Water supply and drainage pipe network inspection order dispatching method and system

Abstract

The invention relates to the technical field of intelligent water affairs, in particular to a water supply and drainage pipe network inspection order dispatching method and system. The method comprises the steps of obtaining historical inspection data of a water supply and drainage pipe network; identifying dynamic inspection behavior information from the historical inspection data by adopting a data mining algorithm; constructing an inspection task evaluation index system based on the identified dynamic inspection behavior information; acquiring water supply and drainage pipe network monitoring data and inspector load data in real time; extracting required indexes from the monitoring data of the water supply and drainage pipe network and the load data of the inspector based on an inspection task evaluation index system; dynamically adjusting the priority of each inspection task and an order dispatching algorithm based on the extracted indexes; and order dispatching of the inspection tasks is carried out based on the priority of each inspection task and the order dispatching algorithm, so that the overall effect and the flexible adjustment capability of the inspection order dispatching of the water supply and drainage pipe network are improved.

Description

Technical Field

[0001] The present invention relates to the field of smart water technology, and in particular to a method and system for dispatching inspections of water supply and drainage networks. Background Art

[0002] With the acceleration of urbanization, the water supply and drainage pipeline network, as an important part of urban infrastructure, its safe and efficient operation has an important impact on the quality of life of urban residents. Therefore, it is necessary to inspect and maintain the water supply and drainage pipeline network to promptly discover or solve problems such as aging, damage, leakage, etc. of the water supply and drainage pipeline network to ensure the normal operation of the pipeline system.

[0003] The traditional method of dispatching inspection orders for water supply and drainage networks often relies on manual labor. Dispatchers manually dispatch orders based on the task information at hand, the inspectors' situations, and personal preferences. Dispatching decisions are mainly based on the dispatchers' experience and intuition rather than data-driven analysis. As a result, inspection dispatching is inefficient, making it difficult to achieve optimal resource allocation, and unable to respond quickly to emergencies, affecting the timely completion of inspection tasks.

[0004] For this reason, the current inspection and dispatching methods of water supply and drainage networks are gradually developing towards digitalization and intelligence. For example, Chinese patent CN116029695A, rail transit inspection work order processing method, device, electronic equipment and storage medium, obtains the first inspection work order, obtains the dispatching time, the equipment type of the equipment to be inspected, the inspection date and the inspection duration, and then can obtain the duty personnel information and unfinished work order statistics at the dispatching time, and then can screen the duty personnel based on the equipment type, inspection date, inspection duration, duty personnel information and unfinished work order statistics to determine the order receiving personnel information, the order receiving personnel information can represent the duty personnel who handles the first inspection work order, and then can dispatch the first inspection work order based on the order receiving personnel information, to a certain extent, realize the automatic dispatching of inspection work orders, and improve the efficiency of dispatching work orders.

[0005] However, with the expansion of the inspection area of ​​the water supply and drainage network and the increase in the number of inspection tasks, the simple dispatching method of the existing technology is mainly based on the pre-set inspection task work order and static information to make dispatching decisions. It lacks an intelligent decision-making mechanism, and it is difficult to achieve dynamic optimization of the inspection dispatching method. It lacks flexible adjustment capabilities and a comprehensive understanding of the characteristics of the inspection tasks. When the inspection task volume is large or the number of inspectors is insufficient, it is often impossible to find the global optimal solution, resulting in waste of resources and time delays. Summary of the invention

[0006] In response to the above technical problems, the present invention proposes a water supply and drainage network inspection dispatching method and system, aiming to improve the overall effect and flexible adjustment capability of water supply and drainage network inspection dispatching.

[0007] In a first aspect, the present application provides a method for dispatching inspection orders for a water supply and drainage network, comprising the following steps: Step 102, obtaining historical inspection data of the water supply and drainage network; Step 104, using a data mining algorithm to identify dynamic inspection behavior information from historical inspection data; Step 106, constructing an inspection task evaluation index system based on the identified dynamic inspection behavior information; Step 108, obtaining water supply and drainage network monitoring data and inspector load data in real time; Step 110, based on the inspection task evaluation index system, extract the required indicators from the water supply and drainage network monitoring data and the inspector load data; Step 112, dynamically adjusting the priority of each inspection task and the dispatching algorithm based on the extracted indicators; Step 114: dispatching inspection tasks based on the priorities of the inspection tasks and the dispatching algorithm.

[0008] In some embodiments, in step 104, a data mining algorithm is used to identify dynamic inspection behavior information from historical inspection data, including: Step 202, using an association rule mining algorithm to perform association rule mining on the historical inspection data to obtain a number of item sets and association rules of the historical inspection data; Step 204, setting a support threshold and a confidence threshold, and screening out frequent item sets and strong association rules from a number of item sets and association rules based on the set support threshold and confidence threshold; Step 206, determining representative dynamic inspection behaviors based on frequent item sets and strong association rules; Step 208: identifying dynamic inspection behavior information from historical inspection data according to the determined dynamic inspection behavior.

[0009] In some embodiments, step 202 uses an association rule mining algorithm to perform association rule mining on historical inspection data to obtain several item sets and association rules of the historical inspection data, including: Clustering algorithm is used to cluster historical inspection data; Analyze the feature distribution in different clusters, identify inspection data with similar features, and aggregate the inspection data into several types of inspection patterns; An association rule mining algorithm is used to mine association rules for historical inspection data of various inspection modes, and several item sets and association rules of historical inspection data of various inspection modes are obtained.

[0010] In some embodiments, in step 104, using a data mining algorithm to identify dynamic inspection behavior information from historical inspection data also includes: Periodically obtain the latest inspection data of the water supply and drainage network; Based on the latest inspection data of the water supply and drainage network, steps 202 to 206 are executed to update representative dynamic inspection behaviors.

[0011] In some embodiments, step 112, dynamically adjusting the priority of each inspection task and the dispatching algorithm based on the extracted indicators, includes: Step 302, prioritizing each inspection task based on the extracted indicators; Step 304: Select a corresponding dispatching algorithm based on the priority ranking of each inspection task. The dispatching algorithm includes a greedy algorithm and an ant colony algorithm.

[0012] In some embodiments, the indicators include the number of inspectors required for the task, the location, workload, availability, and estimated occupancy time of the resources of the inspectors. In step 114, the inspection task dispatching is performed based on the priority of each inspection task and the dispatching algorithm, including: Step 402: when the determined dispatching algorithm is a greedy algorithm, a task list to be dispatched is created; Step 404, selecting the inspection task with the highest priority from the task list, and for the current inspection task, calculating the distance between each inspector and the inspection task area based on the position of the inspector; Step 406, calculating a comprehensive score of each inspector for the current inspection task based on the distance, the inspector's workload, availability, and the estimated occupancy time of the resource; Step 408, based on the comprehensive scores of each inspector and the number of inspectors required for the task, determine the inspector that matches the current inspection task and complete the dispatch of the current inspection task; Step 410, remove the current inspection task from the task list and update the inspector load data; Step 412, repeat steps 404 to 410 until there are no inspection tasks to be dispatched in the task list.

