Water volume scheduling method and system for cascade reservoir pump station group

Abstract

The invention relates to the technical field of water resource dispatching, and provides a water quantity dispatching method and system for a cascade reservoir pump station group. The method comprises the steps that cascade reservoir water supply system modeling is executed, a sensor network is activated, and a real-time state is established according to monitoring data and a modeling result; constructing a scheduling constraint by using a modeling result; acquiring a crop irrigation demand and establishing an irrigation priority; establishing a multi-dimensional objective function; according to the priority, the scheduling constraint, the real-time state and the target function, water volume scheduling adaptive optimization is carried out, and an optimization result is formed; and performing cascade reservoir pump station group water scheduling management according to an optimization result. The method solves the technical problems of low water resource distribution efficiency and high water shortage risk caused by unbalanced water resource supply and demand and complex scheduling constraint in the water scheduling process of the cascade reservoir pump station group, and achieves the purposes of realizing the accuracy and flexibility of water scheduling by introducing a multi-dimensional objective function and real-time state monitoring, and improving the scheduling efficiency of the cascade reservoir pump station group. And the water loss is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of water resource scheduling, and specifically to a water scheduling method and system for a cascade reservoir pump station group. Background Art

[0002] A cascade reservoir and pumping station cluster refers to a series of reservoirs and pumping stations arranged in a specific sequence and jointly operated within the same water source system. These clusters are typically used for large-scale water resource scheduling and supply systems. Each reservoir and pumping station has distinct functions and roles, ensuring the effective scheduling and utilization of water resources in different regions and over different time periods through specific water flow paths and control measures. However, traditional scheduling methods face multiple challenges: First, the water network topology typically uses a fixed hierarchical model, which makes it difficult to dynamically reflect complex water supply relationships (such as direct and indirect water supply paths), resulting in inefficient water transfer path planning and redundant costs. Second, reservoir evaporation loss prediction often relies on historical means or single meteorological factor models, which cannot accurately capture the effects of non-stationary meteorological sequences and multi-factor coupling, resulting in a lack of foresight in scheduling strategies and exacerbated water resource waste. Third, multiple objectives in cascade systems, such as water shortage, energy consumption, and water shortage rate, are mutually constrained. Traditional single-objective optimization methods struggle to achieve global balance and often sacrifice a single objective, impacting overall scheduling efficiency. Furthermore, the spatiotemporal heterogeneity of irrigation demand (e.g., crop growth cycles and priorities) and fluctuations in equipment operating conditions (e.g., pump station failures and abnormal pipeline pressures) further complicate dynamic scheduling. Therefore, a technical solution is urgently needed that can intelligently allocate and optimize water resources based on real-time monitoring data to address water supply demands at different times and in a variety of complex scenarios, ensuring efficient use of water resources and minimizing water losses. Summary of the Invention

[0003] This application provides a water scheduling method and system for a cascade reservoir pump station group, aiming to solve the technical problems of low water resource allocation efficiency and high water shortage risk due to imbalance in water supply and demand and complex scheduling constraints during the water scheduling process of a cascade reservoir pump station group.

[0004] The first aspect disclosed in the present application provides a water scheduling method for a cascade reservoir pump station group, the method comprising: executing modeling of water supply system elements of the cascade reservoir, activating a sensor network deployed at the cascade reservoir, and establishing a real-time status of the cascade reservoir based on monitoring data and modeling results; establishing scheduling constraints using the modeling results, the scheduling constraints including reservoir capacity constraints, pump station group flow constraints, pipeline pressure flow constraints, and water flow path constraints; obtaining irrigation needs of crops, and establishing irrigation priorities based on the irrigation needs; establishing a multidimensional objective function, the multidimensional objective function including a water shortage minimization target, a water shortage rate minimization target, a water loss minimization target, an irrigation priority adaptation target, and an energy consumption minimization target; performing adaptive optimization of water scheduling based on the irrigation priority, the scheduling constraints, the real-time status, and the multidimensional objective function, and establishing an adaptive optimization result; and performing water scheduling management for the cascade reservoir pump station group using the adaptive optimization result.

[0005] Another aspect disclosed in the present application provides a water scheduling system for a cascade reservoir pump station group, the water scheduling system comprising: a real-time status monitoring module: executing modeling of water supply system elements of the cascade reservoir, activating a sensor network deployed in the cascade reservoir, and establishing a real-time status of the cascade reservoir based on monitoring data and modeling results; a constraint establishment module: utilizing the modeling results to establish scheduling constraints, the scheduling constraints including reservoir capacity constraints, pump station group flow constraints, pipeline pressure flow constraints, and water flow path constraints; a priority establishment module: obtaining the irrigation needs of crops, and establishing irrigation priorities based on the irrigation needs; a function establishment module: establishing a multidimensional objective function, the multidimensional objective function including a water shortage minimization objective, a water shortage rate minimization objective, a water loss minimization objective, an irrigation priority adaptation objective, and an energy consumption minimization objective; an adaptive optimization module: performing adaptive optimization of water scheduling based on the irrigation priority, the scheduling constraints, the real-time status, and the multidimensional objective function, and establishing an adaptive optimization result; a scheduling management module: performing water scheduling management of the cascade reservoir pump station group based on the adaptive optimization result.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The water scheduling method for a cascade reservoir pump station cluster described above first models the cascade reservoir's water supply system and uses a sensor network to collect real-time monitoring data to establish the reservoir's real-time status. Based on this data, scheduling constraints are then set, including restrictions on reservoir capacity, pump station cluster flow, pipeline pressure flow, and water flow paths. Next, irrigation priorities are determined based on crop irrigation needs, and a comprehensive optimization is performed using a multidimensional objective function that encompasses objectives such as minimizing water shortages, minimizing water shortage rates, minimizing water losses, adapting irrigation priorities, and minimizing energy consumption. Finally, water scheduling is optimized based on these priorities, constraints, and objective functions to obtain the optimal scheduling result, thereby managing water allocation within the reservoir pump station cluster and ensuring efficient use of water resources and a stable water supply.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0010] Figure 1 Schematic diagram of a flow chart of a water dispatching method for a cascade reservoir pump station group in one embodiment.

