A smart water pump station equipment fault prediction method and automatic switching control system

Abstract

This invention provides a method for predicting equipment failures in smart water pumping stations and an automatic switching control system, belonging to the field of urban water supply technology. The method and control system involve: constructing a hydraulic topology map and a hydraulic dependency matrix; configuring sensors on each pumping station; identifying faulty pumping station nodes using a preset failure prediction model; identifying water-scarce areas along associated edges; setting backup pumping stations based on the hydraulic dependency matrix; dynamically adjusting the combination of backup pumping stations according to a dynamic selection strategy, regional correlation, and regional water scarcity; and generating a set of backup pumping station control commands; combining the backup pumping station control command set with real-time pumping station operation data to control the operation of backup pumping stations, thereby predicting failures and automatically switching modes in the smart water pumping stations. This ensures the continuity of water supply, reduces the occurrence of water supply anomalies, and improves the stability and reliability of the entire water supply system.

Description

Technical Field

[0001] This invention belongs to the field of urban water supply technology, and more specifically, relates to a method for predicting equipment failures in smart water pumping stations and an automatic switching control system. Background Technology

[0002] Currently, urban water supply systems are expanding in scale and increasing in complexity. Smart water management, as a crucial component of smart city construction, is developing towards greater intelligence and precision. Pumping stations, as core facilities of the water supply system, bear the vital responsibility of water transport and pressurization. Their operational status directly impacts the stability and security of urban water supply. Traditional methods of monitoring and handling pumping station faults are insufficient to meet the demands of modern smart water management. It is necessary to predict pumping station equipment malfunctions and water shortage areas, and to dynamically adjust the operation of backup pumping stations in a timely manner.

[0003] Existing technologies have the following problems: equipment status monitoring based on individual pump stations lacks fault analysis of the correlation between pump stations, and the analyzed fault causes are not accurate enough; the fault prediction model is relatively simple, making it difficult to accurately identify faulty pump station nodes and water-scarce areas; the selection and adjustment of backup pump stations adopts a fixed mode, which cannot be dynamically adjusted according to the actual water shortage situation, resulting in the inability to guarantee water supply in a timely and effective manner when a pump station fails; to solve the above problems, this application proposes a smart water pump station equipment fault prediction method and automatic switching control system. Summary of the Invention

[0004] The purpose of this invention is to provide a method for predicting equipment failures in smart water pumping stations, in order to solve the problems existing in the background art.

[0005] This invention is implemented as follows:

[0006] This invention provides a method for predicting equipment failures in smart water pumping stations, comprising the following steps:

[0007] S1: Construct a hydraulic topology map to show the connections between pump stations, and construct a hydraulic dependency matrix to show the hydraulic dependencies between pump stations;

[0008] S2: Sensors are configured on each pumping station to identify faulty pumping station nodes through a preset fault prediction model and to identify water-scarce areas along the associated edges;

[0009] S3: Based on the hydraulic dependency matrix, set up backup pumping stations, dynamically adjust the combination of backup pumping stations according to the dynamic screening strategy, regional correlation and regional water shortage, and generate a set of backup pumping station control commands.

[0010] S4: Combines the set of control commands for standby pump stations with real-time pump station operation data to control the operation of standby pump stations, so as to predict faults and automatically switch modes for smart water pump stations.

[0011] Based on the above technical solution, the intelligent water pumping station equipment fault prediction method and automatic switching control system of the present invention can be further improved as follows:

[0012] Furthermore, S1 includes:

[0013] S11: Based on the pre-set pump station coordinate data, unify the pump station coordinates to the same coordinate system, combine the pipeline connection data between pump stations, take each pump station as an independent node, establish associated edges according to the pipeline connection relationship and the bending flow of the pipeline, and obtain the hydraulic topology map.

[0014] S12: Extract the overlapping areas covered by pipeline connection paths from the hydraulic topology map, determine the intersection areas where multiple pumping stations share water supply, and obtain the intersection water supply areas between pumping stations;

[0015] S13: Calculate the priority of cross-supply areas based on area type, configure pump station dependency analysis strategy, calculate the hydraulic dependency between pump stations, and construct hydraulic dependency matrix.

[0016] Furthermore, S13 includes:

[0017] S131: Determine the corresponding priority base weight according to the area type, calculate the corresponding area weight according to the ratio of the water supply coverage area of ​​the cross-area areas, and add the two weights to obtain the priority score of the cross-area water supply area.

[0018] S132: Analyze the water supply volume between pumping stations to the cross-supply area, obtain the proportion of water supply volume of each pumping station in the cross-area, and calculate the first hydraulic dependence between pumping stations based on priority.

[0019] S133: Collect the water pressure at the water inlet of each pumping station in the cross-water supply area through pressure sensors, calculate the water pressure difference of multiple pumping stations, and correct the first hydraulic dependence by combining the water pressure difference to obtain the second hydraulic dependence.

[0020] S134: Construct a hydraulic dependency matrix based on the second hydraulic dependency between pumping stations.

[0021] Furthermore, S2 includes:

[0022] S21: Configure corresponding sensors for components such as pumps, motors, and pipelines in the pumping station, collect the operating data of the pumping station in real time through the sensors, and preprocess the collected data;

[0023] S22: Based on the collected pump station operation data, extract the corresponding feature data, and the model predicts the operation status of each pump station based on the feature data to identify the pump station nodes that may fail.

[0024] S23: Combining the hydraulic topology map, starting from the faulty pump station node, analyze the water pressure change along the associated edges. When a location with a water pressure difference greater than the preset water pressure threshold is found, that location constitutes a water shortage area.

[0025] Furthermore, S23 includes:

[0026] S231: Starting from the faulty pump station node, calculate the water pressure difference on both sides of each associated pipeline, and search along the associated edge until the water pressure difference is greater than the preset water pressure threshold to obtain the set of water pressure attenuation paths.

[0027] S232: Extract the pipe corresponding to each path in each set of water pressure attenuation paths according to the hydraulic topology map, and take the area covered by the pipe as the first water shortage area.

[0028] S233: Based on the pipeline connection relationship of the first water shortage area and the water pressure difference, the water pressure fluctuation area is identified. The first water shortage area is expanded along the water pressure fluctuation area to obtain the second water shortage area.

[0029] S234: Combine the second water-scarce areas corresponding to each water pressure attenuation path to obtain the water-scarce region.

[0030] Furthermore, S3 includes:

[0031] S31: Based on the hydraulic dependence matrix, extract the second hydraulic dependence value between the faulty pump station and other pump stations, and set the backup pump station at the midpoint between the two pump stations whose second hydraulic dependence value is lower than the preset second hydraulic dependence threshold.

[0032] S32: Based on the priority of water-scarce areas, select the corresponding backup pumping stations for water supply to water-scarce areas with high priority. Based on the correlation between backup pumping stations and faulty pumping stations, select the first backup pumping station combination that can cover water-scarce areas with high correlation.

[0033] S33: Based on the real-time status of the standby pumping stations and the conflict between the standby pumping stations required by the faulty pumping stations, a second standby pumping station combination is obtained by dynamically adjusting and eliminating standby pumping stations with abnormal status.

[0034] S34: Analyze the real-time operating parameters of each standby pump station in the second standby pump station combination, generate specific control commands for each pump station, and accurately control the operating status of each standby pump station.

[0035] Furthermore, S32 includes:

[0036] S321: Calculate the first correlation degree based on the cross-water supply area between the standby pumping station and the faulty pumping station;

[0037] S322: Analyze the real-time water pressure of the standby pump station, calculate the radius of the water supply area, analyze the coverage between the water supply area and the water shortage area, calculate the area of ​​the overlapping area, and obtain the second correlation degree between the standby pump station and the corresponding faulty pump station based on the ratio between the area of ​​the area and the area of ​​each water shortage area.

[0038] S323: Calculate the comprehensive correlation between each standby pumping station and the faulty pumping station by combining the first correlation degree and the second correlation degree;

[0039] S324: Based on the priority of the water-scarce areas, match the backup pumping station with the highest comprehensive correlation for each water-scarce area in descending order of priority, until the water supply area of ​​all matched backup pumping stations can cover the water-scarce areas, thus obtaining the first backup pumping station combination.