[0013] In some embodiments, the indicators include the number of inspectors required for the task, the location, workload, availability, and estimated occupancy time of the resources of the inspectors. In step 114, the inspection task dispatching is performed based on the priority of each inspection task and the dispatching algorithm, including: Step 502: When the determined dispatching algorithm is the ant colony algorithm, each inspection task is regarded as a node, and each inspector is regarded as a moving ant; Step 504, setting the parameters of the ant colony algorithm, initializing the pheromone matrix, and setting the same initial value for the pheromone concentration on all paths; Step 506, based on the priority of each inspection task, the position, workload, availability, estimated occupancy time of the resource, and the pheromone concentration on the path, the probability of each ant selecting each inspection task is calculated; Step 508, based on the probability of each ant selecting each inspection task and the number of inspectors required for the task, determine the inspection task that matches each inspector, and after each ant completes a task matching, update the pheromone concentration on the corresponding path; Step 510: After all ants have completed one iteration, the pheromone concentration is globally updated, and the optimal dispatching and scheduling scheme between inspectors and inspection tasks in the current iteration is recorded; Step 512, repeating steps 506 to 510 to iterate until a predetermined termination condition is met, and outputting the optimal dispatching and scheduling plan between the inspectors and the inspection tasks.

[0014] In some embodiments, after step 114, the method further includes: Step 602, obtaining the inspection point information in each inspection task; Step 604, each inspection point is regarded as a node, each inspector is regarded as a moving ant, and the ant colony algorithm is used to calculate the probability of each ant selecting the next node at the current node; Step 606, based on the probability of each ant selecting the next node at the current node, determine the inspection path planning of each inspector, and after each ant completes an inspection path planning, update the pheromone concentration on the corresponding path; Step 608, when all ants have completed one iteration, the pheromone concentration is globally updated, and the optimal path solution between the inspector and the inspection point in the current iteration is recorded; Step 610, repeating steps 604 to 608 for iteration until a predetermined termination condition is met, and outputting the optimal path plan between the inspector and the inspection point.

[0015] In some embodiments, in step 108, real-time acquisition of water supply and drainage network monitoring data and inspector load data includes: GIS technology is used to collect the geographic information of the water supply and drainage network in real time, and IoT technology is used to collect the operation data of the water supply and drainage network in real time. The geographic information and operation data of the water supply and drainage network constitute the monitoring data of the water supply and drainage network; Receive load data submitted by inspectors.

[0016] In the second aspect, the present application provides a water supply and drainage network inspection dispatching system, including: The data acquisition module is used to obtain the historical inspection data of the water supply and drainage network, and obtain the monitoring data of the water supply and drainage network and the inspector load data in real time; An information identification module is used to identify dynamic inspection behavior information from historical inspection data using a data mining algorithm; The indicator system construction module is used to construct the inspection task evaluation indicator system based on the identified dynamic inspection behavior information; The indicator extraction module is used to extract the required indicators from the water supply and drainage network monitoring data and the inspector load data based on the inspection task evaluation indicator system; Dynamic adjustment module, used to dynamically adjust the priority of each inspection task and the dispatching algorithm based on the extracted indicators; The dispatch scheduling module is used to dispatch inspection tasks based on the priority of each inspection task and the dispatch scheduling algorithm.

[0017] The beneficial technical effects of the present invention include at least: 1. Adopt the inspection dispatch method and system of water supply and drainage network. First, the dynamic inspection behavior information is identified from the historical inspection data through data mining algorithm, which can more comprehensively understand the characteristics of the inspection task and the working status of the inspectors. Secondly, based on the identified dynamic inspection behavior information, the inspection task evaluation index system is constructed to provide scientific quantitative standards for the priority sorting and resource allocation of inspection tasks. Then, the water supply and drainage network monitoring data and inspector load data are obtained in real time to ensure that the dispatch decision is based on the latest information, so that the inspection dispatch system can respond to emergencies quickly. Then, the required indicators are extracted from the real-time data, and the dynamic inspection task evaluation index system is constructed based on the extracted indicators. Adjusting the priority of each inspection task and the dispatching algorithm greatly improves the flexible adjustment capability of the inspection dispatching of the water supply and drainage network compared to the existing technology that makes dispatching decisions based on pre-set inspection task work orders and static information. This solution that dynamically adjusts the priority and dispatching algorithm based on real-time data can adapt to the ever-changing environment and needs, and more accurately match inspectors with inspection tasks, thereby more reasonably allocating inspection resources. Finally, dispatching inspection tasks based on the priority of each inspection task and the dispatching algorithm improves the overall effect of the inspection dispatching of the water supply and drainage network, and provides strong support for water companies to achieve intelligent management and sustainable development.

[0018] 2. By using the association rule mining algorithm to identify representative dynamic inspection behavior information from historical inspection data, it is possible to more comprehensively and accurately identify the dynamic inspection behavior information that affects the inspection task dispatching, providing a more accurate and reliable basis for the subsequent inspection dispatching. At the same time, by periodically obtaining the latest inspection data to update the representative dynamic inspection behavior, it is ensured that the determined dynamic inspection behavior can more accurately reflect the real-time status and dynamic inspection needs of the water supply and drainage network. Moreover, this method emphasizes data-based, extracting valuable rules through data mining technology, reducing subjective judgment, and greatly improving the scientific nature of decision-making; 3. By real-time monitoring of water supply and drainage network monitoring data and inspector load data, and dynamically adjusting the priority of each inspection task based on the changes in indicators extracted from real-time data, it can adapt to the ever-changing actual conditions of the network and the working environment, improve the flexible adjustment ability of water supply and drainage network inspection dispatching, ensure that emergency situations can be responded to quickly, and at the same time, more reasonably dispatch inspection tasks while considering the workload of inspectors, and then select the corresponding dispatching algorithm according to the priority ranking of each inspection task after adjustment, so as to provide more scientific decision support for dispatching, improve the accuracy of decision-making, and thus improve the overall effect of water supply and drainage network inspection dispatching; 4. In scenarios with clear priorities and simple dispatching procedures, the local optimal solution is usually consistent with the global optimal solution. Therefore, this application uses a greedy algorithm to dispatch inspection tasks. According to the priority ranking of each inspection task, the current optimal inspector is directly selected for dispatching for each inspection task based on the current resource situation. This does not involve complex calculation processes and can make decisions more quickly while ensuring the overall effect of the inspection dispatching of the water supply and drainage network. It is more suitable for the dispatching of inspection tasks that require real-time response or in emergency situations. 5. In complex dispatching scenarios where priorities are unclear or where there are multiple paths and multiple tasks, the ant colony algorithm is used to dispatch inspection tasks. This can adapt to the changes in the priorities and complexity of each inspection task, and dynamically adjust the search direction through the update of pheromones, making the dispatching of inspection tasks more flexible. The positive feedback mechanism of pheromones and the random search strategy can effectively avoid premature convergence to the local optimal solution. At the same time, the workload balance of inspectors is considered to avoid the situation where some inspectors are overworked while other inspectors are relatively idle, providing a more comprehensive and robust dispatching solution, which is more suitable for dispatching inspection tasks that need to handle complex and dynamically changing tasks. 6. By further adopting the ant colony algorithm to achieve the optimal path planning between the inspectors and the inspection points in the inspection area, and dynamically adjust the path planning from a global perspective based on real-time data to adapt to changes in the inspection area, it can reduce the inspectors' moving distance and time waste during the inspection process, help to more reasonably allocate inspection resources, and thus improve the overall inspection efficiency.

