Water plant control method and control device

Abstract

The application provides a water plant control method and a control device. The control device estimates the water supply situation of at least one water plant supplying water to a target area in a future period according to historical water supply data of the target area, divides the future period into multiple target periods, determines the water supply demand corresponding to each target period, for each water plant, determines multiple target scheduling strategies from an expert library to meet the scheduling demand of multiple scenarios according to the water supply demand of the multiple target periods and the current working condition of the water plant, determines a corresponding scoring strategy according to the current scheduling demand to select the optimal scheduling strategy of the water plant from the multiple target scheduling strategies, and schedules the operation of the water pump in each water plant in a timely manner to improve the scheduling efficiency of the control device. Since each scheduling strategy in the expert library is determined by standardized analysis of historical data and prediction data, the accuracy and reliability of the optimal scheduling strategy provided by the expert library are ensured.

Description

Technical Field

[0001] This application relates to the field of pump control, and more particularly to a water plant control method and control equipment. Background Technology

[0002] Water supply from waterworks is provided to users through a network of pipes laid at various levels throughout the city. The water company's dispatch center coordinates the water supply from multiple waterworks within its jurisdiction to ensure that the water supply in that area meets the needs of users.

[0003] When dispatching water plants, the dispatch center typically uses real-time monitoring software to monitor the water production process and output indicators of each plant. Based on the data displayed on the software and their own dispatching experience, dispatchers adjust the water supply situation of each plant within the dispatch center's jurisdiction.

[0004] However, the above scheduling strategies rely heavily on the scheduling experience of the schedulers, resulting in different effects of scheduling strategies determined by different schedulers. The schedulers can only generate scheduling strategies based on historical data, which has a lag effect, resulting in low scheduling efficiency. Furthermore, the number of scheduling strategies that the schedulers can provide is limited, which cannot meet the scheduling needs of various scenarios. Summary of the Invention

[0005] This application provides a water plant control method and control equipment to solve the above-mentioned technical problems.

[0006] In a first aspect, this application provides a water plant control method, the method comprising:

[0007] Obtain historical water supply data and scheduling requirements for the target area;

[0008] Based on the historical water supply data, estimate the water supply data of at least one water plant supplying water to the target area in multiple sub-periods within a future period; the water supply data includes water supply volume and water supply pressure, and the future period includes the multiple sub-periods, each of which is independent of the others;

[0009] Based on the water supply data of the multiple sub-periods, the future period is divided into multiple target periods, and the water supply demand corresponding to each target period is determined; each target period includes at least one sub-period.

[0010] For each water plant, based on the water supply demand for the multiple target time periods and the current operating conditions of the water plant, various target scheduling strategies for the water plant are obtained from the expert database.

[0011] Based on the scheduling requirements, a corresponding scoring strategy is determined, and the optimal scheduling strategy is selected from the multiple target scheduling strategies using the scoring strategy.

[0012] In the above technical solution, the control equipment acquires historical water supply data for the target area. Based on the historical water supply data, it estimates the water supply volume and pressure of at least one water plant supplying water to the target area in multiple sub-periods within a future period. This allows the control equipment to determine the water supply situation of the target area in advance. Based on the water supply data for multiple future sub-periods, the future period is divided into multiple target periods, and the corresponding water supply demand for each target period is determined. For each water plant, based on the water supply demand of multiple target periods and the current operating conditions of the water plant, multiple target scheduling strategies are determined from an expert database to meet the scheduling needs of various scenarios. The control equipment can determine the corresponding scoring strategy based on the current scheduling needs and use this scoring strategy to select the optimal scheduling strategy for the water plant from multiple target scheduling strategies. This allows for timely scheduling of the pumps in each water plant, improving the scheduling efficiency of the control equipment. Since each scheduling strategy in the expert database is determined through standardized analysis of historical and predicted data, the accuracy and reliability of the optimal scheduling strategy provided by the expert database are guaranteed.

[0013] Optionally, the historical water supply data includes hourly water consumption and environmental data from multiple pipeline monitoring points within the target area over multiple historical periods; the target area is equipped with multiple pipelines, and each pipeline has at least one pipeline monitoring point;

[0014] Based on the historical water supply data, estimate the water supply data for at least one water plant supplying water to the target area over multiple sub-periods in the future, including:

[0015] For each of the pipelines, a water consumption prediction model is used to process the hourly water consumption and environmental data of multiple network monitoring points associated with the pipeline in multiple historical periods, and to estimate the hourly water consumption of the pipeline in multiple sub-periods in the future period.

[0016] For each water plant, the sum of the hourly water consumption of at least one pipeline associated with the water plant in any sub-time period is taken as the water supply of the water plant in that sub-time period.

[0017] The water consumption prediction model includes a time-dimensional data prediction model.

[0018] Optionally, the water consumption prediction model includes a first LSTM algorithm, a first Prophet algorithm, and a moving weighted algorithm;

[0019] The water consumption prediction model is used to process the hourly water consumption and environmental data of multiple pipeline network monitoring points associated with the pipeline over multiple historical periods, and to estimate the hourly water consumption of the pipeline in multiple sub-periods within a future period, including:

[0020] The first LSTM algorithm is used to process the hourly water consumption and environmental data of multiple pipeline monitoring points associated with the pipeline in multiple historical time periods to obtain the first predicted hourly water consumption of the pipeline in multiple sub-time periods in the future.

[0021] The first Prophet algorithm is used to process the hourly water consumption and environmental data of multiple pipeline monitoring points associated with the pipeline in multiple historical time periods to obtain the second predicted hourly water consumption of the pipeline in multiple sub-time periods in the future.

[0022] The moving weighted algorithm is used to process the hourly water consumption and environmental data of multiple pipeline monitoring points associated with the pipeline in multiple historical time periods to obtain the third predicted hourly water consumption of the pipeline in multiple sub-time periods in the future.

[0023] The water consumption during the first, second, and third predicted periods is fused to obtain the hourly water consumption of the pipeline in multiple sub-periods within a future time period.

[0024] In the above technical solution, the historical hourly water consumption and environmental data of each pipeline monitoring point are analyzed using a multi-time dimension data prediction model to obtain the corresponding predicted hourly water consumption. By fusing multiple predicted hourly water consumption data, the accuracy of the prediction data is ensured, thereby ensuring the accuracy of the optimal scheduling strategy provided by the expert database.

[0025] Optionally, the method further includes:

[0026] For each water plant, the hourly water consumption of the water plant in multiple sub-periods within the future time period is queried from the water plant outlet water characteristic table corresponding to the water plant to obtain the water supply pressure required by the water plant to provide the water supply in each sub-period.

[0027] The water plant effluent characteristic table includes the correspondence between the water plant providing at least one water supply volume and the water supply pressure providing that water supply volume.

[0028] Optionally, the historical water supply data includes water supply pressure and environmental data from multiple pipeline monitoring points within the target area over multiple historical periods; the target area is equipped with multiple pipelines, and each pipeline has at least one network monitoring point;

[0029] Based on the historical water supply data, estimate the water supply data for at least one water plant supplying water to the target area over multiple sub-periods in the future, including:

[0030] For each pipeline monitoring point, the water supply pressure and environmental data of the pipeline monitoring point in multiple historical periods are processed using the pipeline monitoring point pressure prediction model to obtain the water supply pressure of the pipeline monitoring point in multiple sub-periods in the future period;

[0031] For each pipeline, the water supply pressure of at least one network monitoring point associated with the pipeline is processed in multiple sub-periods in the future time period using a pipeline pressure model to obtain the water supply pressure of the pipeline in multiple sub-periods in the future time period.

[0032] For each water plant, the average water supply pressure of the pipelines associated with the water plant in any sub-period is taken as the water supply pressure of the water plant in that sub-period.

[0033] The pressure prediction model for pipeline monitoring points includes a time-dimensional prediction model.