[0011] Figure 2 This is an architecture diagram of a water dispatching system for a cascade reservoir pump station group in one embodiment.

[0012] Explanation of the reference numerals: real-time status monitoring module 11 , constraint establishment module 12 , priority establishment module 13 , function establishment module 14 , adaptation and optimization module 15 , scheduling management module 16 . DETAILED DESCRIPTION

[0013] The embodiments of the present application provide a water scheduling method and system for a cascade reservoir pump station group to solve the technical problems of low water resource allocation efficiency and high water shortage risk caused by imbalance in water resource supply and demand and complex scheduling constraints during the water scheduling process of a cascade reservoir pump station group.

[0014] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0015] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0016] Example 1, as Figure 1 As shown, the present application provides a water dispatching method for a cascade reservoir pump station group, the method comprising:

[0017] Perform modeling of the water supply system elements of the cascade reservoirs, activate the sensor network deployed in the cascade reservoirs, and establish the real-time status of the cascade reservoirs based on monitoring data and modeling results.

[0018] In the embodiment of the present application, the water supply system of the cascade reservoir is first comprehensively modeled, involving various infrastructure facilities, functional modules of the reservoir and the connection relationship with other water supply facilities. Specifically, the main components of the water supply system are defined in the computer simulation software, including hydraulic facilities such as multiple reservoirs, pumping stations, water pipelines, regulating tanks and valves. The boundaries, functions and mutual relationships of each component are also clearly defined. For example, the pumping station is responsible for transporting water from one reservoir to a downstream reservoir or directly supplying water. The reservoirs are connected by pipelines. Then, the model of each reservoir is designed, and its basic parameters such as storage capacity, water level range and effective water depth are set to construct a relationship model between water level and storage capacity. In addition, the modeling of the pump station is also followed closely, and the rated flow, maximum flow and working pressure range of the pump station are set, and the flow-pressure curve of the pump station is designed to reflect the relationship between flow and pressure under different working conditions. In pipeline modeling, the basic parameters of the pipeline are defined, including the diameter, length, material, and friction coefficient of the pipeline. A dynamic model of the pipeline water flow is established, and the water transfer rules between reservoirs are defined. The start and stop conditions of the pump station are set to ensure the smoothness and effectiveness of water transfer. After the construction is completed, the simulation model will be preliminarily verified using historical data, and optimized and adjusted based on the verification results to ensure that the simulation model can operate stably under various conditions for subsequent optimized scheduling. At the same time, the sensor network deployed in the cascade reservoirs is activated to collect reservoir operating data in real time, including key parameters such as water level, flow, and pressure. The real-time monitoring data is then imported into the simulation model, thereby establishing the operating status of the reservoir in real time, accurately reflecting the current situation of each link, and ensuring that subsequent scheduling can be efficiently managed and decided based on real-time information.

[0019] The modeling results are used to establish scheduling constraints, which include reservoir capacity constraints, pump station group flow constraints, pipeline pressure flow constraints, and water flow path constraints.

[0020] In one embodiment, a series of scheduling constraints are set based on the aforementioned reservoir water supply simulation model to ensure the rational allocation of water resources and the stable operation of the system. These constraints include reservoir capacity constraints, pump station group flow constraints, pipeline pressure flow constraints, and water flow path constraints. Reservoir capacity constraints are maximum and minimum water level constraints set based on the boundary information of the reservoir model in the simulation model. This ensures that the reservoir's water storage capacity is not overloaded or over-filled, thereby avoiding overflows or inability to meet water demand. Pump station group flow constraints refer to the maximum flow rate that each pump station can provide per unit time during water scheduling. This is determined based on the flow limits of individual pump stations in the pump station group, avoiding overload operation and ensuring the continuity and stability of water supply. Pipeline pressure flow constraints are used to ensure that the flow rate and pressure of water in the pipeline remain within the range specified by the pipeline design. They are typically determined based on the boundary information of the pipeline model in the simulation model to prevent pipeline leakage or rupture due to excessive pressure. Water flow path constraints refer to the restrictions on the path of water flow in the water supply system. These water flow path constraints can be set according to the water flow diagram to ensure that water flows through the most appropriate path, thereby improving water delivery efficiency and reducing water loss.