[0019] Other features and advantages of the present invention will be disclosed in detail in the following specific embodiments and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The present invention will be further described below in conjunction with the accompanying drawings: Figure 1 The present invention is a flowchart of a method for dispatching inspection orders for water supply and drainage networks according to an embodiment of the present invention.

[0021] Figure 2 It is a schematic diagram of the structure of the water supply and drainage network inspection dispatching system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The technical solutions of the embodiments of the present invention are explained and described below in conjunction with the drawings of the embodiments of the present invention, but the following embodiments are only preferred embodiments of the present invention, not all. Based on the embodiments in the implementation mode, other embodiments obtained by those skilled in the art without creative work are all within the protection scope of the present invention.

[0023] In the following description, terms such as "inside", "outside", "up", "down", "left", "right", etc. that indicate directions or positional relationships are only used to facilitate the description of the embodiments and simplify the description, and do not indicate or imply that the referred device or element must have a specific direction, be constructed and operate in a specific direction. Therefore, it should not be understood as a limitation of the present invention.

[0024] Please see attached Figure 1 , Figure 1 A flow chart of a method for dispatching inspection orders for a water supply and drainage network provided in accordance with an embodiment of the present specification is shown.

[0025] like Figure 1 As shown, the water supply and drainage network inspection dispatching method may at least include the following steps: Step 102, obtaining historical inspection data of the water supply and drainage network.

[0026] Among them, historical inspection data can be obtained from past inspection reports, maintenance records, fault logs and other data.

[0027] Furthermore, the historical inspection data of the water supply and drainage network can be cleaned, missing values ​​can be processed, errors can be corrected, and the format can be standardized to ensure the quality of the data.

[0028] Step 104: Use a data mining algorithm to identify dynamic inspection behavior information from historical inspection data.

[0029] Specifically, in this embodiment, in step 104, a data mining algorithm is used to identify dynamic inspection behavior information from historical inspection data, including: Step 202: Use an association rule mining algorithm to perform association rule mining on the historical inspection data to obtain a number of item sets and association rules of the historical inspection data.

[0030] Among them, association rule mining algorithms include but are not limited to Apriori algorithm, FP-growth algorithm, etc. Association rule mining algorithms are designed to discover the relationship between items in a data set. An item set is a combination of several items. For example, in a historical inspection data set, an item can be the ID of an inspection task, the ID of an inspector, the skills of an inspector, the inspection date, the time required for the inspection, the geographical location of the inspection area, the inspection route, the inspection point, the operation data of the water supply and drainage network, the abnormal value of the network pressure, the equipment type, the fault type, the maintenance measures, etc.

[0031] Furthermore, in this embodiment, in step 202, an association rule mining algorithm is used to perform association rule mining on the historical inspection data to obtain several item sets and association rules of the historical inspection data, including: Clustering algorithm is used to cluster historical inspection data; Analyze the feature distribution in different clusters, identify inspection data with similar features, and aggregate the inspection data into several types of inspection patterns; An association rule mining algorithm is used to mine association rules for historical inspection data of various inspection modes, and several item sets and association rules of historical inspection data of various inspection modes are obtained.

[0032] Among them, the implementation method of clustering historical inspection data using clustering algorithm is as follows: First, select a clustering algorithm: You can select a suitable clustering algorithm according to the characteristics of the historical inspection data and the clustering target, such as K-means, hierarchical clustering (such as AGNES), density-based clustering (such as DBSCAN), etc. Preferably, in this embodiment, the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) clustering algorithm is selected to cluster the historical inspection data, because the DBSCAN clustering algorithm is suitable for discovering clusters of complex shapes and different sizes, and can identify noise points (outliers). In the water supply and drainage network, the inspection data may contain areas of different densities and shapes, and DBSCAN can effectively handle this situation.

[0033] Secondly, determine the clustering parameters: determine the neighborhood radius and the minimum number of points, and then determine the number and shape of clusters by calculating the density between data points.

[0034] Finally, execute the clustering algorithm: use the DBSCAN clustering algorithm to cluster the historical inspection data.

[0035] The implementation method of analyzing the feature distribution in different clusters, identifying the inspection data with similar features, and aggregating the inspection data into several types of inspection modes is as follows: First, the distribution of inspection behavior characteristics of each cluster is observed, and the cohesion and separation of each cluster are evaluated using internal indicators (such as silhouette coefficient) to determine the quality of the cluster; Secondly, the similarities within and between clusters are compared by internal and external distances to identify inspection data with similar characteristics in different clusters; Finally, the clustering results are combined with business background knowledge, and the characteristics of the clustering results are explained. Then, according to the inspection data, they are aggregated into several types of inspection modes and each mode is named. For example, if the pipe network operation data of a cluster is generally stable, the inspection frequency is relatively regular, and the inspection points are evenly distributed in the geographical space, it may indicate a regular inspection mode. If the pipe network operation data of a cluster has obvious mutations, the inspection frequency is irregular, and the inspection points are concentrated in a specific area, it may indicate an emergency inspection mode. If the pipe network operation data of a cluster shows periodic changes and shows an improvement trend after the inspection, and the inspection records are at specific and regular time intervals, it may indicate a regular maintenance mode.

[0036] Through clustering algorithms, we understand the inherent structure and characteristics of inspection tasks, identify inspection data with similar characteristics, and aggregate them into several types of inspection patterns based on the inspection data. We then conduct association rule mining on historical inspection data of different inspection patterns to reveal the correlation between inspection tasks, in order to adapt to the personalized needs of different inspection patterns for inspection resources.