[0034] Optionally, the pressure prediction model for the pipeline monitoring points includes a second LSTM algorithm, a second Prophet algorithm, and a weighted average algorithm;

[0035] The water supply pressure and environmental data of the pipeline monitoring points in multiple historical time periods are processed using a pipeline monitoring point pressure prediction model to obtain the water supply pressure of the monitoring points in multiple sub-time periods in the future, including:

[0036] The second LSTM algorithm is used to process the water supply pressure and environmental data of the pipeline monitoring point in multiple historical periods to obtain the first predicted water supply pressure of the pipeline monitoring point in multiple sub-periods in the future period;

[0037] The second Prophet algorithm is used to process the water supply pressure and environmental data of the pipeline monitoring point in multiple historical periods to obtain the second predicted water supply pressure of the pipeline monitoring point in multiple sub-periods in the future period;

[0038] The weighted average algorithm is used to process the water supply pressure and environmental data of the pipeline monitoring points in multiple historical periods to obtain the third predicted water supply pressure of the pipeline monitoring points in multiple sub-periods in the future period;

[0039] The first predicted water supply pressure, the second predicted water supply pressure, and the third predicted water supply pressure are fused to obtain the water supply pressure of the pipeline monitoring point in multiple sub-periods within a future period.

[0040] In the above technical solution, the predicted water supply pressure for future periods is estimated by utilizing historical water supply pressure and environmental data from each pipeline monitoring point. Furthermore, based on the predicted water pressure at different pipeline monitoring points and the correlation between pipelines and monitoring points for future periods, the pressure of each pipeline in the future period is determined. Finally, based on the relationship between pipelines and water plants, the pressure of each water plant in the future period is determined. Since the pipeline monitoring point pressure prediction model and the fitted pipeline-monitoring point correlation are both trained based on a large amount of historical data, the accuracy of the fitting results can be guaranteed, thereby ensuring the accuracy of the subsequent use of an expert database to provide the optimal scheduling strategy.

[0041] Optionally, the method further includes:

[0042] The water supply pressure of each pipeline in each sub-period is processed using a pipeline flow-pressure model to obtain the water supply volume of the pipeline in each sub-period.

[0043] For each water plant, the sum of the water supply of at least one pipeline associated with the water plant in any sub-time period is taken as the water supply of the water plant in that sub-time period.

[0044] The pipeline flow-pressure model is a model fitted using the historical water supply pressure and corresponding water supply volume of each pipeline.

[0045] Optionally, the method further includes:

[0046] For each water plant, the water supply pressure of the water plant in multiple sub-periods within the future time period is queried in the water plant outlet water characteristic table corresponding to the water plant to obtain the water supply volume provided by the water plant using the water supply pressure in each sub-period.

[0047] The water plant effluent characteristic table includes a correspondence between at least one water supply pressure and the amount of water supplied by the water plant under each of the stated water supply pressures.

[0048] Optionally, based on the water supply data of the multiple sub-time periods, the future time period is divided into multiple target time periods, and the water supply demand corresponding to each target time period is determined, including:

[0049] The water supply data for the multiple sub-periods are processed using the dynamic least squares method, and the future period is divided into multiple target periods.

[0050] For each target time period, the sum of the water supply volume of each sub-time period within the target time period is determined as the water supply volume of the target time period, and the average water supply pressure of each sub-time period within the target time period is determined as the water supply pressure of the target time period; the water supply demand of the target time period includes the water supply volume and the water supply pressure of the target time period.

[0051] Optionally, the current operating status of the water plant includes the status of multiple water pumps in the water plant, and the status of the water pumps includes running status and stopped status;

[0052] For each water plant, based on the water supply demand for multiple target time periods and the current operating status of the water plant, various target scheduling strategies for the water plant are obtained from an expert database, including:

[0053] The water supply range is determined based on the water supply volume during the target period and the preset fluctuation error of the water supply volume.

[0054] The available water supply, the water supply pressure, and the status of multiple water pumps are input into the expert database to obtain various target scheduling strategies; each target scheduling strategy includes the target status of multiple water pumps in the water plant.

[0055] The water supply pressure generated by the water plant applying the target scheduling strategy is the water supply pressure for the target time period, and the available water supply is within the range of the target water supply.

[0056] In the above technical solution, for the water supply volume and pressure of the target time period to be achieved, the control equipment expands the water supply volume within the allowable preset fluctuation error of the water supply volume to the water supply volume range of the target time period, and inputs multiple water supply volumes, water pressures and the current operating conditions of the water plant within this water supply volume range into the expert database for querying, to obtain multiple target scheduling strategies that can achieve the input parameters, and comprehensively evaluates the target scheduling strategy from multiple aspects to determine the optimal scheduling strategy, thus ensuring the accuracy of determining the optimal scheduling strategy.

[0057] Optionally, before obtaining the various target scheduling strategies for the water plant from the expert database, the method further includes:

[0058] For each water plant, obtain the flow-head characteristic curve and flow-power characteristic curve of each water pump in the water plant; the water pump includes a constant speed pump or a variable frequency pump;

[0059] Based on the historical water supply pressure of the water plant, a water supply pressure range is determined; the maximum value of the water supply pressure range is the maximum value of the historical water supply pressure, and the minimum value of the water supply pressure range is the minimum value of the historical water supply pressure.

[0060] Within the constraints, under any state of each water pump, calculate the operating parameters of the water plant, construct a mapping relationship between the operating parameters and the states of multiple water pumps in the water plant, and store the mapping relationship in the expert database;

[0061] The constraints include: the frequency of each variable frequency pump is within the high-efficiency operating frequency range, the flow deviation of each water pump is within the preset deviation range, and the water supply pressure of the water plant is within the water supply pressure range.

[0062] The arbitrary state includes each water pump being in an operating state or a stopped state. When the variable frequency pump is in an operating state, the variable frequency pump is at any frequency within the high-efficiency operating frequency range.

[0063] The operating parameters of the water plant include the unit water consumption for supply, the unit water consumption for distribution, the pressure deviation of the variable frequency pump, the flow deviation, and the total power.

[0064] In the above technical solution, when constructing the expert database, the control equipment determines various pump operating conditions within the achievable water supply pressure range based on the characteristic curves of each pump in each water plant. Based on these different pump operating conditions, it calculates relevant parameters such as water supply volume and water plant power consumption. This ensures that when querying the expert database for target scheduling strategies based on the water supply volume, water supply pressure, and current operating conditions for the target time period, all achievable target scheduling strategies that meet the given water supply volume and pressure can be found. Furthermore, a comprehensive analysis is conducted based on the operating conditions to be achieved in the target scheduling strategy, the current operating conditions, the power consumption generated by achieving the target scheduling strategy, and the deviation in water supply volume and pressure for the target time period. This ensures that the determined optimal scheduling strategy is more reliable.

[0065] Optionally, the target area can be divided into multiple target sub-areas, and each target sub-area includes multiple pipeline measurement points;

[0066] The method further includes:

[0067] Based on the water supply pressure at each of the aforementioned pipeline monitoring points, obtain the water supply pressure of each target sub-area;

[0068] When the water supply pressure of any target sub-area is not within the normal pressure range, an alarm is triggered, and the pressure to be adjusted in the target sub-area is obtained, wherein the pressure to be adjusted is within the normal pressure range.

[0069] The corrected regional water supply pressure and the time information of the sampling of the water supply pressure are processed using a regional water plant pressure fitting model to obtain the water supply pressure to be adjusted for each water plant; the corrected regional water supply pressure includes the pressure to be adjusted for the target sub-region with pressure anomalies and the water supply pressure for other target sub-regions.

[0070] Input the water supply pressure to be adjusted from each water plant into the expert database to obtain the optimal scheduling strategy.

[0071] In a second aspect, this application provides a control device, including: a processor and a memory communicatively connected to the processor;

[0072] The memory stores computer instructions;

[0073] The processor is used to implement the methods involved in the first aspect when executing computer instructions.

