as buffer analysis) to ensure that all spaces within the set distance from the repair point are covered. The generated standard envelope is used for subsequent pipeline data retrieval. For example, query all pipelines within this envelope from the GIS system to ensure that the analysis scope neither misses related pipelines nor expands excessively resulting in data redundancy.

[0069] S203. Obtain all pipeline data in the standard irregular space envelope. The pipeline data includes at least pipeline ID, pipeline spatial coordinates, and pipe diameter. Among them, the pipeline data represents the pipeline attribute data stored in the pipeline network geographic information system (GIS), which is used to describe the physical characteristics and spatial positions of pipelines. The pipeline ID is the code that uniquely identifies each pipeline, such as "DN200-01-001", which is used for data association and tracking. The pipeline spatial coordinates refer to the longitude and latitude coordinates of the starting point and ending point of the pipeline. For example, the starting point (116.4810°E, 39.9219°N) and the ending point (116.4820°E, 39.9225°N), which are used to draw the pipeline route on the map. The pipe diameter refers to the nominal diameter of the pipeline (such as DN50, DN200), which is used to distinguish pipeline levels and water conveyance capabilities. Generally, the larger the pipe diameter, the higher the pipeline level (for example, the diameter of the tertiary main pipeline ≥ DN200).

[0070] This step is executed after the generation of the standard irregular space envelope, and the application scenario is the stage of retrieving pipeline data in the target area from the GIS system. Specifically, the intelligent customer service system sends a request to the GIS database through the spatial query interface to retrieve all pipeline records that are completely or partially located within the standard envelope. The query conditions include: the pipeline geometry intersects with the envelope polygon, and attributes such as pipeline ID, starting point coordinates, ending point coordinates, pipe diameter, material, and burial depth are returned. For example, if the coverage area of a certain standard envelope is 1 square kilometer, the query results may include 100 primary pipelines (DN20 - DN50), 50 secondary pipelines (DN80 - DN150), and 20 tertiary pipelines (DN200 - DN400). The obtained data is stored as a structured data set, such as a CSV file or a JSON array, after deduplication and format conversion, for subsequent atlas construction.

[0071] S204. Construct a three - level pipe network association graph based on the standard irregular space envelope and pipe data.

[0072] Among them, the three - level pipe network association graph refers to a pipe network topology model with the standard envelope as the spatial scope and the pipe level as the structure, which is used to visually display the association relationship between the repair points and pipes at all levels. The primary associated pipe refers to the branch pipe directly connected to the repair point, the secondary associated pipe refers to the branch pipe connected to the primary pipe, and the tertiary associated pipe refers to the main pipe (pipe diameter ≥ preset threshold) connected to the secondary pipe. The graph represents the pipe connection relationship in the form of nodes and edges. For example, the nodes are valves and water meter wells, and the edges are pipes. Different - level pipes are distinguished by different colors or line types (such as the primary pipe is a blue dashed line and the tertiary pipe is a red solid line).

[0073] The timing of this step is after obtaining the standard envelope and pipe data, and the application scenario is the visualization modeling stage of the pipe network topology structure. Specifically, the intelligent customer service system first divides the levels according to the pipe diameter and connection relationship: the pipe with the closest spatial distance to the repair point coordinate in space is set as the primary associated pipe (such as the branch pipe in the user's home). Then, the secondary pipe (the upstream branch pipe of the primary pipe) and the tertiary pipe (the upstream main pipe of the secondary pipe) are recursively searched through the pipe connection relationship table (storing the upstream and downstream nodes of the pipe). Then, a graph is constructed using a graph database or a GIS visualization tool: each pipe is used as an edge, and the nodes are the pipe endpoints, and attributes such as pipe ID, pipe diameter, and level are marked; the standard envelope polygon and the repair point coordinates are overlaid on the graph, and the fault location is marked with an icon (such as a water drop icon indicating a leak repair). For example, a certain repair point is located on a DN50 branch pipe (primary), which is connected to a DN100 branch pipe (secondary), and then connected to a DN300 main pipe (tertiary). The hierarchical relationship of the tertiary pipe is shown in the graph, and a repair icon is marked at the position of the branch pipe.