[0037] Step 204: set a support threshold and a confidence threshold, and select frequent item sets and strong association rules from a plurality of item sets and association rules based on the set support threshold and confidence threshold.

[0038] Among them, support is the frequency of an item set appearing in all transactions, and an item set refers to the set of all items in the data set. The support threshold is a user-specified minimum frequency used to filter out unimportant item sets.

[0039] The implementation method of filtering frequent itemsets can be: 1. The Apriori algorithm calculates the support of item sets by scanning the data set multiple times, and prunes the item sets that do not meet the support, so as to screen out frequent item sets.

[0040] 2. The FP-growth algorithm uses a data structure called a prefix tree to efficiently find frequent itemsets.

[0041] Among them, confidence is the conditional probability of the occurrence of the consequence when the antecedent occurs. The confidence threshold is a user-specified minimum confidence value used to filter out unimportant association rules.

[0042] The implementation method of screening out strong association rules can be: First, for each frequent item set, all possible association rules are generated; Then, the confidence of each association rule is calculated and compared with the confidence threshold, and those association rules whose confidence is not less than the confidence threshold are retained to screen out strong association rules.

[0043] Step 206: Determine representative dynamic inspection behaviors based on frequent item sets and strong association rules.

[0044] For example, frequent item sets and strong association rules can be that certain inspection areas often need to be inspected, the same type of faults often occur on certain specific inspection dates, the same maintenance measures are often taken when the operating data of certain pipelines are abnormal, or certain inspectors often inspect specific types of equipment, etc.

[0045] Among them, the implementation method of determining representative dynamic inspection behaviors based on frequent item sets and strong association rules can be: first, pre-define multiple different dynamic inspection behavior categories based on the inspection business needs of the water supply and drainage network, and then match the frequent item sets and strong association rules with the pre-defined dynamic inspection behavior categories to obtain representative dynamic inspection behaviors.

[0046] By way of example, the representative dynamic inspection behaviors determined in this embodiment include, but are not limited to: regular high-frequency inspections that match frequent item sets of frequently inspected inspection areas, seasonal fault prevention inspections that match strong association rules between specific dates and specific fault types, data-driven predictive inspections, emergency inspections for data mutations, professional skill matching inspections that match strong association rules that specific inspectors frequently inspect specific types of equipment, etc.

[0047] Furthermore, in this embodiment, in step 104, using a data mining algorithm to identify dynamic inspection behavior information from historical inspection data also includes: Periodically obtain the latest inspection data of the water supply and drainage network; Based on the latest inspection data of the water supply and drainage network, steps 202 to 206 are executed to update representative dynamic inspection behaviors, ensuring that the determined dynamic inspection behaviors can more accurately reflect the real-time status and dynamic inspection requirements of the water supply and drainage network.

[0048] Step 208: identifying dynamic inspection behavior information from historical inspection data according to the determined dynamic inspection behavior.

[0049] For example, the dynamic inspection behavior information identified in this embodiment includes but is not limited to high-frequency time periods of inspections, geographical information of the inspection area, inspection routes, operating data and abnormal points of the water supply and drainage network, fault types, inspection task response time, inspector skills, inspection resources (inspectors, equipment), etc.

[0050] This embodiment uses an association rule mining algorithm to identify representative dynamic inspection behavior information from historical inspection data, and can more comprehensively and accurately identify dynamic inspection behavior information that affects inspection task dispatching, providing a more accurate and reliable basis for subsequent inspection dispatching. At the same time, by periodically acquiring the latest inspection data to update representative dynamic inspection behaviors, it is ensured that the determined dynamic inspection behaviors can more accurately reflect the real-time status and dynamic inspection needs of the water supply and drainage network. Moreover, this method emphasizes data-based, and extracts valuable rules through data mining technology, which reduces subjective judgment and greatly improves the scientific nature of decision-making.

[0051] Step 106: construct an inspection task evaluation index system based on the identified dynamic inspection behavior information.

[0052] By way of example, the inspection task evaluation index system constructed in this embodiment includes but is not limited to operating data and abnormal points of the water supply and drainage network, fault types, maintenance difficulty, the number of inspectors required for the task, geographic information of the inspection area, the location of the inspectors, the workload of the inspection task, the skill requirements of the inspectors, and the estimated occupancy time of resources.

[0053] Step 108, obtaining water supply and drainage network monitoring data and inspector load data in real time.

[0054] For example, the pressure, flow, water quality and other monitoring data of the water supply and drainage network can be collected in real time through sensors and pipeline monitoring systems, and the inspector load data such as the inspector's location, work status and current task load can be tracked in real time using mobile devices or positioning systems.

[0055] Furthermore, in this embodiment, in step 108, real-time acquisition of water supply and drainage network monitoring data and inspector load data includes: GIS technology is used to collect the geographic information of the water supply and drainage network in real time, and IoT technology is used to collect the operation data of the water supply and drainage network in real time. The geographic information and operation data of the water supply and drainage network constitute the monitoring data of the water supply and drainage network; Receive load data submitted by inspectors.

[0056] For example, the implementation method of using GIS (geographic information system) technology to collect geographic information of the water supply and drainage network in real time is: through the GIS system, real-time recording of the spatial position, direction, connection relationship and other geographic information of the water supply and drainage network, planning inspection areas and inspection routes, formulating inspection tasks, adding pipeline facilities that need to be inspected and inspection points that do not exist in the GIS system to the tasks, so as to adapt to the complex and changing external environment.

[0057] For example, the use of IoT technology to collect real-time operating data of water supply and drainage pipelines is achieved by: obtaining the real-time flow, pressure operating status, water quality monitoring results and key parameters of the pipeline facilities to be inspected through Internet of Things (IoT) devices. At the same time, mobile inspection equipment, such as smart phones or tablets, can also be used to enter or correct on-site data to ensure the accuracy of the water supply and drainage pipeline operation data.

[0058] For example, the implementation method of receiving the load data submitted by the inspector is: the inspector can actively upload or automatically record the inspector's current workload, work status, work progress and other load data through mobile device applications, handheld terminals, smart wearable devices, etc.

[0059] Furthermore, this embodiment can also construct a pipeline information database, fuse the historical inspection data of the water supply and drainage pipeline network, the real-time water supply and drainage pipeline network monitoring data and the inspector load data and store them in the pipeline information database, providing a good foundation for subsequent data analysis and traceability.

[0060] For example, the data fusion of historical inspection data of the water supply and drainage network, real-time water supply and drainage network monitoring data, and inspector load data can be achieved by: identifying and merging records from different data sources but representing the same entity, merging historical data with real-time water supply and drainage network monitoring data and inspector load data in chronological order to create a continuous data stream, and using data fusion technologies such as GIS (geographic information system) and BIM (building information model) integration to fuse spatial information with inspection data to form a unified network information database.