[0040] Furthermore, S33 includes:

[0041] S331: Select core parameters reflecting the operating status of the standby pumping stations from the real-time status data of the standby pumping stations, construct a status vector based on the core parameters, analyze the operating status of each standby pumping station through a preset abnormal pumping station identification model, and filter out abnormal pumping stations based on the status vector through the model.

[0042] S332: Based on the conflict situation of the backup pump station required by the faulty pump station, the conflicting pump station is selected, and the corresponding water-scarce area is taken as the conflict area. The hydraulic coupling degree between the conflicting pump station and the faulty pump station corresponding to the conflict area is calculated. Among the conflicting pump stations, the pump station with high hydraulic coupling degree is selected to supply water to the conflict area, and the third pump station combination is obtained.

[0043] S333: Remove abnormal pump stations from the first backup pump station combination, and update the pump stations through the third pump station combination. Replace the corresponding pump station information in the first backup pump station combination with the backup pump station information in the third pump station combination to obtain the second backup pump station combination.

[0044] Furthermore, S4 includes:

[0045] S41: Combining the set of control commands for standby pump stations with real-time pump station operation data, the water supply situation is predicted through a preset fault evolution model, and risk areas of abnormal water supply are screened out.

[0046] S42: Optimize the control commands based on the risk area to obtain an optimized set of control commands, and control the operation of the standby pump station.

[0047] An automatic switching control system for implementing the aforementioned smart water pumping station equipment fault prediction method includes:

[0048] The hydraulic analysis module constructs a hydraulic topology map and a hydraulic dependency matrix. The hydraulic topology map uses preset pump stations as nodes, establishes associated edges based on the pipeline connection relationship between pump stations, extracts the cross-water supply area of ​​pump stations based on the hydraulic topology map, configures the pump station dependency analysis strategy, calculates the hydraulic dependency between pump stations, and constructs the hydraulic dependency matrix.

[0049] The water shortage area prediction module collects pump station operation data in real time through sensors configured in each pump station, combines the hydraulic topology map, identifies faulty pump station nodes through a preset fault prediction model, and identifies water shortage areas along the associated edges.

[0050] The backup pump station screening module sets up multiple backup pump stations among the pump stations based on the hydraulic dependence matrix, configures a dynamic screening strategy for pump stations to analyze the regional correlation between the faulty pump station and the backup pump station, dynamically screens the backup pump station combination in combination with the water shortage area, and generates a set of backup pump station control commands.

[0051] The control command optimization module combines the set of control commands for the backup pumping station with real-time pumping station operation data, predicts the water supply situation through a preset fault evolution model, and optimizes the control commands to obtain an optimized set of control commands. This optimizes the operation of the backup pumping station and enables fault prediction and automatic mode switching for the smart water pumping station.

[0052] Compared with existing technologies, the beneficial effects of the intelligent water pumping station equipment fault prediction method and automatic switching control system provided by this invention are as follows: A hydraulic topology map is constructed based on the connection relationship between pumping stations and pipelines; cross-supply areas are extracted and hydraulic dependence is calculated; faulty pumping station nodes are identified using a preset fault prediction model based on real-time sensor data; water-scarce areas are identified by combining the hydraulic topology map and water pressure difference analysis; backup pumping stations are set based on the hydraulic dependence matrix; correlation is calculated based on the priority of water-scarce areas; backup pumping station combinations are selected; and dynamic adjustments are made based on the real-time status and conflict situations of backup pumping stations to optimize their working status in real time. By combining the correlation relationships between pumping stations, faulty pumping station nodes and water-scarce areas can be accurately identified; and the dynamic selection and real-time optimization of backup pumping station combinations improves the utilization efficiency of backup pumping stations, ensures that the water supply area of ​​backup pumping stations can effectively cover water-scarce areas, guarantees the continuity of water supply, reduces the occurrence of water supply anomalies, and improves the stability and reliability of the entire water supply system. Attached Figure Description

[0053] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained from these drawings without creative effort.

[0054] Figure 1 A flowchart illustrating the process of a method for predicting equipment failures in smart water pumping stations;

[0055] Figure 2 A partial schematic diagram of the hydraulic topology for a method of predicting equipment failures in smart water pumping stations;

[0056] Figure 3 A schematic diagram illustrating the water shortage area identification process in a smart water pumping station equipment fault prediction method;

[0057] Figure 4 This is a schematic diagram of an automatic switching control system. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0060] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0061] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0062] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0063] Example 1

[0064] like Figure 1 As shown, the present invention provides a method for predicting equipment failures in smart water pumping stations and an automatic switching control system, which includes the following steps:

[0065] S1: Construct a hydraulic topology map to show the connections between pump stations, and construct a hydraulic dependency matrix to show the hydraulic dependencies between pump stations;

[0066] S2: Sensors are configured on each pumping station to identify faulty pumping station nodes through a preset fault prediction model and to identify water-scarce areas along the associated edges;

[0067] S3: Based on the hydraulic dependency matrix, set up backup pumping stations, dynamically adjust the combination of backup pumping stations according to the dynamic screening strategy, regional correlation and regional water shortage, and generate a set of backup pumping station control commands.

[0068] S4: Combines the set of control commands for standby pump stations with real-time pump station operation data to control the operation of standby pump stations, so as to predict faults and automatically switch modes for smart water pump stations.

[0069] By collecting data on pump station locations (determining node spatial coordinates), pipeline connections (clarifying the pipeline network topology), and water supply range (dividing pump station service boundaries), a hydraulic topology map and hydraulic dependency matrix are constructed. This provides data reference for pump station fault prediction and handling. Combined with a fault prediction model, faulty pump stations and water-scarce areas are identified, and fault problems are predicted in a timely manner. Based on the hydraulic dependency matrix and dynamic screening strategy, backup pump station combinations are dynamically selected according to the water-scarce area conditions, and corresponding control commands are generated. The control commands and real-time data are input into the fault evolution model to predict the water supply status of backup pump stations, optimize the control commands, and control the operation of backup pump stations to achieve faulty pump station handling and water supply restoration.

[0070] In the water system, each preset pump station is used as a node. Corresponding edges are established based on the pipeline connection relationship between the pump stations to construct a hydraulic topology map, which intuitively displays the connection between the pump stations. Based on the hydraulic topology map, the water supply coverage of each pump station is analyzed, and cross-water supply areas with overlapping water supply areas of multiple pump stations are extracted. According to the configured pump station dependency analysis strategy, combined with factors such as the water volume ratio and water supply stability of the cross-water supply areas, the hydraulic dependency between pump stations is calculated. Based on the calculated hydraulic dependency, a hydraulic dependency matrix is ​​constructed to intuitively display the hydraulic dependency relationship between pump stations.

[0071] By constructing a hydraulic topology map, we can grasp the overall structural layout of the water system. Analyzing the cross-supply areas between pumping stations can determine the correlation range between multiple pumping stations, and further analyze the correlation between pumping stations. The hydraulic dependency matrix quantifies the dependency relationship between pumping stations, providing accurate data support for pumping station fault prediction and backup pumping station selection, thus improving the scientific nature of analysis and decision-making.

[0072] Simultaneously, sensor groups are configured at each pumping station to collect real-time data on pumping station operation, such as flow rate, pressure, motor temperature, and vibration frequency, at preset frequencies. The collected data reflects the real-time operating status of the pumping station. When a pumping station malfunctions, the relevant data will show anomalies. Combining this with a pre-constructed hydraulic topology map, a preset fault prediction model is used to analyze the collected data, predicting the faulty pumping station node with the abnormal data. The faulty pumping station cannot supply water normally, and its supply area will experience water shortages. By tracing the edges connected to the faulty pumping station in the hydraulic topology map, the corresponding water-shortage area is selected based on the pumping station's supply range. The real-time data collected by the sensors promptly reflects the operating status of the pumping station, providing real-time and accurate data support for fault identification. The fault prediction model can quickly and accurately identify faulty pumping station nodes, improving the efficiency and accuracy of fault prediction. Combining this with the hydraulic topology map to select water-shortage areas allows for timely analysis of the impact range of the fault on water supply, providing a clear target area for the scheduling of backup pumping stations.