[0074] This application provides a water plant control method and control equipment. The control equipment estimates the water supply situation of at least one water plant supplying water to the target area in the future time period based on historical water supply data of the target area, divides the future time period into multiple target time periods, determines the water supply demand corresponding to each target time period, and for each water plant, determines multiple target scheduling strategies from an expert database based on the water supply demand of multiple target time periods and the current operating condition of the water plant to meet the scheduling needs of multiple scenarios. It also determines the corresponding scoring strategy based on the current scheduling needs to select the optimal scheduling strategy for the water plant from multiple target scheduling strategies, and timely schedules the operation of water pumps in each water plant to improve the scheduling efficiency of the control equipment. Since each scheduling strategy in the expert database is determined by standardized analysis of historical data and predicted data, the accuracy and reliability of the optimal scheduling strategy provided by the expert database are guaranteed. Attached Figure Description

[0075] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0076] Figure 1 This is a schematic diagram of a water plant control method provided in this application according to an exemplary embodiment;

[0077] Figure 2 This is a schematic flowchart of a water plant control method provided in this application according to an exemplary embodiment;

[0078] Figure 3 This is a flowchart illustrating a method for estimating water supply data over a future time period according to an exemplary embodiment of this application.

[0079] Figure 4 This is a schematic diagram of the structure of a water consumption prediction model provided in this application according to an exemplary embodiment;

[0080] Figure 5 This is a flowchart illustrating a method for estimating water supply data over a future time period according to another exemplary embodiment of this application;

[0081] Figure 6 This is a schematic diagram of the structure of a pipeline monitoring point pressure prediction model provided in this application according to an exemplary embodiment;

[0082] Figure 7 This is a schematic diagram of the structure of a pipeline pressure model provided in this application according to an exemplary embodiment;

[0083] Figure 8 This is a flowchart illustrating the process of determining the optimal scheduling strategy according to an exemplary embodiment of this application;

[0084] Figure 9This is a schematic diagram illustrating the construction of an expert library according to an exemplary embodiment of this application;

[0085] Figure 10 This is a schematic flowchart of a method for handling abnormal regional water supply pressure according to an exemplary embodiment of this application;

[0086] Figure 11 This is a schematic diagram of a water plant control method provided in this application according to another exemplary embodiment;

[0087] Figure 12 This is a schematic diagram of the structure of a water plant control device provided in accordance with an exemplary embodiment of this application;

[0088] Figure 13 This is a schematic diagram of the structure of a control device provided according to an embodiment of this application.

[0089] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0090] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0091] This application provides a water plant control method and control equipment, aiming to solve the technical problems of low efficiency and poor reliability in dispatching water supply from various water plants by the dispatch center. The technical concept of this application is as follows: the control equipment in the dispatch center estimates the water supply data for future periods based on historical water supply data from each water plant and the monitoring points of the pipeline network associated with each water plant. It then utilizes an expert database to provide multiple dispatching strategies for the dispatcher to choose from, allowing the dispatcher to select the most suitable scheme as the optimal dispatching strategy based on dispatching needs, thus enabling timely advance control of each water plant, improving the dispatching efficiency of the dispatch center, and enhancing the reliability of the dispatching strategies due to the introduction of the expert database.

[0092] Figure 1 This is a schematic diagram illustrating a scenario of a water plant control method provided in this application according to an exemplary embodiment. Figure 1As shown, the system includes a water company control device 10, a first water plant 11, and a second water plant 12. The target area 16 is the jurisdiction of the water company control device 10, which controls the first water plant 11 and the second water plant 12 to supply water to the target area 16.

[0093] The target area 16 includes a first water supply device 14, a second water supply device 13, and a third water supply device 15. The first water plant 11 supplies water to the second water supply device 13 through one pipeline and to the first water supply device 14 through two pipelines. The second water plant 12 supplies water to the first water supply device 14 through one pipeline and to the third water supply device 15 through two pipelines.

[0094] Each pipeline is equipped with at least one monitoring point (as shown by the black dot in the figure) to monitor water supply data at important nodes along the pipeline. In one embodiment, the water supply data includes water pressure and water supply volume.

[0095] The water company control equipment 10 can obtain water supply data and environmental data sampled at each pipeline monitoring point through monitoring software. It can also obtain information such as the operation status of water pumps in each water plant and the water supply data of each pipeline in the water plant.

[0096] The water company control device 10 has an expert database, or the expert database is located in the server. The water company control device 10 communicates with the server and performs scheduling with the expert database in the server.

[0097] The water company control equipment 10 also stores historical water supply data. It uses the historical water supply data to predict the planned water supply data for future periods, and inputs the planned water supply data and the current operating conditions of each water plant into the expert database to determine the optimal scheduling strategy for each water plant. This allows for the regulation of each water plant before the arrival of the future period, which not only improves the regulation efficiency of the water company control equipment 10, but also ensures the reliability of the determined scheduling strategy.

[0098] Based on the above application scenarios and in conjunction with the accompanying drawings, some embodiments of this application will be described in detail below. Where there is no conflict between the embodiments, the following embodiments and features can be combined with each other. Furthermore, the timing of the steps in the following method embodiments is merely an example and not a strict limitation.

[0099] Figure 2 This is a schematic flowchart of a water plant control method provided in this application according to an exemplary embodiment, such as... Figure 2 As shown, it includes:

[0100] S201. Obtain historical water supply data and scheduling requirements for the target area.

[0101] Multiple pipelines are located within the target area, and each pipeline has at least one network monitoring point.

[0102] Historical water supply data includes hourly water consumption and environmental data from multiple pipeline monitoring points within the target area over multiple historical periods.

[0103] Among them, multiple historical periods are those prior to the current moment, including the current moment and multiple moments adjacent to the current moment.

[0104] Scheduling requirements are at least one factor that dispatchers consider when scheduling water pumps in a water plant. For example: the deviation between the water supply pressure that the pumps can provide and the water supply pressure to be achieved in the scheduling strategy is small; the deviation between the water supply volume that the pumps can provide and the water supply volume to be achieved in the scheduling strategy is small; the total power consumption of water supply is low; and the number of pump start-ups and shutdowns is low.

[0105] S202. Based on historical water supply data, estimate the water supply data of at least one water plant supplying water to the target area for multiple sub-periods in the future period.

[0106] Water supply data includes water supply volume and water pressure.

[0107] The control equipment estimates the water supply data for multiple sub-periods in the future time period adjacent to the current time, based on the historical water supply data of the target area.

[0108] The future time period includes multiple sub-time periods, and these sub-time periods do not overlap with each other.

[0109] For example, if we define the future day after the previous moment as the future time period, and divide the future day into hours, then the future time period includes 24 sub-time periods.

[0110] S203. Based on the water supply data of multiple sub-periods, divide the future period into multiple target periods and determine the water supply demand corresponding to each target period.

[0111] Each target time period includes at least one sub-time period. For example, the next day can be divided into four target time periods, each lasting more than or equal to one hour.

[0112] The control equipment determines the water supply demand for each target time period based on the estimated water supply data for future time periods within at least one sub-time period within each target time period, and uses the water supply demand as the planned water supply pressure and planned water supply volume for that target time period.

[0113] In one embodiment, for each target time period, the sum of the estimated water supply corresponding to each sub-time period within the target time period is taken as the planned water supply for the target time period, and the average of the estimated water supply pressure corresponding to each sub-time period is taken as the planned water supply pressure for the target time period.

[0114] S204. For each water plant, based on the water supply demand for multiple target time periods and the current operating conditions of the water plant, obtain multiple target scheduling strategies for the water plant from the expert database.

[0115] The operating conditions of the water plant include the operating status of each water pump. The water pumps include constant-speed pumps and variable-frequency pumps. For constant-speed pumps, the operating status is either in operation or stopped; for variable-frequency pumps, the operating status is either in operation at a certain frequency or stopped.

[0116] The expert database includes multiple preset scheduling strategies, each of which assumes that each pump in the water plant is in one operating condition. These multiple preset scheduling strategies involve arranging and combining multiple pumps in any given operating condition to obtain various operating conditions for the water plant.

[0117] In the expert database, for each preset scheduling strategy, the characteristic curves and energy consumption of each water pump are used to determine the water supply volume, water pressure and power required when the water plant applies various preset scheduling strategies.

[0118] For each water plant, the water supply demand for multiple target time periods after the current time obtained in step S203, as well as the current operating status of the water plant, are input into the expert database and matched with multiple preset scheduling strategies in the expert database to determine the multiple target scheduling strategies required to achieve the aforementioned water supply demand.