[0061] Through GIS technology and Internet of Things technology, the geographic information, operation data and inspector load data of the pipeline network are collected in real time, and historical data and real-time data are integrated to form a unified pipeline information database, which will help to achieve more efficient, intelligent and reliable water supply and drainage pipeline network management.

[0062] Step 110, based on the inspection task evaluation index system, extract the required indicators from the water supply and drainage network monitoring data and the inspector load data.

[0063] By way of example, the indicators extracted from the water supply and drainage network monitoring data and the inspector load data in this embodiment include but are not limited to the operation data and abnormal points of the water supply and drainage network, water quality monitoring results, fault type, maintenance difficulty, the number of inspectors required for the task, geographic information of the inspection area, the location of the inspectors, the workload of the inspection task, the workload and availability of the inspectors, the skill requirements of the inspectors, and the estimated occupancy time of the resources.

[0064] Step 112: dynamically adjust the priority of each inspection task and the dispatching algorithm based on the extracted indicators.

[0065] Optionally, a decision tree, rule engine or machine learning model can be used to dynamically adjust the task priority based on the extracted indicators, and then a greedy algorithm, ant colony algorithm or other algorithm can be selected as the dispatching algorithm based on the adjusted task priority and the indicator data of the inspection task.

[0066] Furthermore, in this embodiment, step 112 dynamically adjusts the priority of each inspection task and the dispatching algorithm based on the extracted indicators, including: Step 302: Prioritize each inspection task based on the extracted indicators.

[0067] Specifically, in this embodiment, the inspection tasks are prioritized based on the extracted indicators according to the preset priority rules. For example, the preset priority rules in this embodiment include but are not limited to the following points: Rule 1: Determine whether there is a leak / burst based on the operating data and abnormal points of the water supply and drainage network. Emergency inspection tasks such as leaks / bursts have the highest priority to reduce water waste and potential property losses.

[0068] Rule 2: Determine whether there are water quality problems based on water quality monitoring results, with water quality problems taking second priority to ensure water supply safety.

[0069] Rule 3: Based on the geographic information of the inspection area, determine whether the problematic water supply pipeline is the main water supply pipeline or a branch pipeline, and whether it is a key inspection point. Problems with the main water supply pipeline have a higher priority than those with the branch pipeline. Problems at key inspection points such as water plant outlets and large regulating tanks need to be handled first.

[0070] Rule 4: If historical inspection data shows that a certain area is prone to failure, the priority of the inspection task in this area is increased.

[0071] Rule 5: Prioritize preventive maintenance tasks based on the age of the facility and when it was last maintained.

[0072] Rule 6: Adjust the priority of seasonal fault prevention inspection tasks based on seasonal changes and historical inspection data. For example, give priority to inspecting pipe network areas prone to water accumulation before the rainy season, and give priority to inspecting pipe network areas prone to freezing and cracking before winter.

[0073] Step 304: Based on the priority ranking of each inspection task, a corresponding dispatching algorithm is selected. The dispatching algorithm includes a greedy algorithm and an ant colony algorithm.

[0074] It is understandable that the greedy algorithm is suitable for scenarios with clear priorities and simple dispatching processes, while the ant colony algorithm is suitable for complex dispatching scenarios with unclear priorities or multiple paths and multiple tasks. Specifically, if the priority ranking of each inspection task is relatively clear and there are enough inspectors to match each inspection task, the greedy algorithm is selected as the dispatching algorithm; if there are multiple inspection tasks with the same priority level or in complex dispatching scenarios with multiple paths and multiple tasks or there are not enough inspectors to match each inspection task (that is, one inspector needs to perform many inspection tasks), the ant colony algorithm is selected as the dispatching algorithm. .

[0075] It is understandable that when changes occur in the water supply and drainage network monitoring data and the inspector load data (including changes in the water supply and drainage network monitoring data that exceed the threshold and the arrival of new inspection tasks / data), the priority adjustment mechanism is triggered.

[0076] By real-time monitoring of the water supply and drainage network monitoring data and inspector load data, and dynamically adjusting the priority of each inspection task based on the changes in indicators extracted from real-time data, it can adapt to the ever-changing actual conditions of the pipeline network and the working environment, and improve the flexible adjustment capabilities of the water supply and drainage network inspection dispatching, ensuring that emergencies can be responded to quickly. At the same time, the inspection tasks can be dispatched more reasonably considering the workload of the inspectors, and then the corresponding dispatching algorithm can be selected according to the priority ranking of each inspection task after adjustment, providing more scientific decision-making support for dispatching, improving the accuracy of decision-making, and thereby improving the overall effect of water supply and drainage network inspection dispatching.

[0077] Step 114: dispatching inspection tasks based on the priorities of the inspection tasks and the dispatching algorithm.

[0078] On the one hand, in this embodiment, in a scenario where the priority is clear and the dispatching process is simple, in step 114, dispatching of the inspection tasks is performed based on the priority of each inspection task and the dispatching scheduling algorithm, including: Step 402: When the determined dispatch scheduling algorithm is a greedy algorithm, a task list to be dispatched is created.

[0079] Among them, the task list to be dispatched is used to store the inspection tasks to be assigned and the corresponding inspectors.

[0080] Step 404: Select the inspection task with the highest priority from the task list, and for the current inspection task, calculate the distance between each inspector and the inspection task area based on the position of the inspector.

[0081] The calculation of the distance between each inspector and the inspection task area based on the position of the inspector can be achieved through a geographic information system (GIS) or similar positioning services, which will not be described in detail in this embodiment.

[0082] Step 406 , based on the distance, the inspector's workload, availability, and the estimated occupancy time of the resource, calculate the comprehensive score of each inspector in the current inspection task.

[0083] Among them, the inspector's workload includes the number of assigned tasks and the estimated completion time. The inspector's availability indicates whether the inspector is in a working state and whether the number of tasks assigned to the inspector is saturated. If the inspector has been assigned multiple tasks and the task volume is large, his availability is reduced.

[0084] Specifically, step 406 is implemented as follows: Firstly, the data of each indicator is standardized to obtain the standardized value of each indicator.

[0085] It is understandable that, since different indicators may have different dimensions and numerical ranges, the indicators need to be standardized to ensure that their contributions to the score are equivalent. Commonly used standardization methods include minimum-maximum standardization and Z-score standardization, which are not limited in this embodiment.

[0086] Next, the following formula can be used to calculate the comprehensive score of each inspector in the current inspection task: Comprehensive score = w1 * (1 / distance) + w2 * (1 / workload) + w3 * availability + w4 * (1 / estimated resource occupancy time) Among them, w1, w2, w3, w4 are pre-set weight coefficients used to adjust the influence of each factor.