[0074] In the embodiment of the present application, since the minimum circumscribed polygon algorithm is used to generate the minimum closed irregular space envelope and extend it outward by a set distance to generate the standard irregular space envelope, the problem that the pipe network analysis range is fuzzy and the fault - associated area cannot be accurately defined in the prior art is effectively solved. Furthermore, the technical effect of accurately framing the distribution range of the repair points through a geometric algorithm and covering the potentially affected pipes through extension is achieved, providing a clear spatial reference for subsequent pipe data retrieval and level division.

[0075] In some embodiments, after the step of generating a corresponding intelligent work order according to the fault type, the following steps are further included: obtaining the current time information; if the time information is within the set working hours, obtaining the ability information and location information of the maintenance personnel; combining the ability information and the location information, and determining the maintenance personnel with the closest geographical coordinate distance to the abnormal water - using pipeline area and matching ability according to the fault type. Among them, the current time information represents the system time obtained in real - time by the intelligent customer service system, which is used to determine whether it is within the set working time period (such as 8:00 - 18:00 from Monday to Friday). The ability information of the maintenance personnel refers to the set of skill tags of the maintenance personnel, which is generated through data such as historical maintenance records and training certifications, and is used to represent the professional skill range of the maintenance personnel, such as "pipe welding", "valve repair", "GIS operation", etc. The location information refers to the real - time geographical location coordinates of the maintenance personnel obtained through GPS, Beidou or Internet of Things base stations, which is used to calculate the spatial distance from the fault point. For example, the current coordinates of a certain maintenance personnel are (116.4815°E, 39.9220°N). The geographical coordinates refer to the central longitude and latitude of the abnormal water - using pipeline area, which is obtained through the three - level pipe network map. For example, the coordinates of the fault area are (116.4800°E, 39.9200°E). The timing of this step is after the intelligent work order is generated, and the application scenario is the real - time scheduling stage of maintenance resources. Specifically, the intelligent customer service system first obtains the current system time (such as 10:30 on April 26, 2025), and determines whether it belongs to the set working time (such as 8:00 - 18:00). If it is within the working time, the system obtains the real - time location information of all on - duty maintenance personnel through the mobile terminal APP or in - vehicle positioning device, and at the same time retrieves their ability tags from the personnel management database (such as the ability tags of maintenance personnel A are "pipe rupture repair" and "tertiary pipe network maintenance"). Then, the system filters out the subset of maintenance personnel with corresponding ability tags based on the fault type (such as "pipe rupture"), for example, filtering out the personnel whose ability tags include "pipe welding" and "leak repair operation". Next, the spatial distance calculation algorithm is used to calculate the straight - line distance between each qualified maintenance personnel and the fault point, and the person with the closest distance is preferentially selected. If there are multiple persons with the same distance, their current task loads (such as the number of unfinished work orders) are further compared, and the person with the lowest load is selected. For example, maintenance personnel B is 1.2 kilometers away from the fault point, holds the "pipe rupture repair" tag and has no pending work orders at present, and the system determines him as the optimal candidate and pushes a task notification through the work order system.

[0076] In some embodiments, if the time information is outside the set working hours, the ability information and home address information of the maintenance personnel are obtained; by combining the ability information and home address information, the maintenance personnel with the closest geographical coordinate distance to the water-using abnormal pipeline area and matching ability are determined according to the fault type. Herein, outside the set working hours refers to the time range exceeding the preset working period, such as at night (18:00 - 8:00 the next day), on weekends or holidays. The home address information refers to the coordinates of the permanent residence address registered by the maintenance personnel in the system, which is used to replace the real-time positioning data. The geographical coordinates are also the central coordinates of the fault area, maintaining the consistency of the spatial reference. The timing of this step is after the intelligent work order is generated and the system determines that the current time belongs to a non-working period (such as 22:00 on April 26, 2025). Specifically, the intelligent customer service system first confirms that the current time is not within the working hours, and then retrieves the home address information of the maintenance personnel from the personnel database (which needs to be collected and encrypted in advance). Then, based on the fault type, a subset of personnel with corresponding ability labels is screened (for example, for "valve failure", those with "valve repair" and "seal replacement" label holders need to be screened), and then the distance between each personnel's home address and the fault point is calculated through a preset formula, and the person with the closest distance is selected. For example, the home address of maintenance personnel D is 3.5 kilometers away from the fault point, the ability label includes "valve torque detection", and the person is on standby during non-working hours, and the system determines him as the priority dispatching object. In addition, the system also needs to consider the standby status of the maintenance personnel (such as "responsive" or "unresponsive"), and confirm whether they can accept the order through double verification by text message or phone call to avoid task delays caused by poor communication.