[0119] S205. Based on the scheduling requirements, determine the corresponding scoring strategy, and use the scoring strategy to select the optimal scheduling strategy from multiple target scheduling strategies.

[0120] Different schedulers have different scheduling preferences and corresponding scheduling needs. Therefore, when selecting the optimal scheduling strategy, the weights of each scheduling indicator in the target scheduling strategy will vary in the scoring strategy adopted.

[0121] In one embodiment, the scheduling indicators in the schedulable strategy include: pressure deviation relative to the planned water supply pressure, flow deviation relative to the planned water supply volume, water supply unit consumption, water distribution unit consumption, total operating power, and the number of pump start-ups and shutdowns relative to the target operating condition corresponding to the target scheduling strategy for achieving the current operating condition.

[0122] The above evaluation data are weighted and summed to obtain the total score of each target scheduling strategy. The data are then sorted in ascending order, and the control device can determine the target scheduling strategy with the smallest total score as the optimal scheduling strategy.

[0123] In other embodiments, the dispatcher's dispatching needs for different water plants may not be exactly the same.

[0124] The scheduling requirements include the scheduling sub-requirements corresponding to each water plant. For each water plant, a corresponding scoring strategy is determined based on the corresponding scheduling sub-requirements, and the optimal scheduling strategy is obtained from the multiple target scheduling strategies corresponding to that water plant using the scoring strategy.

[0125] In the above technical solution, the control equipment acquires historical water supply data for the target area. Based on the historical water supply data, it estimates the water supply volume and pressure of at least one water plant supplying water to the target area in multiple sub-periods within a future period. This allows the control equipment to determine the water supply situation of the target area in advance. Based on the water supply data for multiple future sub-periods, the future period is divided into multiple target periods, and the corresponding water supply demand for each target period is determined. For each water plant, based on the water supply demand of multiple target periods and the current operating conditions of the water plant, multiple target scheduling strategies are determined from an expert database to meet the scheduling needs of various scenarios. The control equipment can determine the corresponding scoring strategy based on the current scheduling needs and use this scoring strategy to select the optimal scheduling strategy for the water plant from multiple target scheduling strategies. This allows for timely scheduling of the pumps in each water plant, improving the scheduling efficiency of the control equipment. Since each scheduling strategy in the expert database is determined through standardized analysis of historical and predicted data, the accuracy and reliability of the optimal scheduling strategy provided by the expert database are guaranteed.

[0126] The estimation of water supply data for future periods can be achieved through several of the following embodiments.

[0127] Figure 3 This is a flowchart illustrating a method for estimating water supply data over a future time period according to an exemplary embodiment of this application, as shown below. Figure 3 As shown, it includes:

[0128] S301. For each pipeline, use a water consumption prediction model to process the hourly water consumption and environmental data of multiple network monitoring points associated with the pipeline in multiple historical periods, and estimate the hourly water consumption of the pipeline in multiple sub-periods in the future period.

[0129] Water consumption prediction models include time-dimensional data prediction models.

[0130] In one embodiment, a schematic diagram of the water consumption prediction model is shown below. Figure 4 As shown, the water consumption prediction model includes a first LSTM algorithm, a first Prophet algorithm, and a moving weighted algorithm. The control equipment uses the first LSTM algorithm to process the hourly water consumption and environmental data from multiple pipeline monitoring points associated with the pipeline over multiple historical periods to obtain the first predicted hourly water consumption for multiple sub-periods in the future period. It then uses the first Prophet algorithm to process the hourly water consumption and environmental data from multiple pipeline monitoring points associated with the pipeline over multiple historical periods to obtain the second predicted hourly water consumption for multiple sub-periods in the future period.

[0131] The first LSTM algorithm and the first Prophet algorithm are both trained using historical data from multiple first time periods and corresponding historical data from multiple second time periods. In each set of historical data from the first time period and the second time period, the maximum time of the first time period is less than or equal to the minimum time of the second time period.

[0132] As time progresses, when a future time period is reached, the current water supply data is added to the training set of the two algorithms as historical water supply data. The two algorithms are then retrained using the updated training set at preset time intervals to ensure the accuracy of the algorithms.

[0133] The system also utilizes a moving weighted algorithm to process hourly water consumption and environmental data from multiple pipeline monitoring points across various historical time periods, obtaining a third predicted hourly water consumption for multiple sub-time periods in the future. The first, second, and third predicted hourly water consumption are then fused to obtain the hourly water consumption for multiple sub-time periods within the future timeframe.

[0134] In one embodiment, data fusion means weighting and summing the predicted hourly water consumption, for example, using the sum of 0.2 times the first predicted hourly water consumption, 0.3 times the second predicted hourly water consumption, and 0.5 times the third predicted hourly water consumption as the hourly water consumption of the pipeline in the future period.

[0135] S302. For each water plant, the sum of the hourly water consumption of at least one pipeline associated with the water plant in any sub-period shall be taken as the water supply of the water plant in the sub-period.

[0136] At least one pipeline associated with the water plant indicates a pipeline that is connected to the water plant.

[0137] For example, if a water plant is connected to two pipelines, and the hourly water consumption of each pipeline in two sub-time periods (the first sub-time period and the second sub-time period) has been predicted, then the sum of the hourly water consumption of the two pipelines in the first sub-time period is determined as the water supply of the water plant in the first sub-time period, and the sum of the hourly water consumption of the two pipelines in the second sub-time period is determined as the water supply of the water plant in the second sub-time period.

[0138] S303. For each water plant, the hourly water consumption of the water plant in multiple sub-periods within the future time period is queried in the corresponding water plant outlet water characteristic table to obtain the water supply pressure required by the water plant to provide water supply in each sub-period.

[0139] The water plant effluent characteristics table includes the correspondence between the water plant providing at least one water supply volume and the water supply pressure that provides that volume.

[0140] The water plant's effluent characteristics table is a statistical analysis of the plant's historical water supply data, determining the correspondence between the water supply volume and pressure that the plant can provide.

[0141] In the above technical solution, the historical hourly water consumption and environmental data of each pipeline monitoring point are analyzed using a multi-time dimension data prediction model to obtain the corresponding predicted hourly water consumption. By fusing multiple predicted hourly water consumption data, the accuracy of the prediction data is ensured, thereby ensuring the accuracy of the optimal scheduling strategy provided by the expert database.

[0142] Figure 5 This is a flowchart illustrating a method for estimating water supply data over a future time period according to another exemplary embodiment of this application, such as... Figure 5 As shown, it includes:

[0143] S401. For each pipeline monitoring point, use the pipeline monitoring point pressure prediction model to process the water supply pressure and environmental data of the pipeline monitoring point in multiple historical periods to obtain the water supply pressure of the pipeline monitoring point in multiple sub-periods in the future period.

[0144] The pressure prediction model for pipeline monitoring points includes a time-dimensional prediction model.

[0145] In one embodiment, a schematic diagram of the pressure prediction model for pipeline monitoring points is shown below. Figure 6 As shown, the pressure prediction model for pipeline monitoring points includes the second LSTM algorithm, the second Prophet algorithm, and the weighted average algorithm.

[0146] The second LSTM algorithm is used to process water supply pressure and environmental data from pipeline monitoring points across multiple historical time periods to obtain the first predicted water supply pressure for multiple sub-time periods within the future time period. The second Prophet algorithm is then used to process the same data to obtain the second predicted water supply pressure for the same data within the future time period. The training process for this second LSTM algorithm and the second Prophet algorithm is similar to that of the first LSTM algorithm and the first Prophet algorithm in step S301, the difference being that the algorithm training in step S301 is for hourly water consumption, while the algorithm training in this step is for water supply pressure.

[0147] A weighted average algorithm is used to process water supply pressure and environmental data from pipeline monitoring points across multiple historical time periods to obtain the third predicted water supply pressure for multiple sub-time periods within the future timeframe. The first, second, and third predicted water supply pressures are then fused to obtain the water supply pressure for the pipeline monitoring points within the future timeframe.

[0148] In one embodiment, data fusion is a weighted summation, wherein the explanation of data fusion has been explained in detail in step S301 and will not be repeated here.