[0073] Specifically, based on the hydraulic dependency matrix analysis, the fundamental hydraulic relationships between pumping stations are analyzed. According to a dynamic screening strategy, the matching degree between faulty and standby pumping stations in areas of overlapping water supply is quantified to determine whether standby pumping stations can effectively cover the faulty area. This includes, but is not limited to, water supply complementarity and pipeline connection convenience. Combining the predicted location of the water-scarce area, water consumption, and the real-time operating status of the standby pumping stations, a dynamic screening process is used to select standby pumping station combinations that can effectively cover the water-scarce area and have matching water supply capacity. A corresponding set of control commands is generated based on the operating parameters of the standby pumping stations. Setting standby pumping stations based on the hydraulic dependency matrix ensures coordinated water supply capabilities between standby pumping stations, improving the reliability of water supply. The dynamic screening strategy, combined with regional correlation and water-scarce area conditions, dynamically adjusts and optimizes the standby pumping station combinations. The selected standby pumping station combinations are more targeted and effective, enabling rapid response to water shortage problems. The generated set of standby pumping station control commands allows for real-time control of the standby pumping station's operating status, ensuring the orderly operation of the standby water supply process.

[0074] The set of control commands for standby pumping stations, real-time collected pumping station operation data, and fault evolution trend data (such as the rate of expansion of the fault's impact area and the rate of change in water shortage) are input into a preset fault evolution model. Based on the prediction results, the control commands are optimized and adjusted, including but not limited to adjusting the water supply flow and start-up sequence of each standby pumping station, resulting in an optimized set of control commands to control the actual operation of the standby pumping stations. By predicting water supply conditions, water supply problems can be detected in advance, and control commands can be optimized in a timely manner, making the operation of standby pumping stations more reasonable, avoiding standby pumping station failures, and improving the stability and reliability of the water supply system.

[0075] Furthermore, in the above technical solution, S1 includes:

[0076] S11: Based on the pre-set pump station coordinate data, unify the pump station coordinates to the same coordinate system, combine the pipeline connection data between pump stations, take each pump station as an independent node, establish associated edges according to the pipeline connection relationship and the bending flow of the pipeline, and obtain the hydraulic topology map.

[0077] S12: Extract the overlapping areas covered by pipeline connection paths from the hydraulic topology map, determine the intersection areas where multiple pumping stations share water supply, and obtain the intersection water supply areas between pumping stations; determine the water supply range boundary of each pumping station based on its water supply range.

[0078] S13: Calculate the priority of cross-supply areas based on area type, configure pump station dependency analysis strategy, calculate the hydraulic dependency between pump stations, and construct hydraulic dependency matrix.

[0079] like Figure 2As shown, the specific data on pipeline connection relationships includes, but is not limited to, connected pump station pairs (i.e., a pair of pump stations directly connected to the pipeline), pipeline types, etc. The constructed hydraulic topology map can intuitively display the physical structure of the water supply system, intuitively reflect the water flow transmission path between pump stations, and can quickly analyze the connection relationship and water flow path between pump stations, providing data support and structural support for the extraction of cross-water supply areas and the calculation of pump station dependence.

[0080] Specifically, the cross-regions of pipeline flow are screened out from the hydraulic topology map to obtain the cross-water supply areas between pump stations. The cross-water supply area refers to the area where the water supply ranges of multiple pump stations overlap and water flow can be interconnected through pipelines. The water supply in this area can be guaranteed by any one of the overlapping pump stations alone, or by multiple pump stations working together to supply water and replenish each other. The water supply stability of the cross-water supply area depends on the coordinated operation of multiple pump stations.

[0081] In the hydraulic topology graph, for each pipe with an associated edge, analyze whether there is flow interaction in the water supply range of the pump station connected to it, and analyze whether the pipe is used by two pump stations to supply water to the same overlapping area. For each pipe (i.e. "associated edge") in the hydraulic topology graph, analyze all the pump stations that supply water to the pipe, and whether there is a dynamic relationship of water flow overlap, distribution or complementarity in their respective water supply coverage areas. Confirm whether there is flow overlap in the water supply of different pump stations in the pipe and corresponding area.

[0082] For bidirectional pipelines, the water supply range of both pumping stations is analyzed simultaneously based on the water pressure conditions. For unidirectional pipelines, the water supply range of both the supply and receiving pumping stations is analyzed based on the water pressure conditions to obtain the pipeline's flow association range, i.e., which areas this pipeline specifically connects to for water supply. Each water supply range is overlaid to calculate the intersection area of ​​different pumping station water supply ranges, thus obtaining the cross-supply areas between pumping stations. By identifying the cross-supply areas of coordinated pumping station operations, data support is provided for the hydraulic dependence analysis between pumping stations.

[0083] Specifically, the priority of cross-supply areas is analyzed in conjunction with regional type analysis, and a pump station dependency analysis strategy is configured to calculate the hydraulic dependency between pump stations and construct a hydraulic dependency matrix. The priority of cross-supply areas is calculated based on regional type and water supply importance. Regional types include, but are not limited to, densely populated residential areas and industrial areas. Higher priority areas have a stronger dependency on the coordinated operation of pump stations. A dependency calculation strategy is configured in conjunction with priority analysis to quantify the degree of hydraulic dependency between pump stations caused by cross-regional water supply, and a hydraulic dependency matrix is ​​constructed to provide a dependency reference for the selection of backup pump stations. Combining regional priority analysis makes the dependency calculation more consistent with the actual water supply importance needs, improves the guarantee priority of key areas, and provides data support for the selection of backup pump stations through the quantitatively calculated hydraulic dependency matrix.

[0084] Furthermore, in the above technical solution, S13 includes:

[0085] S131: Determine the corresponding priority base weight according to the area type, calculate the corresponding area weight according to the ratio of the water supply coverage area of ​​the cross-area areas, and add the two weights to obtain the priority score of the cross-area water supply area.

[0086] S132: Analyze the water supply volume between pumping stations to the cross-supply area, obtain the proportion of water supply volume of each pumping station in the cross-area, and calculate the first hydraulic dependence between pumping stations based on priority.

[0087] S133: Collect the water pressure at the water inlet of each pumping station in the cross-water supply area through pressure sensors, calculate the water pressure difference of multiple pumping stations, and correct the first hydraulic dependence by combining the water pressure difference to obtain the second hydraulic dependence.

[0088] S134: Construct a hydraulic dependency matrix based on the second hydraulic dependency between pumping stations.

[0089] The priority base weights are determined according to the area type. Higher weights are set for areas with a large concentration of hospitals, intermediate weights are set for general residential areas, and lower weights are set for low-population-density residential areas. The coverage area of ​​each cross-water supply area is measured, and the corresponding area weight is calculated based on the ratio of the coverage areas. The larger the area, the greater the area weight.

[0090] Wherein, the coverage area ratio (R) = Sintersection ÷ Sparameter;

[0091] Actual coverage area of ​​the intersection zone (S_intersection): Only the effective water use coverage area within the intersection water supply area is counted (excluding non-water use areas such as green space and open space).

[0092] Reference total area (S parameter): Choose one of two and maintain consistency throughout the process: ① Single pump station scenario: the total water supply area of ​​a certain pump station; ② Dual pump station collaborative scenario: the sum of the total water supply areas of the two pump stations.

[0093] The priority score of each cross-water supply area is obtained by adding the priority base weight and the area weight. The priority of the cross-water supply area reflects the importance of the area in water supply security. The priority calculated by combining the area type and coverage area can take into account both the importance of water use and the scope of impact. By calculating the priority, cross-areas that need to be given priority can be quickly identified, thereby improving the pertinence of emergency response.

[0094] Specifically, the water supply volume of pumping stations to the cross-water supply area is analyzed. Combined with priority, the contribution relationship of different pumping stations to the water supply of the cross-water area is quantified, and the first hydraulic dependence between pumping stations is calculated. The first hydraulic dependence reflects the basic correlation between pumping stations due to the joint water supply to the cross-water supply area.