[0077] The intelligent customer service system in the embodiments of the present invention application will be described from the perspective of hardware processing. Please refer to Figure 3 , which is a schematic structural diagram of an entity device of the intelligent customer service system in the embodiments of the present application.

[0078] It should be noted that Figure 3 the structure of the intelligent customer service system shown is only an example, and should not bring any restrictions to the functions and usage scopes of the embodiments of the present invention.

[0079] Such as Figure 3As shown, the intelligent customer service system includes a Central Processing Unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the Read-Only Memory (ROM) 302 or the program loaded from the storage section 308 into the Random Access Memory (RAM) 303, such as executing the methods described in the above embodiments. In the RAM 303, various programs and data required for system operation are also stored. The CPU 301, ROM 302, and RAM 303 are connected to each other via a bus 304. An Input / Output (I / O) interface 305 is also connected to the bus 304.

[0080] The following components are connected to the I / O interface 305: an input section 306 including an audio input device, a button switch, etc.; an output section 307 including a Liquid Crystal Display (LCD), an audio output device, an indicator light, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that a computer program read from it can be installed into the storage section 308 as needed.

[0081] Specifically, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network via the communication section 309, and / or installed from the removable medium 311. When the computer program is executed by the Central Processing Unit (CPU) 301, various functions defined in the present invention are executed.

[0082] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fibers, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium that contains or stores a program, which can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0083] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. Among them, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the above-mentioned module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings.

[0084] Specifically, the intelligent customer service system of this embodiment includes a processor and a memory. A computer program is stored on the memory. When the computer program is executed by the processor, the repair handling method based on intelligent customer service provided in the above embodiment is implemented.

[0085] On the other hand, the present invention also provides a computer-readable storage medium, which may be included in the intelligent customer service system described in the above embodiment; or it may exist alone and not be assembled into the intelligent customer service system. The above storage medium carries one or more computer programs. When the above one or more computer programs are executed by a processor of the intelligent customer service system, the intelligent customer service system is enabled to implement the repair handling method based on intelligent customer service provided in the above embodiment.

[0086] As described above, the above embodiments 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 foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the various embodiments of the present application.

[0087] As used in the foregoing embodiments, depending on the context, the term "when" may be construed to mean "if", or "after", or "in response to determining", or "in response to detecting". Similarly, depending on the context, the phrase "when determining" or "if (the stated condition or event) is detected" may be construed to mean "if determined", or "in response to determining", or "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0088] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the foregoing embodiments can be implemented by a computer program instructing relevant hardware. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the foregoing method embodiments. The foregoing storage media include: various media such as ROM or random access memory RAM, magnetic disks, or optical discs that can store program code.

Claims

1. A repair handling method based on intelligent customer service, applied to an intelligent customer service system, characterized in that The method includes: obtaining user repair information and determining an associated repair information set; extracting the geographical location information and fault phenomenon keywords of the associated repair points in the associated repair information set through a natural language processing algorithm; combining the geographical location information to generate a three-level pipe network association map covering all the associated repair points and extending outward by a set distance. The three-level pipe network association map includes primary associated pipes, secondary associated pipes, and tertiary associated pipes. The primary associated pipes are the pipes directly connected to the repair points, the secondary associated pipes are the branch pipes connected to the primary associated pipes, and the tertiary associated pipes are the main pipes connected to the secondary associated pipes and with a pipe diameter greater than or equal to a preset pipe diameter threshold; obtaining the sub-period water consumption data of each pipe area in the three-level pipe network association map in real time through intelligent metering devices; determining the pipe areas with abnormal water use in combination with the sub-period water consumption data; obtaining the real-time hardware status data of the pipe areas with abnormal water use, where the real-time hardware status data at least includes continuous pressure fluctuation data, valve torque, switch frequency, pressure curve, and acoustic frequency spectrum signal; determining the fault type in combination with the real-time hardware status data; generating a corresponding intelligent work order according to the fault type, where the work order at least includes the fault type, the geographical coordinates and pipe network level of the pipe area with abnormal water use, and the list of tools and accessories required for repair.

2. The method according to claim 1, characterized in that, In the step of obtaining user repair information and determining an associated repair information set, it specifically includes: extracting the time data and community information in the user repair information; extracting the fault keywords in the user repair information; if the similarity of the fault keywords reaches a set similarity threshold, and the community information is the same or within a set distance range, and the time data is within a set time range, then the corresponding user repair information is merged and determined as an associated repair information set.