[0087] It can be understood that the comprehensive scoring formula represents the overall suitability score of the inspector when performing the current inspection task. The higher the score, the more suitable the inspector is for the task. The closer the distance between the inspector and the inspection task area, the higher the score, which means that the inspector can reach the task location faster; the smaller the inspector's workload, the higher the score, indicating that the inspector has more time and energy to accept new tasks; the higher the availability, the higher the score, indicating that the inspector can perform the task immediately or quickly; the shorter the estimated occupancy time of the resource, the higher the score, indicating that the inspector can complete the task more efficiently. It helps to more comprehensively and quickly identify the inspector who is most suitable for performing the current inspection task, thereby improving the efficiency and response speed of task dispatch.

[0088] Step 408, based on the comprehensive score of each inspector and the number of inspectors required for the task, determine the inspector that matches the current inspection task and complete the dispatch of the current inspection task.

[0089] Based on the comprehensive score, the inspector with the highest score is selected as the best match. If multiple inspectors have the same score, other factors can be further considered, such as whether the inspector's skills match the current inspection task, or a random selection can be made.

[0090] Step 410, remove the current inspection task from the task list and update the inspector load data.

[0091] Among them, updating the inspector load data includes updating the inspector's workload, resource usage, and inspector's location information.

[0092] Step 412, repeat steps 404 to 410 until there are no inspection tasks to be dispatched in the task list.

[0093] In scenarios with clear priorities and simple dispatching processes, the local optimal solution is usually consistent with the global optimal solution. Therefore, this embodiment adopts a greedy algorithm to dispatch inspection tasks. According to the priority ranking of each inspection task, the current optimal inspector is directly selected for dispatching each inspection task based on the current resource situation. It does not involve complex calculation processes and can make decisions more quickly while ensuring the overall effect of water supply and drainage network inspection dispatching. It is more suitable for inspection task dispatching that requires real-time response or in emergency situations.

[0094] However, since the greedy algorithm only considers the optimal choice of the current inspection task, but does not consider the optimality of the overall inspection dispatch scheduling, in complex dispatch scenarios where the priority is not clear or there are multiple paths and multiple tasks, the local optimal solution may not be the global optimal solution. Therefore, on the other hand, in this embodiment, in step 114, the dispatch scheduling of the inspection task is performed based on the priority of each inspection task and the dispatch scheduling algorithm, including: Step 502: When the determined dispatch scheduling algorithm is the ant colony algorithm, each inspection task is regarded as a node, and each inspector is regarded as a moving ant.

[0095] Furthermore, attributes are set for each task node, such as task priority, inspection area location, resources required for inspection, etc. Attributes are set for each ant, such as the inspector's current location, current workload, availability, occupied resources, etc.

[0096] Step 504, setting the parameters of the ant colony algorithm, initializing the pheromone matrix τ, and setting the same initial value for the pheromone concentration on all paths.

[0097] The parameters of the ant colony algorithm include but are not limited to the pheromone importance factor α, the heuristic function importance factor β, the pheromone evaporation coefficient ρ, the pheromone intensity Q, the number of ants m, etc.

[0098] Step 506 , based on the priority of each inspection task, the position, workload, availability, estimated occupancy time of the resource and the pheromone concentration on the path, the probability of each ant selecting each inspection task is calculated.

[0099] For example, for each inspection task to be dispatched , according to the task priority, ant To the task Distance, ants The current status (such as workload, resource usage, etc.) and the pheromone concentration on the path , computing ants Select Task Probability .

[0100] in, is the heuristic function, which represents the ant To the task Non-pheromone heuristic factors, for example, the heuristic function can be proportional to the task priority or the ant The inverse of the workload of ants To the task The inverse of the distance, etc. T represents the total number of inspection tasks to be dispatched.

[0101] Step 508, based on the probability of each ant selecting each inspection task and the number of inspectors required for the task, determine the inspection task that matches each inspector, and update the pheromone concentration on the corresponding path after each ant completes a task dispatch.

[0102] The implementation method of determining the inspection task matching each inspector based on the probability of each ant selecting each inspection task may use a roulette selection method or other selection strategies, which is not limited in this embodiment.

[0103] For example, after each ant completes a task assignment, the local update rule for updating the pheromone concentration on the corresponding path can be: ,in, Indicates the local pheromone enhancement amount, which is a small fixed value or based on ant To the task The evaluation value can be calculated based on factors such as task priority, the number of tasks completed by ants, and the task completion time.

[0104] Step 510: After all ants have completed one iteration, the pheromone concentration is globally updated, and the optimal dispatching and scheduling scheme between inspectors and inspection tasks in the current iteration is recorded.

[0105] It is understandable that only the best or better dispatching scheme will enhance the pheromone concentration on the path it passes through. The global update rule for globally updating the pheromone concentration can be ,in , Q represents a positive constant, represents the total cost of the kth ant to complete all inspection tasks, which shows that the global update depends on the quality of dispatch scheduling.

[0106] Among them, if the new optimal dispatching scheduling scheme is better than the historical optimal dispatching scheduling scheme, then the optimal dispatching scheduling scheme between the inspectors and the inspection tasks in the current iteration is the updated optimal dispatching scheduling scheme. Specifically, the implementation method for judging whether the new dispatching scheduling scheme is better than the historical optimal dispatching scheduling scheme is: (1) Defining the optimization goal: In this embodiment, the optimization goal can be to minimize the time to complete the task, balance the workload, etc.

[0107] (2) Calculate the performance indicators of the dispatching scheme: For the new dispatching scheme, calculate its performance indicators, such as total task completion time, average workload, etc. These indicators must correspond to the optimization objectives.

[0108] (3) Comparison of performance indicators: If the optimization goal is to minimize the time to complete the task, then compare the total time to complete the task of the new dispatch scheduling plan with the total time to complete the task of the historical optimal dispatch scheduling plan. If the new dispatch scheduling plan has a shorter time to complete the task, it is considered to be better; If the optimization goal is to balance the workload, then compare the workload distribution of the new dispatch scheduling scheme with the workload distribution of the historical optimal dispatch scheduling scheme. If the workload of the new dispatch scheduling scheme is more balanced, it is considered to be better.

[0109] (4) Considering constraints: When making comparisons, it is also necessary to consider whether the dispatching plan meets all constraints, such as inspection resource constraints, inspection time constraints, inspector skill matching, etc. If the new dispatching plan is better than the historical optimal dispatching plan while meeting all constraints, then the new dispatching plan is considered to be better.

[0110] Step 512, repeating steps 506 to 510 to iterate until a predetermined termination condition is met, and outputting the optimal dispatching and scheduling plan between the inspectors and the inspection tasks.