[0095] Collect historical water supply data for each cross-water supply area and water supply data for each pumping station participating in the water supply of that area, and calculate the water supply share of each pumping station in the cross-water supply area. The calculation formula is as follows: Water supply share of a certain pumping station = (Water supply of the pumping station to the cross-water supply area / Total water supply of the cross-water supply area) × 100%.

[0096] The water supply volume of the pumping station to the cross-region is the total water flow actually delivered to the cross-region through pipeline by a designated pumping station; the total water supply volume of the cross-region is the total water flow obtained by the cross-region from all water supply sources (all pumping stations, backup routes, etc.).

[0097] The first hydraulic dependence between pumping stations is calculated using the formula for calculating the first hydraulic dependence, as follows:

[0098]

[0099] In the formula, Let N be the first hydraulic dependence between pump station i and pump station j, and let N be the number of cross-supply zones between pump station i and pump station j. Priority is given to the cross-water supply area n between pump station i and pump station j. Let i be the proportion of water supplied by pump station i to the cross-supply area n. Let J be the percentage of water supplied by pump station J to the cross-supply area N.

[0100] The calculation of the first hydraulic dependence by combining the actual water supply ratio and regional priority reflects the collaborative relationship of pumping stations in cross-water supply areas. The calculation results are more consistent with the actual water supply scenario. By summarizing and calculating data from multiple cross-regional areas, the water supply relationship between pumping stations in different cross-water supply areas is comprehensively considered, avoiding the limitations of single-region analysis and obtaining more accurate hydraulic dependence analysis results.

[0101] After calculating the first hydraulic dependence, the water pressure difference at the water supply inlet of multiple pumping stations in the cross-supply area is calculated. The first hydraulic dependence is then corrected based on this water pressure difference to obtain the second hydraulic dependence. Water pressure is a key factor in ensuring water supply quality and stability. The water pressure difference at the water supply inlet of different pumping stations in the cross-supply area affects the actual coordinated water supply effect between pumping stations. Analyzing the impact of the water pressure difference on the water supply correlation and correcting the first hydraulic dependence yields a more accurate second hydraulic dependence, which more accurately reflects the hydraulic correlation between pumping stations. Water pressure at each pumping station in the cross-supply area is collected using pressure sensors. The average water pressure of pumping station i and pumping station j within a preset time period is statistically calculated. The absolute value of the difference between the average water pressure of pumping station i and the average water pressure of pumping station j is then calculated to obtain the water pressure difference.

[0102] Preferably, the pressure sensor is installed on the water supply inlet pipe of each pump station in the cross-water supply area. Preferably, the installation point needs to meet the requirements of being ≤5m away from the inlet node, at the center of the pipe cross section, and free from elbows or valve interference.

[0103] Specifically, the calculation rules for the water pressure difference correction coefficient are as follows: when the water pressure difference ≤ 0.1 MPa, it indicates good water pressure coordination, and the correction coefficient is set to 1; when 0.1 MPa < water pressure difference ≤ 0.3 MPa, it indicates moderate water pressure coordination, and the correction coefficient is set to 0.8; when the water pressure difference > 0.3 MPa, it indicates poor water pressure coordination, and the correction coefficient is set to 0.5. When there are multiple cross-supply areas between pump station i and pump station j, the average value of the correction coefficients for multiple cross-supply areas is calculated to obtain the total correction coefficient. The first hydraulic dependence is corrected by multiplying the total correction coefficient, resulting in the second hydraulic dependence. By correcting the first hydraulic dependence through water pressure difference, and considering the influence of hydraulic parameters on pump station correlation, the corrected second hydraulic dependence more accurately reflects the actual coordinated water supply capacity between pump stations. It better matches the hydraulic correlation state of the real water supply system, providing accurate data support for the selection of backup pump stations.

[0104] Based on the second hydraulic dependence between pump stations, a hydraulic dependence matrix is ​​constructed. The calculated second hydraulic dependence is arranged in matrix form. For two pump stations that do not have a common water supply area, the corresponding second hydraulic dependence is set to 0, and the corresponding matrix elements are filled with 0 to obtain the hydraulic dependence matrix. The hydraulic dependence matrix can intuitively display the dependence relationship between all pump stations, quickly check and analyze the relationship between pump stations, and improve the efficiency of fault prediction and analysis.

[0105] Furthermore, in the above technical solution, S2 includes:

[0106] S21: Configure corresponding sensors for components such as pumps, motors, and pipelines in the pumping station, collect the operating data of the pumping station in real time through the sensors, and preprocess the collected data;

[0107] S22: Based on the collected pump station operation data, extract the corresponding feature data, and the model predicts the operation status of each pump station based on the feature data to identify the pump station nodes that may fail.

[0108] S23: Combining the hydraulic topology map, starting from the faulty pump station node, analyze the water pressure change along the associated edges. When a location with a water pressure difference greater than the preset water pressure threshold is found, that location constitutes a water shortage area.

[0109] Preferably, an electromagnetic flow sensor is installed on the pump outlet pipe; pressure transmitters are installed on the inlet and outlet pipes respectively; a temperature sensor is installed on the motor housing; and a vibration sensor is installed on the pump bearing housing. The sensor group collects the pump station's operating data in real time, and the preprocessing methods include, but are not limited to, format conversion, outlier removal, and normalization. By collecting the pump station's operating data in real time through multiple sensors, the operating status can be monitored in real time, and comprehensive operating parameters of the pump station can be obtained, avoiding the limitations of single parameter monitoring and improving the accuracy of fault prediction results.

[0110] Specifically, the fault scenarios that need to be identified include equipment faults: motor overload, bearing wear, impeller blockage, seal leakage, abnormal pump body temperature; operational faults: sudden drop / rise in flow rate, pressure imbalance (too high / too low), abnormal power fluctuations, unstable voltage / current; and water supply faults: insufficient water supply, abnormal water supply range (such as supply interruption in overlapping areas), and coordination faults with other pumping stations.

[0111] Fault prediction models include, but are not limited to, XGBoost models. Feature parameters, including but not limited to mean, variance, and peak values ​​of vibration signals, are extracted from a large amount of historical pump station operation data. The XGBoost model is trained using historical feature data and corresponding fault probabilities and types to obtain a pre-trained model. Corresponding feature data is extracted from pump station operation data and input into the pre-trained XGBoost model. Based on this feature data, the model predicts the operating status of each pump station and identifies pump station nodes prone to failure. By using a multi-feature fusion fault prediction model to predict the fault status of pump stations, the accuracy of fault identification and early warning capabilities can be improved, reducing missed and false alarms. Timely prediction of the location and condition of faulty pump stations allows for the timely deployment of backup pump stations, mitigating the risk of large-scale water supply disruptions.

[0112] The hydraulic topology map includes the connections between pump stations and pipelines. A failed or inefficiently operating pump station will cause a drop in water pressure within its corresponding supply area. When a location with a pressure difference greater than a preset threshold is found, it indicates that other pump stations supply water at that location at a higher pressure than the failed pump station. Water can then be obtained from other pump stations, the search stops, and the searched areas constitute a water-scarce region. Path search based on the topology map, combined with pressure difference analysis, can accurately locate the affected area of ​​a failed pump station, avoiding overestimation or omission of water-scarce areas, and identifying accurate water-scarce region results. This provides accurate location information for controlling standby pump stations.

[0113] Furthermore, in the above technical solution, S23 includes:

[0114] S231: Starting from the faulty pump station node, calculate the water pressure difference on both sides of each associated pipeline, and search along the associated edge until the water pressure difference is greater than the preset water pressure threshold to obtain the set of water pressure attenuation paths.

[0115] S232: Extract the pipe corresponding to each path in each set of water pressure attenuation paths according to the hydraulic topology map, and take the area covered by the pipe as the first water shortage area.