3. The method according to claim 1, characterized in that In the step of combining the geographical location information to generate a three-level pipe network association map covering all the associated repair points and extending outward by a set distance, it specifically includes: generating a minimum closed irregular spatial envelope based on the geographical location information through the minimum bounding polygon algorithm; extending outward by a set extension distance with the minimum closed irregular spatial envelope as a reference to generate a standard irregular spatial envelope; obtaining all pipe data in the standard irregular spatial envelope, where the pipe data at least includes pipe ID, pipe spatial coordinates, and pipe diameter; constructing a three-level pipe network association map according to the standard irregular spatial envelope and the pipe data.

4. The method according to claim 1, characterized in that, In the step of determining the pipe areas with abnormal water use in combination with the sub-period water consumption data, it specifically includes: calculating the deviation rate in real time in combination with the sub-period water consumption data; if the deviation rate exceeds a preset deviation rate threshold, then determine the pipe area corresponding to the deviation rate as the pipe area with abnormal water use.

5. The method according to claim 1, characterized in that, In the step of determining the fault type in combination with the real-time hardware status data, it specifically includes: If the deviation rate is greater than the first deviation rate threshold, and the pressure curve drops by more than the first pressure drop threshold within the set time, and the acoustic wave spectrum is greater than the first frequency within the set continuous time, then it is determined that the fault type is pipeline rupture; If the deviation rate is less than the second deviation rate threshold, and the amplitude by which the valve torque exceeds the standard value is greater than the first valve torque amplitude, and the switching frequency increases abnormally, then it is determined that the fault type is pipe network blockage; If the deviation rate is less than the third deviation rate threshold, and the amplitude of the pressure curve is greater than the first pressure curve amplitude, and the proportion of low-frequency components in the acoustic wave spectrum is greater than the first acoustic wave spectrum proportion, then it is determined that the fault type is valve failure.

6. The method according to claim 1, characterized in that After the step of generating a corresponding intelligent work order according to the fault type, it further includes: Obtain the current time information; If the time information is within the set working hours, then obtain the ability information and location information of the maintenance personnel; Combining the ability information and the location information, determine the maintenance personnel with the closest geographical coordinate distance to the water-abnormal pipeline area and matching ability according to the fault type; If the time information is outside the set working hours, then obtain the ability information and home address information of the maintenance personnel; Combining the ability information and the home address information, determine the maintenance personnel with the closest geographical coordinate distance to the water-abnormal pipeline area and matching ability according to the fault type.

7. The method according to claim 1, characterized in that After the step of generating a corresponding intelligent work order according to the fault type, it further includes: Obtain the voice information of the user, and convert the voice information into text information through automatic speech recognition technology; Input the text information into the customer service quality inspection large model to determine the preliminary quality inspection score. The customer service quality inspection large model is constructed through deep learning in advance according to multiple sensitive vocabulary sets and emotion vocabulary sets with quality inspection score annotations; Combining the voice information, determine the emotion information of the user towards the intelligent customer service handling of the repair through the emotion recognition model. The emotion recognition model is constructed through deep learning in advance through multiple voice information sets with emotion information annotations. The emotion information at least includes anger, satisfaction, and neutral; If the emotion information is anger, then subtract points according to the preset deduction value based on the preliminary quality inspection score to determine the final quality inspection score; If the emotion information is satisfaction, then add points according to the preset bonus value based on the preliminary quality inspection score to determine the final quality inspection score; If the emotion information is neutral, then determine the preliminary quality inspection score as the final quality inspection score; If the final quality inspection score is lower than the set passing score threshold, then trigger the intervention of the human customer service.

8. An intelligent customer service system, characterized in that, The intelligent customer service system includes: one or more processors and a memory; the memory is coupled to the one or more processors. The memory is used to store computer program code, and the computer program code includes computer instructions. The one or more processors call the computer instructions to enable the intelligent customer service system to execute the method according to any one of claims 1-7.

9. A computer-readable storage medium, comprising instructions, characterized in that, When the instruction runs on the intelligent customer service system, the intelligent customer service system is caused to execute the method described in any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product runs on the intelligent customer service system, the intelligent customer service system is caused to execute the method described in any one of claims 1-7.

Images