[0149] S402. For each pipeline, use the pipeline pressure model to process the water supply pressure of at least one network monitoring point associated with the pipeline in multiple sub-periods in the future time period, and obtain the water supply pressure of the pipeline in multiple sub-periods in the future time period.

[0150] The pipeline pressure model uses the water supply pressure at each monitoring point in the pipeline network to fit and set the water supply pressure of the pipeline at that monitoring point.

[0151] In one embodiment, the pipeline pressure model adopts the XGBoost model, the structural diagram of which is shown below. Figure 7 As shown. When using this pipeline pressure model, the basic data of water supply pressure at at least one pipeline monitoring point associated with the pipeline in multiple sub-periods within a future period is constructed. The basic data is used as the input of the xgboost model, and the sum of the results obtained from processing multiple weak learners is used to determine the water supply pressure of the pipeline.

[0152] The XGBoost model constructs a strong learner F using multiple weak learners f1 to fN trained on historical data of pipeline water supply pressure and the water supply pressure at monitoring points along those pipelines. During training, weak learner f1 is trained using the historical water supply pressure at each monitoring point along the pipeline and the historical water supply pressure of the pipeline itself. The residual between the estimated water supply pressure of each pipeline by weak learner f1 and the actual historical water supply pressure is then used to train weak learner f2, and so on, until the training conditions for weak learners are met. These multiple weak learners are then used to construct the strong learner.

[0153] S403. For each water plant, the average water supply pressure of the pipelines associated with the water plant in any sub-period shall be taken as the water supply pressure of the water plant in that sub-period.

[0154] S404. Use the pipeline flow and pressure model to process the water supply pressure of each pipeline in each sub-period and obtain the water supply volume of the pipeline in each sub-period.

[0155] The pipeline flow-pressure model is a model fitted using the historical water supply pressure and corresponding water supply volume of each pipeline. In one embodiment, the pipeline flow-pressure model is a linear model.

[0156] S405. For each water plant, the sum of the water supply of at least one pipeline associated with the water plant in any sub-period shall be taken as the water supply of the water plant in that sub-period.

[0157] In other embodiments, in step S403, the water supply pressure of each water plant in each sub-time period is determined, and the corresponding water supply volume can also be determined through the water plant effluent characteristic table. More specifically, this includes:

[0158] For each water plant, the water supply pressure for multiple sub-periods within the future time period is looked up in the corresponding water plant outlet characteristic table to obtain the water supply volume provided by the water plant using the supply pressure in each sub-period. The water plant outlet characteristic table includes at least one correspondence between the supply pressure and the water supply volume provided by the water plant under each supply pressure.

[0159] In the above technical solution, the predicted water supply pressure for future periods is estimated by utilizing historical water supply pressure and environmental data from each pipeline monitoring point. Furthermore, based on the predicted water pressure at different pipeline monitoring points and the correlation between pipelines and monitoring points for future periods, the pressure of each pipeline in the future period is determined. Finally, based on the relationship between pipelines and water plants, the pressure of each water plant in the future period is determined. Since the pipeline monitoring point pressure prediction model and the fitted pipeline-monitoring point correlation are both trained based on a large amount of historical data, the accuracy of the fitting results can be guaranteed, thereby ensuring the accuracy of the subsequent use of an expert database to provide the optimal scheduling strategy.

[0160] After determining the water supply data for multiple future sub-periods, the control equipment uses an expert database to process the water supply data and obtain the optimal scheduling strategy. Figure 8 The following is a flowchart illustrating the process for determining the optimal scheduling strategy according to an exemplary embodiment of this application, as shown in the figure, including:

[0161] S501. Use dynamic least squares method to process water supply data for multiple sub-periods and divide the future period into multiple target periods.

[0162] Using water supply volume as the independent variable and water supply pressure as the dependent variable, a linear fit is performed to obtain the fitting function. The predicted water supply pressure corresponding to each water supply volume in the fitting function is determined. The absolute value of the difference between each predicted water supply pressure and the corresponding actual water supply pressure is calculated. When the absolute value of the difference is greater than a preset threshold, the time period corresponding to the water supply volume corresponding to the difference is determined as the start time period of a sub-time period.

[0163] S502. For each target time period, the sum of the water supply of each sub-time period in the target time period is determined as the water supply of the target time period, and the average water supply pressure of each sub-time period in the target time period is determined as the water supply pressure of the target time period.

[0164] The water supply demand for the target period includes the water supply volume and water supply pressure for the target period.

[0165] S503. For each water plant, input the water supply demand corresponding to multiple target time periods and the status of multiple water pumps in the water plant into the expert database to obtain multiple target scheduling strategies to achieve the water supply demand.

[0166] The status of multiple water pumps in the water plant represents the current operating condition of the water plant, and each target scheduling strategy includes the target status of multiple water pumps in the water plant.

[0167] More specifically, for each water plant, the control equipment determines the water supply range based on the target time period's water supply volume and the preset fluctuation error of the water supply volume. It then inputs the available water supply volume, water pressure, and the status of multiple water pumps into an expert database to obtain various target scheduling strategies. The water pressure generated by the water plant applying each target scheduling strategy is the water supply pressure for the target time period, and the available water supply volume falls within the specified range.

[0168] S504. Use the comprehensive scoring method to score multiple target scheduling strategies, and determine the optimal scheduling strategy based on the scores.

[0169] Based on the comprehensive evaluation data and according to actual needs, the controller selects other scheduling strategies as the optimal control strategy.

[0170] Before step S503, the expert database needs to be constructed. Below, we will combine... Figure 9 The process of constructing the expert database is explained. Among other things, Figure 9 This is a schematic diagram illustrating the construction of an expert library according to an exemplary embodiment of this application.

[0171] For each water plant, obtain the flow-head characteristic curve and flow-power characteristic curve of each water pump in the water plant, where the water pumps include constant speed pumps or variable frequency pumps.

[0172] The flow-head characteristic curve of a constant-speed pump is H = H x +S x Q 2 The flow-head characteristic curve of the variable frequency pump is H = H x S 2 +S x Q 2 Where H represents the actual head of the pump (m), Q represents the flow rate of the pump, and H x S represents the virtual total head generated by the pump when Q=0. x This represents the virtual resistance coefficient within the pump body.

[0173] The average values ​​of the coefficients in the above characteristic curves were determined by fitting standard operating data provided by the pump manufacturer.

[0174] Based on the historical water supply pressure of the water plant, the water supply pressure range is determined. The maximum value of the water supply pressure range is the maximum value of the historical water supply pressure, and the minimum value of the water supply pressure range is the minimum value of the historical water supply pressure.

[0175] Within the constraints, and with each pump in any state, calculate the operating parameters of the water plant, construct the mapping relationship between the operating parameters and the states of multiple pumps within the water plant, and store the mapping relationship in the expert database.

[0176] The constraints include: the frequency of each variable frequency pump is within the high-efficiency operating frequency range; the flow deviation of each pump is within the preset deviation range; and the water supply pressure of the water plant is within the supply pressure range. The preset deviation is ±20%.

[0177] "Any state" includes both the pumps being in operation and stopped. When the variable frequency pump is in operation, it operates at any frequency within its high-efficiency operating frequency range. The high-efficiency operating frequency range is 35–50 Hz.

[0178] The operating parameters of a water plant include water supply unit consumption, water distribution unit consumption, variable frequency pump pressure deviation, flow deviation, and total power.

[0179] In addition, while each water plant operates according to the optimal scheduling strategy, the control equipment also monitors the water supply situation in each area within its jurisdiction. When it is determined that the water supply situation in at least one area is abnormal, the operation of each water plant is adjusted again using the expert database. The following is a detailed explanation of the handling process for abnormal regional water supply pressure.

[0180] Figure 10 This is a schematic flowchart of a method for handling abnormal regional water supply pressure according to an exemplary embodiment of this application, as shown below. Figure 10 As shown, it includes:

[0181] S601. Obtain the water supply pressure of each target sub-area based on the water supply pressure of each pipeline monitoring point.

[0182] In one embodiment, the water supply pressure of the target area is the average water supply pressure of each network monitoring point in the area.