[0116] S233: Based on the pipeline connection relationship of the first water shortage area and the water pressure difference, the water pressure fluctuation area is identified. The first water shortage area is expanded along the water pressure fluctuation area to obtain the second water shortage area.

[0117] S234: Combine the second water-scarce areas corresponding to each water pressure attenuation path to obtain the water-scarce region.

[0118] like Figure 3 As shown, after a faulty pump station stops or operates inefficiently, the water pressure in its water supply pipeline will gradually decrease along the direction of water flow. "Taking the faulty pump station as the starting point, along the direction of water flow in the hydraulic topology diagram, and combining the pressure / flow decay law, the pipeline-node chain of the impact diffusion is traced to determine the propagation path of multiple fault effects, providing path support for the screening of water shortage areas."

[0119] Locate the faulty pump station node in the hydraulic topology map, extract all pipes directly connected to the node (i.e., "associated edges"), and use each associated edge as an initial search path. Analyze the real-time water pressure data of all pressure monitoring points on each path, calculate the water pressure difference between the two ends of each pipe based on the predicted water pressure change of the faulty pump station node, and stop the search when the water pressure difference is greater than the preset water pressure threshold. The corresponding position is used as the end point of the path to obtain the set of water pressure attenuation paths.

[0120] By searching for water pressure attenuation paths based on topology maps, the propagation range and direction of fault effects can be presented intuitively, avoiding omission of branch pipes affected by faults and preventing disorderly spread of fault effects, thus obtaining accurate water pressure attenuation paths and providing accurate path references for screening water-scarce areas.

[0121] exist Figure 3 In this study, based on the faulty pump station, the nodes connected to the faulty pump station are analyzed in the hydraulic topology map. Water pressure is collected at each pump station node, and the water pressure value of the faulty pump station node is subtracted from the water pressure value of each pump station node to obtain the water pressure difference. If the calculated water pressure difference is greater than a preset water pressure threshold, it indicates that the water pressure between the two pump stations cannot meet the water demand, and this area is identified as a water-scarce region in the map.

[0122] Preferably, operational data is collected in real time by sensor arrays deployed at each pumping station. These sensor arrays include flow, pressure, temperature, and vibration sensors. After preprocessing, the operational data is input into a pre-trained fault prediction model. The model analyzes the correlation between real-time data characteristics and historical fault modes to identify and output pumping station nodes that are abnormal or predicted to fail, using these nodes as the starting point for faulty pumping stations. The hydraulic topology diagram includes every connecting pipe between two pumping station nodes. In actual operation, real-time data read by pressure sensors installed upstream and downstream of the pipes verifies and corrects the designed flow direction. According to basic principles of fluid mechanics, in a stable pipe network, the flow direction is always from high-pressure areas to low-pressure areas. By comparing the pressure values ​​at the pumping station nodes at both ends of the pipe in real time, if a pressure difference consistent with the designed flow direction is detected (upstream pressure > downstream pressure), the designed flow direction is confirmed as the current actual flow direction. If an abnormal pressure difference is detected, it may indicate abnormal pipe network operation or valve status changes, thus identifying the abnormal state. During path search, the designed flow direction is usually the primary consideration, combined with the latest effective pressure difference data to confirm or correct the search direction.

[0123] Furthermore, the fault causes a drop or loss of pump station output pressure, which in turn leads to a decrease in the inlet pressure of the directly connected downstream pipelines. Starting with the faulty pump station node, calculations are performed pipe by pipe along the determined water flow direction. For the currently analyzed pipeline, the real-time pressure values ​​of its two endpoints are read, and the pressure difference is calculated. A depth-first traversal is performed along the water flow direction of the topology, prioritizing the analysis of the next directly connected pipeline from the current node along the water flow direction. The pressure difference of this pipeline is calculated. If the pressure difference is less than or equal to a preset pressure threshold (which represents the minimum pressure gradient required to maintain normal water supply), the fault effect is considered to have propagated to this pipeline, and the end node of this pipeline is taken as the new current node, continuing the same analysis and search for its downstream pipelines. This process iterates sequentially downstream (i.e., in the direction of water flow) to simulate the gradual propagation of pressure disturbances in the pipeline network.

[0124] Specifically, if the water pressure difference in a pipeline exceeds a preset water pressure threshold, for example, if a pump station node is connected to multiple pump station nodes, and one pump station fails, other pump stations can provide water pressure. The pressure from other water sources can effectively compensate for the impact of the failure, and the main transmission of the failure's impact ends at this pump station node. The complete sequence of nodes and edges from the failure origin to the upstream node of the pipeline is recorded as a water pressure attenuation path. Since a failed pump station connects to multiple downstream main pipelines, the search process proceeds in parallel from the origin along different branches. When all search branches meet the termination condition and the process ends, a set of water pressure attenuation paths is obtained, where each water pressure attenuation path represents a channel through which the impact of the failure pressure is transmitted along the pipeline network topology.

[0125] The geographical coverage area of ​​all pipelines along each water pressure attenuation path is extracted as the first water-deficient area. Further analysis is performed on other hydraulically connected pipelines surrounding these paths. By calculating the fluctuation values ​​of the water pressure in these surrounding pipelines compared to their historical normal average water pressure, if the fluctuation value exceeds the water pressure fluctuation threshold, it is considered that the pressure fluctuation caused by the fault has indirectly affected these areas. These fluctuating areas are then merged with the corresponding first water-deficient areas to obtain the second water-deficient area. All second water-deficient areas are spatially combined and overlapping parts are removed to obtain the complete water-deficient region.

[0126] Specifically, the area covered by each water pressure attenuation path is designated as the first water shortage area. For each path in the set of water pressure attenuation paths, the pipe area corresponding to the pipe of that path is extracted from the hydraulic topology map, and the area covered by the pipe area is designated as the first water shortage area. Identifying the first water shortage area based on the coverage of the water pressure attenuation path can ensure that no core affected area of ​​the faulty pumping station is missed and accurately identify the initial water shortage area.

[0127] Water pressure fluctuations are conductive. Areas connected by pipes around the first water-deficient area will experience secondary fluctuations due to water pressure imbalance, leading to insufficient water supply. Based on the pipe connections along the first water-deficient area and the water pressure difference, water pressure fluctuation regions are identified. The first water-deficient area is then expanded to obtain the second water-deficient area. Starting from the boundary pipe of the first water-deficient area, all connected branch pipes are searched in the hydraulic topology map as associated fluctuation pipes. Real-time water pressure data on the associated fluctuation pipes are analyzed, and the absolute value of the difference between the current water pressure and the historical average water pressure is calculated to obtain the water pressure fluctuation value. A water pressure fluctuation threshold is set according to the water-deficient area identification accuracy. Pipes with water pressure fluctuation values ​​greater than the water pressure fluctuation threshold are selected as water pressure fluctuation regions. The first water-deficient area is then expanded along the water pressure fluctuation region to obtain the second water-deficient area.

[0128] By analyzing the transmissibility of water pressure fluctuations, the water-scarce area can be expanded to avoid overlooking indirectly water-scarce areas, achieving full coverage of the impact and improving the comprehensiveness of water scarcity identification. This allows for timely implementation of corresponding preventative measures in complete water-scarce affected areas.

[0129] The second water shortage areas corresponding to different water pressure attenuation paths may overlap or be adjacent. By combining the second water shortage areas corresponding to each water pressure attenuation path, the water shortage area is obtained. In the process of combination, the duplicate areas are removed and the degree of water shortage in each area is determined. The combined water shortage area is more accurate and can comprehensively reflect the water shortage area affected by the faulty pump station, providing an accurate basis for the dynamic screening and scheduling of standby pump stations.

[0130] Furthermore, in the above technical solution, S3 includes:

[0131] S31: Based on the hydraulic dependence matrix, extract the second hydraulic dependence value between the faulty pump station and other pump stations, and set the backup pump station at the midpoint between the two pump stations whose second hydraulic dependence value is lower than the preset second hydraulic dependence threshold.

[0132] S32: Based on the priority of water-scarce areas, select the corresponding backup pumping stations for water supply to water-scarce areas with high priority. Based on the correlation between backup pumping stations and faulty pumping stations, select the first backup pumping station combination that can cover water-scarce areas with high correlation.