[0111] For example, the predetermined termination condition in this embodiment may be a preset number of iterations, or may be that the optimal dispatching scheduling scheme between inspectors and inspection tasks has not been improved in multiple consecutive iterations.

[0112] By adopting the ant colony algorithm to dispatch inspection tasks, it can adapt to the changes in the priority and complexity of each inspection task, and dynamically adjust the search direction through the update of pheromones, making the dispatching of inspection tasks more flexible. Through the positive feedback mechanism of pheromones and the random search strategy, it can effectively avoid converging to the local optimal solution too early. At the same time, it takes into account the workload balance of inspectors, avoids the situation where some inspectors are overworked while other inspectors are relatively idle, and provides a more comprehensive and robust dispatching solution, which is more suitable for the dispatching of inspection tasks that need to deal with complex and dynamically changing.

[0113] In summary, this embodiment first identifies dynamic inspection behavior information from historical inspection data through a data mining algorithm, which can more comprehensively understand the characteristics of inspection tasks and the working status of inspectors; secondly, based on the identified dynamic inspection behavior information, a patrol task evaluation index system is constructed to provide scientific quantitative standards for the priority sorting and resource allocation of inspection tasks; then, the water supply and drainage network monitoring data and inspector load data are acquired in real time to ensure that the dispatch decision is based on the latest information, so that the inspection dispatch scheduling system can respond quickly to emergencies; then, the required indicators are extracted from the real-time data, and the priority of each inspection task and the dispatch scheduling algorithm are dynamically adjusted based on the extracted indicators, which greatly improves the efficiency of the inspection task. It improves the flexible adjustment capability of inspection dispatching of water supply and drainage network. Moreover, in the dispatching dispatching algorithm, the greedy algorithm is suitable for scenarios with clear priorities and simple dispatching process, while the ant colony algorithm is suitable for complex dispatching scenarios with unclear priorities or multiple paths and multiple tasks. This solution of dynamically adjusting priorities and dispatching algorithms based on real-time data can adapt to the ever-changing environment and needs, more accurately match inspectors with inspection tasks, and thus more reasonably allocate inspection resources. Finally, the dispatching of inspection tasks is carried out based on the priority of each inspection task and the dispatching dispatching algorithm, which improves the overall effect of inspection dispatching of water supply and drainage network, and provides strong support for water companies to achieve intelligent management and sustainable development.

[0114] In another embodiment of the present specification, after step 114, the following further steps are included: Step 602, obtaining the inspection point information in each inspection task; Step 604: Consider each inspection point as a node and each inspector as a moving ant, and use the ant colony algorithm to calculate the movement of each ant at the current node. Select the next node probability.

[0115] For example, for each inspection point to be inspected , according to Ant Slave Node To the next node The distance and pheromone concentration along the path , computing ants Slave Node Select the next node Probability .

[0116] in, is the heuristic function, which represents the ant For non-pheromone heuristic factors of the inspection points, for example, the heuristic function can be ant Slave Node To the next node The reciprocal of the distance. represents the weight of pheromone concentration, represents the weight of the heuristic function, and N represents the total number of inspection points to be inspected.

[0117] Step 606, based on the probability of each ant selecting the next node at the current node, the inspection path planning of each inspector is determined, and after each ant completes an inspection path planning, the pheromone concentration on the corresponding path is updated.

[0118] Step 608: After all ants have completed one iteration, the pheromone concentration is globally updated, and the optimal path solution between the inspector and the inspection point in the current iteration is recorded.

[0119] Among them, in this embodiment, after each ant completes a patrol path planning, the local update rule of updating the pheromone concentration on the corresponding path and the global update rule of globally updating the pheromone concentration after all ants have completed an iteration are similar to the local update rule and the global update rule of updating the pheromone concentration on the corresponding path after each ant completes a task dispatch in the aforementioned embodiment, and this embodiment will not be repeated here.

[0120] Step 610, repeating steps 604 to 608 for iteration until a predetermined termination condition is met, and outputting the optimal path plan between the inspector and the inspection point.

[0121] This embodiment further adopts the ant colony algorithm to achieve optimal path planning between inspectors and each inspection point in the inspection area, and dynamically adjusts the path planning from a global perspective based on real-time data to adapt to changes in the inspection area, thereby reducing the moving distance and time wasted by inspectors during the inspection process, helping to more reasonably allocate inspection resources, thereby improving overall inspection efficiency.

[0122] See also Figure 2 , Figure 2 A schematic diagram of the structure of a multimodal model pre-training device provided for one embodiment of this specification.

[0123] like Figure 2 As shown, the water supply and drainage network inspection dispatching system may at least include a data acquisition module 1, an information identification module 2, an indicator system construction module 3, an indicator extraction module 4, a dynamic adjustment module 5 and a dispatching module 6, wherein: Data acquisition module 1, used to obtain historical inspection data of the water supply and drainage network, and obtain water supply and drainage network monitoring data and inspector load data in real time; Information identification module 2, used to identify dynamic inspection behavior information from historical inspection data using data mining algorithms; The indicator system construction module 3 is used to construct an inspection task evaluation indicator system based on the identified dynamic inspection behavior information; The indicator extraction module 4 is used to extract the required indicators from the water supply and drainage network monitoring data and the inspector load data based on the inspection task evaluation indicator system; Dynamic adjustment module 5, used to dynamically adjust the priority of each inspection task and the dispatching algorithm based on the extracted indicators; The dispatching and scheduling module 6 is used to perform dispatching and scheduling of inspection tasks based on the priority of each inspection task and the dispatching and scheduling algorithm.

[0124] It can be understood that the technical concept of the water supply and drainage network inspection and dispatching system provided in this embodiment is similar to the technical concept of the aforementioned water supply and drainage network inspection and dispatching method, and this embodiment will not be repeated here.

[0125] The above description is only the preferred embodiment disclosed in this application and the description of the technical principle used. Those skilled in the art should understand that the scope of protection involved in this disclosure is not limited to the technical solution formed by the specific combination of the above technical features, but also should cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in this disclosure (but not limited to) to form a technical solution.

[0126] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

Claims

1. Inspection and dispatching method for water supply and drainage network. It is characterized in that The following steps are involved: Step 102, obtaining historical inspection data of the water supply and drainage network; Step 104, using a data mining algorithm to identify dynamic inspection behavior information from historical inspection data; Step 106, constructing an inspection task evaluation index system based on the identified dynamic inspection behavior information; Step 108, obtaining water supply and drainage network monitoring data and inspector load data in real time; Step 110, based on the inspection task evaluation index system, extract the required indicators from the water supply and drainage network monitoring data and the inspector load data; Step 112, dynamically adjusting the priority of each inspection task and the dispatching algorithm based on the extracted indicators; Step 114: dispatching inspection tasks based on the priorities of the inspection tasks and the dispatching algorithm.