[0183] S602. When the water supply pressure of any target sub-area is not within the normal pressure range, an alarm is triggered, and the pressure to be adjusted in the target sub-area is obtained.

[0184] When obtaining the pressure to be adjusted in the target sub-region, the pressure can be adjusted according to the standard pressure interval based on the current abnormal water supply pressure until the adjusted water supply pressure is within the normal pressure range.

[0185] When the abnormal water supply pressure is greater than the maximum value of the normal pressure range, the abnormal water supply pressure is cyclically reduced by the standard pressure interval until the adjusted pressure value is less than the maximum value of the normal pressure range.

[0186] When the abnormal water supply pressure is less than the minimum value of the normal pressure range, the abnormal water supply pressure is cyclically increased by the standard pressure interval until the adjusted pressure value is greater than the minimum value of the normal pressure range.

[0187] In one embodiment, the standard pressure interval is 0.5 MPa.

[0188] S603. Using the regional water plant pressure fitting model to process the corrected regional water supply pressure and the time information of the sampled water supply pressure, the water supply pressure to be adjusted for each water plant is obtained.

[0189] The corrected regional water supply pressure includes the pressure to be adjusted in the target sub-region with abnormal pressure, as well as the water supply pressure in other target sub-regions.

[0190] The regional water plant pressure fitting model is obtained by training the model using the correspondence between the regional pressure and the pressure of each water plant supplying water to the region in each sub-period of historical water supply data.

[0191] In one embodiment, the pressure fitting model for the regional water plant is a neural network model.

[0192] S604. Input the water supply pressure to be adjusted for each water plant into the expert database to obtain the optimal scheduling strategy.

[0193] The water supply pressure to be regulated at each water plant is input into the expert database to obtain multiple target scheduling strategies. The optimal scheduling strategy is determined by a comprehensive scoring method.

[0194] The following explains the complete control process of each water plant by the expert database. A schematic diagram of the water plant control method is shown below. Figure 11 As shown.

[0195] After obtaining basic data 70, which includes historical water supply data, meteorological data, and water plant operating data, the control device inputs the data into the data analysis and data processing module 71 to clean the basic data. In one embodiment, abnormal data and missing data are selected from the basic data. For these two types of data, in one case, interpolation calculations are performed using relevant data to replace abnormal data or supplement missing data; in another case, these two types of data are deleted.

[0196] The data is input into the primary optimization scheduling algorithm unit 721 within the algorithm unit 72. Multiple models within this unit process the cleaned basic data to obtain the planned pressure and planned water output. This planned pressure and planned water output are then input into the expert database 722 for querying, yielding a secondary optimization scheduling scheme: the operating conditions of multiple pump groups and the energy consumption required to achieve these conditions. By analyzing multiple scheduling schemes, the optimal scheduling scheme is determined.

[0197] In addition, when an anomaly occurs in at least one area within the water company's jurisdiction, an alarm is triggered, and the pressure of each water plant is determined for the area with the anomaly. The system then matches the data with experts in the database to find a suitable dispatching plan and implement the corresponding dispatching.

[0198] Figure 12 This is a schematic diagram of a water plant control device 800 according to an embodiment of the present application. The water plant control device 800 includes an acquisition module 801 and a processing module 802, wherein...

[0199] The acquisition module 801 is used to acquire historical water supply data and scheduling requirements for the target area.

[0200] The processing module 802 is used to estimate the water supply data of at least one water plant supplying water to the target area in multiple sub-periods within a future time period based on historical water supply data; the water supply data includes water supply volume and water supply pressure, and the future time period includes multiple sub-periods, which do not overlap with each other.

[0201] The processing module 802 is also used to divide the future period into multiple target periods based on the water supply data of multiple sub-periods, and determine the water supply demand corresponding to each target period; each target period includes at least one sub-period.

[0202] The processing module 802 is also used to obtain various target scheduling strategies for each water plant from the expert database based on the water supply demand for multiple target time periods and the current operating conditions of the water plant.

[0203] The processing module 802 is also used to determine the corresponding scoring strategy according to the scheduling requirements, and to select the optimal scheduling strategy from multiple target scheduling strategies using the scoring strategy.

[0204] In one embodiment, the processing module 802 is specifically used for:

[0205] For each pipeline, a water consumption prediction model is used to process the hourly water consumption and environmental data of multiple pipeline network monitoring points associated with the pipeline in multiple historical periods, and to estimate the hourly water consumption of the pipeline in multiple sub-periods in the future period.

[0206] For each water plant, the sum of the hourly water consumption of at least one pipeline associated with the water plant in any sub-period is taken as the water supply of the water plant in that sub-period.

[0207] Water consumption prediction models include time-dimensional data prediction models;

[0208] Historical water supply data includes hourly water consumption and environmental data from multiple pipeline monitoring points within the target area over multiple historical periods; multiple pipelines are located within the target area, and each pipeline has at least one pipeline monitoring point.

[0209] In one embodiment, the processing module 802 is specifically used for:

[0210] The first LSTM algorithm is used to process the hourly water consumption and environmental data of multiple pipeline monitoring points associated with the pipeline in multiple historical time periods, and to obtain the first predicted hourly water consumption of the pipeline in multiple sub-time periods in the future.

[0211] The first Prophet algorithm is used to process the hourly water consumption and environmental data of multiple pipeline monitoring points associated with the pipeline in multiple historical time periods to obtain the second predicted hourly water consumption of the pipeline in multiple sub-time periods in the future.

[0212] The moving weighted algorithm is used to process the hourly water consumption and environmental data of multiple pipeline monitoring points associated with the pipeline in multiple historical periods to obtain the third predicted hourly water consumption of the pipeline in multiple sub-periods in the future period.

[0213] The water consumption at the first, second, and third forecast times is fused to obtain the hourly water consumption of the pipeline in multiple sub-time periods within the future period.

[0214] The water consumption prediction model includes the first LSTM algorithm, the first Prophet algorithm, and the moving weighted algorithm.

[0215] In one embodiment, the processing module 802 is specifically used for:

[0216] For each water plant, the hourly water consumption of the water plant in multiple sub-periods within the future time period is queried in the corresponding water plant effluent characteristic table to obtain the water supply pressure required by the water plant to provide water supply in each sub-period.

[0217] The water plant effluent characteristics table includes the correspondence between the water plant providing at least one water supply volume and the water supply pressure that provides that volume.

[0218] In one embodiment, the processing module 802 is specifically used for:

[0219] For each pipeline monitoring point, the pipeline monitoring point pressure prediction model is used to process the water supply pressure and environmental data of the pipeline monitoring point in multiple historical periods to obtain the water supply pressure of the pipeline monitoring point in multiple sub-periods in the future period.

[0220] For each pipeline, the pipeline pressure model is used to process the water supply pressure of at least one network monitoring point associated with the pipeline in multiple sub-periods in the future time period, so as to obtain the water supply pressure of the pipeline in multiple sub-periods in the future time period.

[0221] For each water plant, the average water supply pressure of the pipelines associated with the water plant in any sub-period is taken as the water supply pressure of the water plant in that sub-period.

[0222] The pipeline monitoring point pressure prediction model includes a time-dimensional prediction model;

[0223] The historical water supply data includes water pressure and environmental data from multiple pipeline monitoring points within the target area at multiple historical time periods; there are multiple pipelines within the target area, and each pipeline has at least one network monitoring point.

[0224] In one embodiment, the processing module 802 is specifically used for:

[0225] The second LSTM algorithm is used to process the water supply pressure and environmental data of the pipeline monitoring points in multiple historical periods to obtain the first predicted water supply pressure of the pipeline monitoring points in multiple sub-periods in the future period.

[0226] The second Prophet algorithm is used to process the water supply pressure and environmental data of the pipeline monitoring points in multiple historical periods to obtain the second predicted water supply pressure of the pipeline monitoring points in multiple sub-periods in the future period;

[0227] The weighted average algorithm is used to process the water supply pressure and environmental data of pipeline monitoring points in multiple historical periods to obtain the third predicted water supply pressure of pipeline monitoring points in multiple sub-periods in the future period.