[0133] S33: Based on the real-time status of the standby pumping stations and the conflict between the standby pumping stations required by the faulty pumping stations, a second standby pumping station combination is obtained by dynamically adjusting and eliminating standby pumping stations with abnormal status.

[0134] S34: Analyze the real-time operating parameters of each standby pump station in the second standby pump station combination, generate specific control commands for each pump station, and accurately control the operating status of each standby pump station.

[0135] In this embodiment, based on the hydraulic dependency matrix, multiple backup pumping stations are set up between the main pumping stations. The hydraulic dependency matrix reflects the degree of hydraulic correlation between the pumping stations. The higher the dependency, the stronger the correlation between the pumping stations in terms of water supply range, water pressure coordination, etc. A low second hydraulic dependency value indicates that the water supply area between the two pumping stations is far apart. When one pumping station fails, the water supply capacity of the other pumping station cannot cover the water shortage area caused by the failure of the failed pumping station. At this time, a backup pumping station is set up at the midpoint between the two pumping stations so that the backup pumping station can cover the service range of both main pumping stations at the same time. This can make up for the insufficient water supply range of the other pumping station when the pumping station fails and improve the reliability of emergency water supply. Specifically, based on the priority of water-scarce areas, the correlation between standby pumping stations and faulty pumping stations is calculated. The first standby pumping station combination that can cover the water-scarce areas is selected according to the correlation from high to low. Based on the priority of water-scarce areas, corresponding standby pumping stations are selected for water supply to areas with higher priority. Combining the correlation between standby pumping stations and faulty pumping stations, the selection is sorted from high to low, which can quickly identify the first standby pumping station combination that can effectively cover the water-scarce areas and has the fewest number of stations. By sorting and selecting based on correlation, standby pumping stations with high adaptability are prioritized, improving water supply coverage efficiency and ensuring that the water supply capacity of the standby pumping station combination meets the actual water shortage needs, avoiding insufficient water supply capacity or resource waste.

[0136] Based on the real-time status of the backup pumping stations and the conflicting backup pumping station requirements of the faulty pumping stations, the first backup pumping station combination is dynamically adjusted to obtain the second backup pumping station combination. The real-time status of the backup pumping stations and the conflicting backup pumping station requirements of the faulty pumping stations affect the availability of the backup pumping station combination. The real-time status of the backup pumping stations includes whether they are operating normally and the current load. The conflicting backup pumping station conditions include multiple faulty pumping stations simultaneously requiring the same backup pumping station. Through dynamic adjustment, backup pumping stations with abnormal status can be eliminated, and conflicting pumping stations can be dynamically adjusted to ensure the actual operational reliability of the backup pumping station combination and improve the overall emergency response capability of the system to water-scarce areas.

[0137] After determining the second backup pump station combination, the operating status of each backup pump station in the combination is analyzed to generate a set of backup pump station control commands. Based on the real-time operating parameters of each pump station in the second backup combination, the adjustable flow range of the pump station is analyzed to determine the optimal operating range of the backup pump station. Specific control commands are generated for each pump station, including but not limited to target flow, target outlet pressure, and speed setting, resulting in a set of backup pump station control commands. Task allocation based on the operating status of the pump stations allows the backup pump stations to operate within a safe and efficient range, improving the matching accuracy of water supply demand and accurately controlling the operating status of each backup pump station.

[0138] Furthermore, in the above technical solution, S32 includes:

[0139] S321: Calculate the first correlation degree based on the cross-water supply area between the standby pumping station and the faulty pumping station;

[0140] S322: Analyze the real-time water pressure of the standby pump station, calculate the radius of the water supply area, analyze the coverage between the water supply area and the water shortage area, calculate the area of ​​the overlapping area, and obtain the second correlation degree between the standby pump station and the corresponding faulty pump station based on the ratio between the area of ​​the area and the area of ​​each water shortage area.

[0141] S323: Calculate the comprehensive correlation between each standby pumping station and the faulty pumping station by combining the first correlation degree and the second correlation degree;

[0142] S324: Based on the priority of the water-scarce areas, match the backup pumping station with the highest comprehensive correlation for each water-scarce area in descending order of priority, until the water supply area of ​​all matched backup pumping stations can cover the water-scarce areas, thus obtaining the first backup pumping station combination.

[0143] Specifically, the overlapping water supply areas between the standby pumping station and the faulty pumping station are extracted from the hydraulic topology map. The proportion of the overlapping water supply area in the water shortage area caused by the faulty pumping station is calculated by dividing the size of the overlapping water shortage area under the fault by the size of the total water shortage area caused by the fault, and the first correlation degree is obtained. Based on the overlapping water supply areas between the standby pumping station and the faulty pumping station, the correlation between the standby pumping station and the faulty pumping station can be quickly analyzed, providing accurate data support for the selection of standby pumping stations.

[0144] Specifically, the water supply radius of the standby pumping station is calculated based on the real-time water pressure. The water supply area of ​​the standby pumping station is determined with the standby pumping station as the center and the water supply radius as the radius. The coverage between the water supply area and each water-scarce area is analyzed, the overlapping area is calculated, and the ratio between the overlapping area and the area of ​​each water-scarce area is calculated to obtain the second correlation degree between the standby pumping station and the corresponding faulty pumping station. The water supply radius is calculated using the real-time water pressure of the standby pumping station to reflect its real-time water supply capacity. The overlap between the water supply area and the water-scarce area is analyzed, and the corresponding pumping station correlation degree is calculated to obtain an accurate correlation degree value.

[0145] Specifically, based on the fundamental role of historical collaborative relationships and the impact of current coverage capabilities, weight values ​​for the first and second correlation degrees are set respectively. The first and second correlation degrees are then weighted and summed to obtain the comprehensive correlation degree between each standby pump station and the faulty pump station. By integrating historical correlation and real-time coverage capabilities, the comprehensive correlation degree results are more comprehensive, avoiding the limitations of single-dimensional evaluation and obtaining accurate correlation degree analysis results.

[0146] Specifically, based on the region type and coverage area of ​​the water-scarce areas, the priority of each water-scarce area is calculated. These areas are then sorted from highest to lowest priority score to obtain the order of water-scarce areas. The backup pumping stations with the highest overall correlation are matched to these water-scarce areas in sequence. For the second-ranked water-scarce area, it is checked whether the water supply area of ​​the currently selected backup pumping station already covers its water-scarce area. If it does, the backup pumping station is not selected; otherwise, the backup pumping station with the highest overall correlation is selected, and so on, until the water supply area of ​​all matched backup pumping stations can cover the water-scarce area, resulting in the first backup pumping station combination. Prioritizing the matching of high-priority water-scarce areas with corresponding backup pumping stations ensures water supply security in critical areas. This progressively accumulating matching method avoids resource waste and selects the backup pumping station combination with the fewest required numbers to meet the needs of the water-scarce areas, thus preventing resource waste.

[0147] Furthermore, in the above technical solution, S33 includes:

[0148] S331: Select core parameters reflecting the operating status of the standby pumping stations from the real-time status data of the standby pumping stations, construct a status vector based on the core parameters, analyze the operating status of each standby pumping station through a preset abnormal pumping station identification model, and filter out abnormal pumping stations based on the status vector through the model.

[0149] S332: Based on the conflict situation of the backup pump station required by the faulty pump station, the conflicting pump station is selected, and the corresponding water-scarce area is taken as the conflict area. The hydraulic coupling degree between the conflicting pump station and the faulty pump station corresponding to the conflict area is calculated. Among the conflicting pump stations, the pump station with high hydraulic coupling degree is selected to supply water to the conflict area, and the third pump station combination is obtained.

[0150] S333: Remove abnormal pump stations from the first backup pump station combination, and update the pump stations through the third pump station combination. Replace the corresponding pump station information in the first backup pump station combination with the backup pump station information in the third pump station combination to obtain the second backup pump station combination.