2. The water supply and drainage network inspection dispatching method according to claim 1, It is characterized in that In step 104, a data mining algorithm is used to identify dynamic inspection behavior information from historical inspection data, including: Step 202, using an association rule mining algorithm to perform association rule mining on the historical inspection data to obtain a number of item sets and association rules of the historical inspection data; Step 204, setting a support threshold and a confidence threshold, and screening out frequent item sets and strong association rules from a number of item sets and association rules based on the set support threshold and confidence threshold; Step 206, determining representative dynamic inspection behaviors based on frequent item sets and strong association rules; Step 208: identifying dynamic inspection behavior information from historical inspection data according to the determined dynamic inspection behavior.

3. The water supply and drainage network inspection dispatching method according to claim 2, It is characterized in that Step 202, using an association rule mining algorithm to perform association rule mining on historical inspection data, including: Clustering algorithm is used to cluster historical inspection data; Analyze the feature distribution in different clusters, identify inspection data with similar features, and aggregate the inspection data into several types of inspection patterns; An association rule mining algorithm is used to mine association rules for historical inspection data of various inspection modes, and several item sets and association rules of historical inspection data of various inspection modes are obtained.

4. The water supply and drainage network inspection dispatching method according to claim 2, It is characterized in that In step 104, a data mining algorithm is used to identify dynamic inspection behavior information from historical inspection data, which also includes: Periodically obtain the latest inspection data of the water supply and drainage network; Based on the latest inspection data of the water supply and drainage network, steps 202 to 206 are executed to update representative dynamic inspection behaviors.

5. The water supply and drainage network inspection dispatching method according to claim 1, It is characterized in that Step 112, dynamically adjusting the priority of each inspection task and the dispatching algorithm based on the extracted indicators, includes: Step 302, prioritizing each inspection task based on the extracted indicators; Step 304: Select a corresponding dispatching algorithm based on the priority ranking of each inspection task. The dispatching algorithm includes a greedy algorithm and an ant colony algorithm.

6. The method for dispatching inspection orders for water supply and drainage network according to claim 5, It is characterized in that The indicators include the number of inspectors required for the task, the location, workload, availability and estimated occupancy time of the resources of the inspectors. In step 114, the inspection task dispatching is performed based on the priority of each inspection task and the dispatching algorithm, including: Step 402: when the determined dispatching algorithm is a greedy algorithm, a task list to be dispatched is created; Step 404, selecting the inspection task with the highest priority from the task list, and for the current inspection task, calculating the distance between each inspector and the inspection task area based on the position of the inspector; Step 406, calculating a comprehensive score of each inspector for the current inspection task based on the distance, the inspector's workload, availability, and the estimated occupancy time of the resource; Step 408, based on the comprehensive scores of each inspector and the number of inspectors required for the task, determine the inspector that matches the current inspection task and complete the dispatch of the current inspection task; Step 410, remove the current inspection task from the task list and update the inspector load data; Step 412, repeat steps 404 to 410 until there are no inspection tasks to be dispatched in the task list.

7. The method for dispatching inspection orders for water supply and drainage network according to claim 5, It is characterized in that The indicators include the number of inspectors required for the task, the location, workload, availability and estimated occupancy time of the resources of the inspectors. In step 114, the inspection task dispatching is performed based on the priority of each inspection task and the dispatching algorithm, including: Step 502: When the determined dispatching algorithm is the ant colony algorithm, each inspection task is regarded as a node, and each inspector is regarded as a moving ant; Step 504, setting the parameters of the ant colony algorithm, initializing the pheromone matrix, and setting the same initial value for the pheromone concentration on all paths; Step 506, based on the priority of each inspection task, the position, workload, availability, estimated occupancy time of the resource, and the pheromone concentration on the path, the probability of each ant selecting each inspection task is calculated; Step 508, based on the probability of each ant selecting each inspection task and the number of inspectors required for the task, determine the inspection task that matches each inspector, and after each ant completes a task matching, update the pheromone concentration on the corresponding path; Step 510: After all ants have completed one iteration, the pheromone concentration is globally updated, and the optimal dispatching and scheduling scheme between inspectors and inspection tasks in the current iteration is recorded; Step 512, repeating steps 506 to 510 to iterate until a predetermined termination condition is met, and outputting the optimal dispatching and scheduling plan between the inspectors and the inspection tasks.

8. The method for dispatching inspection orders for water supply and drainage network according to claim 7, It is characterized in that After step 114, the method further includes: Step 602, obtaining the inspection point information in each inspection task; Step 604, each inspection point is regarded as a node, each inspector is regarded as a moving ant, and the ant colony algorithm is used to calculate the probability of each ant selecting the next node at the current node; Step 606, based on the probability of each ant selecting the next node at the current node, determine the inspection path planning of each inspector, and after each ant completes an inspection path planning, update the pheromone concentration on the corresponding path; Step 608, when all ants have completed one iteration, the pheromone concentration is globally updated, and the optimal path solution between the inspector and the inspection point in the current iteration is recorded; Step 610, repeating steps 604 to 608 for iteration until a predetermined termination condition is met, and outputting the optimal path plan between the inspector and the inspection point.

9. The water supply and drainage network inspection dispatching method according to claim 1, It is characterized in that In step 108, the water supply and drainage network monitoring data and inspector load data are obtained in real time, including: GIS technology is used to collect the geographic information of the water supply and drainage network in real time, and IoT technology is used to collect the operation data of the water supply and drainage network in real time. The geographic information and operation data of the water supply and drainage network constitute the monitoring data of the water supply and drainage network; Receive load data submitted by inspectors.

10. Inspection and dispatching system for water supply and drainage network, It is characterized in that include: The data acquisition module is used to obtain the historical inspection data of the water supply and drainage network, and obtain the monitoring data of the water supply and drainage network and the inspector load data in real time; An information identification module is used to identify dynamic inspection behavior information from historical inspection data using a data mining algorithm; The indicator system construction module is used to construct the inspection task evaluation indicator system based on the identified dynamic inspection behavior information; The indicator extraction module is used to extract the required indicators from the water supply and drainage network monitoring data and the inspector load data based on the inspection task evaluation indicator system; A dynamic adjustment module, used to dynamically adjust the priority of each inspection task and the dispatching algorithm based on the extracted indicators, wherein the dispatching algorithm includes a greedy algorithm and an ant colony algorithm; The dispatching and scheduling module is used to dispatch inspection tasks based on the priority of each inspection task and adopt a certain dispatching and scheduling algorithm to perform dispatching and scheduling of inspection tasks.

Images