[0228] Data fusion of the first, second, and third predicted water supply pressures is performed to obtain the water supply pressure of pipeline monitoring points in multiple sub-periods within a future period.

[0229] The pressure prediction model for pipeline monitoring points includes the second LSTM algorithm, the second Prophet algorithm, and the weighted average algorithm.

[0230] In one embodiment, the processing module 802 is specifically used for:

[0231] The water supply pressure of each pipeline in each sub-period is processed using a pipeline flow-pressure model to obtain the water supply volume of the pipeline in each sub-period.

[0232] For each water plant, the sum of the water supply of at least one pipeline associated with the water plant in any sub-period is taken as the water supply of the water plant in that sub-period.

[0233] The pipeline flow-pressure model is a model that fits the historical water supply pressure and corresponding water supply volume of each pipeline.

[0234] In one embodiment, the processing module 802 is specifically used for:

[0235] For each water plant, the water supply pressure of the water plant in multiple sub-periods within the future time period is queried in the corresponding water plant outlet water characteristic table to obtain the water supply volume provided by the water plant using the water supply pressure in each sub-period.

[0236] The water plant effluent characteristics table includes at least one water supply pressure and the corresponding relationship between the water supply volume provided by the water plant under each water supply pressure.

[0237] In one embodiment, the processing module 802 is specifically used for:

[0238] The dynamic least squares method is used to process water supply data for multiple sub-periods, dividing the future period into multiple target periods;

[0239] For each target time period, the sum of the water supply volume of each sub-time period within the target time period is determined as the water supply volume of the target time period, and the average water supply pressure of each sub-time period within the target time period is determined as the water supply pressure of the target time period; the water supply demand of the target time period includes the water supply volume and the water supply pressure of the target time period.

[0240] In one embodiment, the processing module 802 is specifically used for:

[0241] The water supply range is determined based on the water supply volume for the target period and the preset fluctuation error of the water supply volume.

[0242] The available water supply, water pressure, and status of multiple water pumps are input into the expert database to obtain various target scheduling strategies. Each target scheduling strategy includes the target status of multiple water pumps in the water plant. The current operating condition of the water plant includes the status of multiple water pumps in the water plant. The status of the water pumps includes running status and stopped status.

[0243] The water supply pressure generated by the water plant using the target scheduling strategy is the water supply pressure for the target time period, and the available water supply is within the water supply range.

[0244] In one embodiment, the processing module 802 is specifically used for:

[0245] For each water plant, obtain the flow-head characteristic curve and flow-power characteristic curve of each water pump in the water plant; the water pumps include constant speed pumps or variable frequency pumps;

[0246] Based on the historical water supply pressure of the water plant, the water supply pressure range is determined; the maximum value of the water supply pressure range is the maximum value of the historical water supply pressure, and the minimum value of the water supply pressure range is the minimum value of the historical water supply pressure.

[0247] Within the constraints, and with each pump in any state, calculate the operating parameters of the water plant, construct the mapping relationship between the operating parameters and the states of multiple pumps in the water plant, and store the mapping relationship in the expert database.

[0248] The constraints include: the frequency of each variable frequency pump is within the high-efficiency operating frequency range, the flow deviation of each water pump is within the preset deviation range, and the water supply pressure of the water plant is within the water supply pressure range.

[0249] Any state includes whether each water pump is in operation or stopped. When the variable frequency pump is in operation, it is at any frequency within the high-efficiency operating frequency range.

[0250] The operating parameters of a water plant include water supply unit consumption, water distribution unit consumption, variable frequency pump pressure deviation, flow deviation, and total power.

[0251] In one embodiment, the processing module 802 is specifically used for:

[0252] Based on the water supply pressure at each pipeline monitoring point, obtain the water supply pressure of each target sub-area;

[0253] An alarm is triggered when the water supply pressure of any target sub-area is not within the normal pressure range, and the pressure to be adjusted for the target sub-area is obtained. If the pressure to be adjusted is within the normal pressure range, an alarm is triggered.

[0254] The corrected regional water supply pressure and the time information of the sampling of the water supply pressure are processed by the regional water plant pressure fitting model to obtain the water supply pressure to be adjusted for each water plant; the corrected regional water supply pressure includes the pressure to be adjusted for the target sub-region with pressure anomalies and the water supply pressure for other target sub-regions;

[0255] Input the water supply pressure to be adjusted from each water plant into the expert database to obtain the optimal scheduling strategy;

[0256] The target area can be divided into multiple target sub-areas, and each target sub-area includes multiple pipeline measurement points.

[0257] Figure 13 This is a schematic diagram of a control device according to an embodiment of the present application. The control device 900 includes a memory 901 and a processor 902. The memory 901 stores computer instructions executable by the processor. The memory 901 may include high-speed random access memory (RAM), and may also include non-volatile memory (NVM), such as at least one disk drive, or a USB flash drive, portable hard drive, read-only memory, magnetic disk, or optical disk, etc.

[0258] When executing computer instructions, processor 902 implements the various steps of the water plant control method with the control device as the execution entity in the above embodiments. For details, please refer to the relevant descriptions in the foregoing method embodiments. The processor 902 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly manifested as being executed by a hardware processor, or executed by a combination of hardware and software modules within the processor.

[0259] Optionally, the memory 901 can be either independent or integrated with the processor 902. When the memory 901 is configured independently, the control device 900 also includes a bus for connecting the memory 901 and the processor 902. This bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be categorized as an address bus, data bus, control bus, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0260] This application also provides a computer-readable storage medium storing computer instructions. When a processor executes the computer instructions, it implements the various steps of the water plant control method described above.

[0261] This application also provides a computer program product, including computer instructions, which, when executed by a processor, implement the various steps in the water plant control method described above.

[0262] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0263] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A water plant control method, characterized in that, The method includes: Obtain historical water supply data and scheduling requirements for the target area; Based on the historical water supply data, estimate the water supply data of at least one water plant supplying water to the target area in multiple sub-periods within a future period; the water supply data includes water supply volume and water supply pressure, and the future period includes the multiple sub-periods, each of which is independent of the others; Based on the water supply data of the multiple sub-periods, the future period is divided into multiple target periods, and the water supply demand corresponding to each target period is determined; each target period includes at least one sub-period. Specifically, the dynamic least squares method is used to process the water supply data of the multiple sub-periods, dividing the future period into the multiple target periods; the water supply volume is used as the independent variable and the water supply pressure is used as the dependent variable, and linear fitting is performed to obtain the fitting function. The predicted water supply pressure corresponding to each water supply volume in the fitting function is determined, and the absolute value of the difference between each predicted water supply pressure and the corresponding actual water supply pressure is calculated. When the absolute value of the difference is greater than a preset threshold, the period corresponding to the water supply volume corresponding to the difference is determined as the start period of a sub-period. For each water plant, based on the water supply demand for the multiple target time periods and the current operating conditions of the water plant, various target scheduling strategies for the water plant are obtained from the expert database. Based on the scheduling requirements, a corresponding scoring strategy is determined, and the optimal scheduling strategy is selected from the multiple target scheduling strategies using the scoring strategy. The expert database is constructed in the following way: For each water plant, obtain the flow-head characteristic curve and flow-power characteristic curve of each water pump in the water plant; the water pump includes a constant speed pump or a variable frequency pump; Based on the historical water supply pressure of the water plant, a water supply pressure range is determined; the maximum value of the water supply pressure range is the maximum value of the historical water supply pressure, and the minimum value of the water supply pressure range is the minimum value of the historical water supply pressure. Within the constraints, under any state of each water pump, calculate the operating parameters of the water plant, construct a mapping relationship between the operating parameters and the states of multiple water pumps in the water plant, and store the mapping relationship in the expert database; The constraints include: the frequency of each variable frequency pump is within the high-efficiency operating frequency range, the flow deviation of each water pump is within the preset deviation range, and the water supply pressure of the water plant is within the water supply pressure range. The arbitrary state includes each water pump being in an operating state or a stopped state. When the variable frequency pump is in an operating state, the variable frequency pump is at any frequency within the high-efficiency operating frequency range. The operating parameters of the water plant include the unit water consumption for supply, the unit water consumption for distribution, the pressure deviation of the variable frequency pump, the flow deviation, and the total power.