[0151] Specifically, core parameters include, but are not limited to, real-time flow rate, outlet pressure, motor temperature, and vibration amplitude. The abnormal pump station identification model includes, but is not limited to, a random forest model. This model is trained using extensive historical state data from the pump stations to obtain a pre-trained model. The state vector of each standby pump station is then input into the pre-trained random forest model, which identifies standby pump stations with abnormal states based on the state vectors. By comprehensively evaluating the pump station's operating status through multi-parameter state vectors, scattered real-time data is transformed into a comprehensive state profile that is "quantifiable, traceable, and predictable." This avoids misjudgments or omissions caused by the one-sidedness of single-parameter evaluation, improves the accuracy of abnormal identification results, and promptly filters out abnormal pump stations, preventing them from entering the standby combination and causing water supply interruptions or insufficient pressure, thus ensuring the reliability of the combination.

[0152] Specifically, when multiple faulty pumping stations simultaneously need to call the same backup pumping station, pumping station conflicts occur. The backup pumping station requirements of all faulty pumping stations are analyzed, and the backup pumping stations selected by each faulty pumping station are compared. The backup pumping stations selected by multiple faulty pumping stations are identified as conflicting pumping stations, and the water-scarce areas corresponding to these conflicting pumping stations are designated as conflict areas. The hydraulic dependence of the conflicting pumping station and the faulty pumping station is summed with the overlap rate of the cross-region to calculate the hydraulic coupling degree between them. The correspondence between the conflicting pumping stations and each conflict area is sorted from high to low hydraulic coupling degree. Conflicting pumping stations are preferentially assigned to the conflict areas with the highest coupling degree. For conflict areas that do not receive pumping station assignments, the pumping station with the second highest comprehensive correlation degree and no conflict is selected from the backup pumping stations to obtain the third pumping station combination. The adaptability of the hydraulic coupling quantification can be achieved by first clarifying the conflict type (replenishment / …). (Overload), and then select core indicators such as pressure matching degree, flow matching degree, and pipeline transmission efficiency to calculate the coupling degree; at the same time, use multi-parameter state vectors to supplement non-hydraulic dimension evaluation, and finally form a comprehensive adaptability judgment; by calculating the hydraulic coupling degree, the adaptability of conflict pumping stations to different conflict areas can be quantified, and high coupling degree pumping stations can be allocated first to ensure that conflict areas obtain the most suitable backup resources, while avoiding conflict leading to some areas having no backup resources, and ensuring water supply coverage for all conflict areas.

[0153] Specifically, removing abnormal pump stations eliminates potential risks, ensuring that each pump station in the combination has reliable water supply capacity. Updating and replacing pump stations can avoid pump station conflicts and ensure that the total water supply capacity of the backup pump station combination meets the demand.

[0154] Furthermore, in the above technical solution, S4 includes:

[0155] S41: Combining the set of control commands for standby pump stations with real-time pump station operation data, the water supply situation is predicted through a preset fault evolution model, and risk areas of abnormal water supply are screened out.

[0156] S42: Optimize the control commands based on the risk area to obtain an optimized set of control commands, and control the operation of the standby pump station.

[0157] Specifically, key parameters from the control command set of standby pump stations are collected, including target flow rate, target pressure, and number of pumps started for each standby pump station. Real-time operating data is collected, including but not limited to the current actual flow rate, outlet pressure, and motor load rate of the standby pump stations, as well as real-time pressure and flow rate data of key monitoring points in the water-scarce area. The fault evolution model includes, but is not limited to, an LSTM model. The LSTM model is trained using a large amount of historical pump station operating data to obtain a pre-trained LSTM model. This model can simulate the fault development trend and water supply changes after the standby pump stations are started, predicting the overall operating status of the water supply system and the water supply recovery status in the water-scarce area. Based on the predicted water supply situation, risk areas where the water supply does not meet the needs of the water-scarce area are identified. By predicting and identifying risk areas of water supply anomalies in advance through the fault evolution model, control parameters can be adjusted and optimized in a timely manner to ensure the stability and reliability of the water supply.

[0158] Specifically, the risk area information is input into the particle swarm optimization model, and the model iteratively solves for the optimized control commands, resulting in an optimized control command set. By optimizing the control commands in real time, the control commands can meet the actual water shortage situation in the water-scarce area, thereby improving the resource utilization rate of the standby pumping station.

[0159] like Figure 4 As shown, an automatic switching control system is used to implement a method for predicting equipment failures in a smart water pumping station, comprising:

[0160] The hydraulic analysis module constructs a hydraulic topology map and a hydraulic dependency matrix. The hydraulic topology map uses preset pump stations as nodes, establishes associated edges based on the pipeline connection relationship between pump stations, extracts the cross-water supply areas of pump stations based on the hydraulic topology map, and configures a pump station dependency analysis strategy to calculate the hydraulic dependency between pump stations and construct a hydraulic dependency matrix.

[0161] The water shortage area prediction module collects pump station operation data in real time through sensors configured in each pump station, combines it with the hydraulic topology map, identifies faulty pump station nodes through a preset fault prediction model, and identifies water shortage areas along the associated edges.

[0162] The backup pump station screening module sets up multiple backup pump stations among the pump stations based on the hydraulic dependency matrix, configures a dynamic screening strategy for pump stations, analyzes the regional correlation between the faulty pump station and the backup pump station, dynamically screens the backup pump station combination in combination with the water shortage area, and generates a set of backup pump station control instructions.

[0163] The control command optimization module combines the set of control commands for the standby pump station with real-time pump station operation data. It predicts the water supply situation through a preset fault evolution model and optimizes the control commands to obtain an optimized set of control commands. This optimizes the operation of the standby pump station and enables fault prediction and automatic mode switching for the smart water pump station.

[0164] In this embodiment, the hydraulic analysis module uses preset pump stations as nodes, establishes associated edges based on the pipeline connection relationships between pump stations to form a hydraulic topology graph, extracts cross-supply areas, calculates the hydraulic dependence between pump stations by configuring a dependency analysis strategy and constructs a matrix, analyzes the association relationships between pump stations, and provides data support for pump station fault prediction and automatic switching systems. The topology graph can clearly analyze pump station fault information, and the dependency matrix provides dependency support for backup pump station screening and fault impact analysis, thereby improving the accuracy of system analysis and decision-making.

[0165] The water shortage area prediction module collects operational data in real time through sensors configured at the pumping station. Combined with the hydraulic topology map, it uses a preset fault prediction model to accurately identify faulty pumping station nodes and identify water shortage areas along the associated edges. This allows for rapid fault detection and impact range localization. Real-time sensor data ensures timely fault identification, and the water shortage area identification combined with the topology map can quickly identify the affected area of ​​water supply, providing a clear target for dynamic selection of backup pumping stations and reducing the impact of faults on water supply.

[0166] The backup pump station selection module sets up multiple backup pump stations based on the hydraulic dependency matrix. Through dynamic selection strategy, it analyzes the regional correlation between the faulty pump station and the backup pump station, and dynamically selects backup pump station combinations based on the water shortage area conditions and generates a set of control commands. This can ensure the rational scheduling of backup resources. The dynamic selection strategy, which combines regional correlation and water shortage area needs, can select the optimal pump station combination. The generated control commands provide definite operational guidance for the operation of backup pump stations, improving the efficiency of emergency water supply.

[0167] The control command optimization module inputs the set of control commands for the standby pumping station and real-time operating data into a preset fault evolution model to predict water supply conditions and optimize control commands. The resulting optimized command set controls the operation of the standby pumping station, enabling dynamic and precise regulation of the water supply system and early prediction of potential water supply anomalies, providing a basis for command optimization. Optimization and adjustment for risk areas ensure that the operating parameters of the standby pumping station match actual needs, preventing the escalation of water supply problems. At the same time, the optimized commands make the standby pumping station more stable and efficient, enhancing the system's emergency response capability and water supply reliability.

[0168] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the invention by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the invention should be included within the scope of protection of the invention.

Claims

1. A method for predicting equipment failures in intelligent water pumping stations, characterized in that, Includes the following steps: S1: Construct a hydraulic topology map to show the connections between pump stations, and construct a hydraulic dependency matrix to show the hydraulic dependencies between pump stations; S2: Sensors are configured on each pumping station to identify faulty pumping station nodes through a preset fault prediction model and to identify water-scarce areas along the associated edges; S3: Based on the hydraulic dependency matrix, set up backup pumping stations, dynamically adjust the combination of backup pumping stations according to the dynamic screening strategy, regional correlation and regional water shortage, and generate a set of backup pumping station control commands. S4: Combines the set of control commands for the backup pump station with real-time pump station operation data to control the operation of the backup pump station, so as to predict faults and automatically switch modes for the smart water pump station. S1 includes: S11: Based on the pre-set pump station coordinate data, unify the pump station coordinates to the same coordinate system, combine the pipeline connection data between pump stations, take each pump station as an independent node, establish associated edges according to the pipeline connection relationship and the bending flow of the pipeline, and obtain the hydraulic topology map. S12: Extract the overlapping areas covered by pipeline connection paths from the hydraulic topology map, determine the intersection areas where multiple pumping stations share water supply, and obtain the intersection water supply areas between pumping stations; S13: Calculate the priority of cross-supply areas based on area type, configure pump station dependency analysis strategy, calculate hydraulic dependency between pump stations, and construct hydraulic dependency matrix; S13 includes: S131: Determine the corresponding priority base weight according to the area type, calculate the corresponding area weight according to the ratio of the water supply coverage area of ​​the cross-area areas, and add the two weights to obtain the priority score of the cross-area water supply area. S132: Analyze the water supply volume between pumping stations to the cross-supply area, obtain the proportion of water supply volume of each pumping station in the cross-area, and calculate the first hydraulic dependence between pumping stations based on priority. S133: Collect the water pressure at the water inlet of each pumping station in the cross-water supply area through pressure sensors, calculate the water pressure difference of multiple pumping stations, and correct the first hydraulic dependence by combining the water pressure difference to obtain the second hydraulic dependence. S134: Construct a hydraulic dependency matrix based on the second hydraulic dependency between pumping stations.

2. The method for predicting equipment failures in a smart water pumping station according to claim 1, characterized in that, S2 includes: S21: Configure corresponding sensors for the pumps, motors, and pipelines in the pumping station, collect the operating data of the pumping station in real time through the sensors, and preprocess the collected data; S22: Based on the collected pump station operation data, extract the corresponding feature data, and the model predicts the operation status of each pump station based on the feature data to identify the pump station nodes that may fail. S23: Combining the hydraulic topology map, starting from the faulty pump station node, analyze the water pressure change along the associated edges. When a location with a water pressure difference greater than the preset water pressure threshold is found, that location constitutes a water shortage area.

3. The method for predicting equipment failures in a smart water pumping station according to claim 2, characterized in that, S23 includes: S231: Starting from the faulty pump station node, calculate the water pressure difference on both sides of each associated pipeline, and search along the associated edge until the water pressure difference is greater than the preset water pressure threshold to obtain the set of water pressure attenuation paths. S232: Extract the pipe corresponding to each path in each set of water pressure attenuation paths according to the hydraulic topology map, and take the area covered by the pipe as the first water shortage area. S233: Based on the pipeline connection relationship of the first water shortage area and the water pressure difference, the water pressure fluctuation area is identified. The first water shortage area is expanded along the water pressure fluctuation area to obtain the second water shortage area. S234: Combine the second water-scarce areas corresponding to each water pressure attenuation path to obtain the water-scarce region.

4. The method for predicting equipment failures in a smart water pumping station according to claim 1, characterized in that, S3 includes: S31: Based on the hydraulic dependence matrix, extract the second hydraulic dependence value between the faulty pump station and other pump stations, and set the backup pump station at the midpoint between the two pump stations whose second hydraulic dependence value is lower than the preset second hydraulic dependence threshold. S32: Based on the priority of water-scarce areas, select the corresponding backup pumping stations for water supply to water-scarce areas with high priority. Based on the correlation between backup pumping stations and faulty pumping stations, select the first backup pumping station combination that can cover water-scarce areas with high correlation. S33: Based on the real-time status of the standby pumping stations and the conflict between the standby pumping stations required by the faulty pumping stations, a second standby pumping station combination is obtained by dynamically adjusting and eliminating standby pumping stations with abnormal status. S34: Analyze the real-time operating parameters of each standby pump station in the second standby pump station combination, generate specific control commands for each pump station, and accurately control the operating status of each standby pump station.

5. The method for predicting equipment failures in a smart water pumping station according to claim 4, characterized in that, S32 includes: S321: Calculate the first correlation degree based on the cross-water supply area between the standby pumping station and the faulty pumping station; S322: Analyze the real-time water pressure of the standby pump station, calculate the radius of the water supply area, analyze the coverage between the water supply area and the water shortage area, calculate the area of ​​the overlapping area, and obtain the second correlation degree between the standby pump station and the corresponding faulty pump station based on the ratio between the area of ​​the area and the area of ​​each water shortage area. S323: Calculate the comprehensive correlation between each standby pumping station and the faulty pumping station by combining the first correlation degree and the second correlation degree; S324: Based on the priority of the water-scarce areas, match the backup pumping station with the highest comprehensive correlation for each water-scarce area in descending order of priority, until the water supply area of ​​all matched backup pumping stations can cover the water-scarce areas, thus obtaining the first backup pumping station combination.

6. The method for predicting equipment failures in a smart water pumping station according to claim 4, characterized in that, S33 includes: S331: Select core parameters reflecting the operating status of the standby pumping stations from the real-time status data of the standby pumping stations, construct a status vector based on the core parameters, analyze the operating status of each standby pumping station through a preset abnormal pumping station identification model, and filter out abnormal pumping stations based on the status vector through the model. S332: Based on the conflict situation of the backup pump station required by the faulty pump station, the conflicting pump station is selected, and the corresponding water-scarce area is taken as the conflict area. The hydraulic coupling degree between the conflicting pump station and the faulty pump station corresponding to the conflict area is calculated. Among the conflicting pump stations, the pump station with high hydraulic coupling degree is selected to supply water to the conflict area, and the third pump station combination is obtained. S333: Remove abnormal pump stations from the first backup pump station combination, and update the pump stations through the third pump station combination. Replace the corresponding pump station information in the first backup pump station combination with the backup pump station information in the third pump station combination to obtain the second backup pump station combination.

7. The method for predicting equipment failures in a smart water pumping station according to claim 1, characterized in that, S4 includes: S41: Combining the set of control commands for standby pump stations with real-time pump station operation data, the water supply situation is predicted through a preset fault evolution model, and risk areas of abnormal water supply are screened out. S42: Optimize the control commands based on the risk area to obtain an optimized set of control commands, and control the operation of the standby pump station.

8. An automatic switching control system for implementing the intelligent water pumping station equipment fault prediction method as described in claim 1, comprising: The hydraulic analysis module constructs a hydraulic topology map and a hydraulic dependency matrix. The hydraulic topology map uses preset pump stations as nodes, establishes associated edges based on the pipeline connection relationship between pump stations, extracts the cross-water supply area of ​​pump stations based on the hydraulic topology map, configures the pump station dependency analysis strategy, calculates the hydraulic dependency between pump stations, and constructs the hydraulic dependency matrix. The water shortage area prediction module collects pump station operation data in real time through sensors configured in each pump station, combines the hydraulic topology map, identifies faulty pump station nodes through a preset fault prediction model, and identifies water shortage areas along the associated edges. The backup pump station screening module sets up multiple backup pump stations among the pump stations based on the hydraulic dependence matrix, configures a dynamic screening strategy for pump stations to analyze the regional correlation between the faulty pump station and the backup pump station, dynamically screens the backup pump station combination in combination with the water shortage area, and generates a set of backup pump station control commands. The control command optimization module combines the set of control commands for the backup pumping station with real-time pumping station operation data, predicts the water supply situation through a preset fault evolution model, and optimizes the control commands to obtain an optimized set of control commands. This optimizes the operation of the backup pumping station and enables fault prediction and automatic mode switching for the smart water pumping station.

Images