2. The method according to claim 1, characterized in that, The historical water supply data includes hourly water consumption and environmental data from multiple pipeline monitoring points within the target area over multiple historical periods; the target area is equipped with multiple pipelines, and each pipeline has at least one pipeline monitoring point; Based on the historical water supply data, estimate the water supply data for at least one water plant supplying water to the target area over multiple sub-periods in the future, including: For each of the pipelines, a water consumption prediction model is used to process the hourly water consumption and environmental data of multiple network monitoring points associated with the pipeline in multiple historical periods, and to estimate the hourly water consumption of the pipeline in multiple sub-periods in the future period. For each water plant, the sum of the hourly water consumption of at least one pipeline associated with the water plant in any sub-time period is taken as the water supply of the water plant in that sub-time period. The water consumption prediction model includes a time-dimensional data prediction model.

3. The method according to claim 2, characterized in that, The water consumption prediction model includes a first LSTM algorithm, a first Prophet algorithm, and a moving weighted algorithm; The water consumption prediction model is used to process the hourly water consumption and environmental data of multiple pipeline network monitoring points associated with the pipeline over multiple historical periods, and to estimate the hourly water consumption of the pipeline in multiple sub-periods within a future period, including: The first LSTM algorithm is used to process the hourly water consumption and environmental data of multiple pipeline monitoring points associated with the pipeline in multiple historical time periods to obtain the first predicted hourly water consumption of the pipeline in multiple sub-time periods in the future. The first Prophet algorithm is used to process the hourly water consumption and environmental data of multiple pipeline monitoring points associated with the pipeline in multiple historical time periods to obtain the second predicted hourly water consumption of the pipeline in multiple sub-time periods in the future. The moving weighted algorithm is used to process the hourly water consumption and environmental data of multiple pipeline monitoring points associated with the pipeline in multiple historical time periods to obtain the third predicted hourly water consumption of the pipeline in multiple sub-time periods in the future. The water consumption during the first, second, and third predicted periods is fused to obtain the hourly water consumption of the pipeline in multiple sub-periods within a future time period.

4. The method according to claim 2 or 3, characterized in that, The method further includes: For each water plant, the hourly water consumption of the water plant in multiple sub-periods within the future time period is queried from the water plant outlet water characteristic table corresponding to the water plant to obtain the water supply pressure required by the water plant to provide the water supply in each sub-period. The water plant effluent characteristic table includes the correspondence between the water plant providing at least one water supply volume and the water supply pressure providing that water supply volume.

5. The method according to claim 1, characterized in that, The historical water supply data includes water supply pressure and environmental data from multiple pipeline monitoring points within the target area at multiple historical time periods; the target area is equipped with multiple pipelines, and each pipeline has at least one network monitoring point; Based on the historical water supply data, estimate the water supply data for at least one water plant supplying water to the target area over multiple sub-periods in the future, including: For each pipeline monitoring point, the water supply pressure and environmental data of the pipeline monitoring point in multiple historical periods are processed using the pipeline monitoring point pressure prediction model to obtain the water supply pressure of the pipeline monitoring point in multiple sub-periods in the future period; For each pipeline, the water supply pressure of at least one network monitoring point associated with the pipeline is processed in multiple sub-periods in the future time period using a pipeline pressure model to obtain the water supply pressure of the pipeline in multiple sub-periods in the future time period. For each water plant, the average water supply pressure of the pipelines associated with the water plant in any sub-period is taken as the water supply pressure of the water plant in that sub-period. The pressure prediction model for pipeline monitoring points includes a time-dimensional prediction model.

6. The method according to claim 5, characterized in that, The pipeline monitoring point pressure prediction model includes the second LSTM algorithm, the second Prophet algorithm, and the weighted average algorithm; The water supply pressure and environmental data of the pipeline monitoring points in multiple historical time periods are processed using a pipeline monitoring point pressure prediction model to obtain the water supply pressure of the monitoring points in multiple sub-time periods in the future, including: The second LSTM algorithm is used to process the water supply pressure and environmental data of the pipeline monitoring point in multiple historical periods to obtain the first predicted water supply pressure of the pipeline monitoring point in multiple sub-periods in the future period; The second Prophet algorithm is used to process the water supply pressure and environmental data of the pipeline monitoring point in multiple historical periods to obtain the second predicted water supply pressure of the pipeline monitoring point in multiple sub-periods in the future period; The weighted average algorithm is used to process the water supply pressure and environmental data of the pipeline monitoring points in multiple historical periods to obtain the third predicted water supply pressure of the pipeline monitoring points in multiple sub-periods in the future period; The first predicted water supply pressure, the second predicted water supply pressure, and the third predicted water supply pressure are fused to obtain the water supply pressure of the pipeline monitoring point in multiple sub-periods within a future period.

7. The method according to claim 5 or 6, characterized in that, The method further includes: The water supply pressure of each pipeline in each sub-period is processed using a pipeline flow-pressure model to obtain the water supply volume of the pipeline in each sub-period. For each water plant, the sum of the water supply of at least one pipeline associated with the water plant in any sub-time period is taken as the water supply of the water plant in that sub-time period. The pipeline flow-pressure model is a model fitted using the historical water supply pressure and corresponding water supply volume of each pipeline.

8. The method according to claim 5 or 6, characterized in that, The method further includes: For each water plant, the water supply pressure of the water plant in multiple sub-periods within the future time period is queried in the water plant outlet water characteristic table corresponding to the water plant to obtain the water supply volume provided by the water plant using the water supply pressure in each sub-period. The water plant effluent characteristic table includes a correspondence between at least one water supply pressure and the amount of water supplied by the water plant under each of the stated water supply pressures.

9. The method according to claim 1, characterized in that, Based on the water supply data of the multiple sub-time periods, the future time period is divided into multiple target time periods, and the water supply demand corresponding to each target time period is determined, including: The water supply data for the multiple sub-periods are processed using the dynamic least squares method, and the future period is divided into multiple target periods. For each target time period, the sum of the water supply volume of each sub-time period within the target time period is determined as the water supply volume of the target time period, and the average water supply pressure of each sub-time period within the target time period is determined as the water supply pressure of the target time period; the water supply demand of the target time period includes the water supply volume and the water supply pressure of the target time period.

10. The method according to claim 9, characterized in that, The current operating status of the water plant includes the status of multiple water pumps in the water plant, and the status of the water pumps includes running status and stopped status; For each water plant, based on the water supply demand for multiple target time periods and the current operating status of the water plant, various target scheduling strategies for the water plant are obtained from an expert database, including: The water supply range is determined based on the water supply volume during the target period and the preset fluctuation error of the water supply volume. The available water supply, the water supply pressure, and the status of multiple water pumps are input into the expert database to obtain various target scheduling strategies; each target scheduling strategy includes the target status of multiple water pumps in the water plant. The water supply pressure generated by the water plant applying the target scheduling strategy is the water supply pressure for the target time period, and the available water supply is within the range of the water supply.

11. The method according to claim 2 or 5, characterized in that, The target area can be divided into multiple target sub-areas, and each target sub-area includes multiple pipeline measurement points; The method further includes: Based on the water supply pressure at each of the aforementioned pipeline monitoring points, obtain the water supply pressure of each target sub-area; When the water supply pressure of any target sub-area is not within the normal pressure range, an alarm is triggered, and the pressure to be adjusted in the target sub-area is obtained, wherein the pressure to be adjusted is within the normal pressure range. The corrected regional water supply pressure and the time information of the sampling of the water supply pressure are processed using a regional water plant pressure fitting model to obtain the water supply pressure to be adjusted for each water plant; the corrected regional water supply pressure includes the pressure to be adjusted for the target sub-region with pressure anomalies and the water supply pressure for other target sub-regions. Input the water supply pressure to be adjusted from each water plant into the expert database to obtain the optimal scheduling strategy.

12. A control device, characterized in that, include: A processor and a memory communicatively connected to the processor; The memory stores computer instructions; When executing the computer instructions, the processor is used to implement the water plant control method as described in any one of claims 1 to 11.

Images