[0021] Furthermore, the water flow path constraint is constructed as follows:

[0022] A water flow direction diagram is generated based on the modeling results, wherein the nodes in the water flow direction diagram represent reservoirs, pumping stations, and pipelines, and the edges in the water flow direction diagram represent water flow paths; a fitting analysis of the path feasibility is performed based on the water flow direction diagram, and water flow path constraints are established based on the fitting analysis results.

[0023] Preferably, first, the connection relationship between water conservancy facilities such as reservoirs, pumping stations, and pipelines is extracted from the simulation model in the modeling results to understand which facilities have water flow transmission and which facilities are the starting point or end point of the water flow. Subsequently, each reservoir, pumping station, and pipeline is regarded as a node, wherein the reservoir node represents the water source or the location of the water storage volume, the pumping station node represents the water lifting and transmission station, and the pipeline node represents the water flow transmission path. On this basis, according to the connection relationship of the water conservancy facilities, the edges between the nodes are constructed to represent the water flow path from the reservoir to the pumping station and from the pumping station to the pipeline. At the same time, the attributes of the edges are marked on the edges, such as the direction, flow rate, length and other parameters of the water flow, so as to construct a water flow direction diagram, showing the flow path of water resources and the connection relationship between the facilities. Next, the operating conditions and characteristics of the nodes involved in each path in the flow diagram are obtained, such as the maximum flow rate, minimum flow rate, and pressure limit of each pumping station. The actual carrying capacity of the pipeline is calculated using hydraulic dynamics formulas (such as the Darcy-Weisbach formula) to represent the flow limit. A complexity threshold is then set based on the complexity of the flow path. This factor includes factors such as the number of pipelines, pumping station control points, and valve settings, all of which affect the stability and control difficulty of the flow. The higher the path complexity, the stricter the flow path constraints, requiring more sophisticated scheduling and monitoring. After setting the complexity threshold, a flow path that meets this threshold is selected to ensure stable flow through multiple control points and changing environmental conditions. Then, a fitting analysis is performed combining the node conditions, flow limit, and complexity threshold to derive the constraints for each flow path. These constraints take into account the flow limit, pipeline carrying capacity, and the dynamic characteristics of the flow, setting limits such as maximum flow and minimum pressure for each path to ensure that the flow is transmitted within the specified range. Finally, these path restrictions are summarized to obtain the water flow path constraint. This water flow path constraint includes flow restriction, pressure restriction, and path complexity restriction, etc. It can ensure that the actual flow of water follows a reasonable route and avoid unreasonable water flow paths or situations that exceed the system carrying capacity during the scheduling process, thereby ensuring the stability and efficiency of the water supply system.

[0024] Furthermore, the flow constraint of the pump station group is constructed as follows:

[0025] Obtain the flow limit of a single pumping station in a pumping station group; perform a pumping station linkage analysis on the pumping station group, and establish a total flow limit for the pumping station group based on the pumping station linkage analysis results; perform equipment status prediction within the pumping station group, and generate predicted abnormality limits based on the equipment status prediction results; establish a pumping station group flow constraint based on the single pumping station flow limit, the total flow limit for the pumping station group, and the predicted abnormality limits.

[0026] Preferably, first, the flow limits of individual pumping stations in the pumping station group are obtained from the simulation model, that is, the maximum flow rate of each pumping station is obtained to ensure that each pumping station does not exceed its designed flow capacity during operation. This is to prevent failures or efficiency reduction caused by overload operation of the pumping station. Subsequently, a pumping station linkage analysis is performed, which is a comprehensive analysis of the operation of multiple pumping stations within the pumping station group. In this process, the connection relationship between pipelines is determined based on the connection relationship of the pumping stations, that is, whether multiple pumping stations are supplying water to the same pipeline or whether the pumping stations are pressurizing and supplying water step by step. By combining the connection relationship with the maximum carrying capacity of the pipeline, flow constraints are set under different states. For parallel connections, the sum of the flow rates output by multiple pumping stations cannot exceed the maximum carrying capacity of the pipeline. For series connections, the flow rate output by a single pumping station cannot exceed the flow limit of a single pumping station. Afterwards, these flow constraints obtained through the pumping station linkage analysis are summarized to establish the total flow limit of the pumping station group, that is, the maximum flow rate that the entire pumping station group can provide within a certain period of time, to ensure that the total flow of the pumping station group is within a reasonable range and avoid overload operation. At the same time, the status of equipment within the pump station group will be predicted. During this process, the real-time monitoring data of the pump station group equipment (such as vibration, temperature, pressure, flow, etc.) will be input into the anomaly assessment model. This anomaly assessment model can be obtained by training the long short-term memory network. The data used for training are historical pump station group equipment status data and historical health status levels (such as normal, warning, alert, fault, etc.). The training is carried out iteratively through steps such as forward propagation, loss calculation (such as cross-entropy loss function), backpropagation, and parameter optimization (Adam). The anomaly assessment model will evaluate the health status of the current pump station group equipment based on the input real-time monitoring data, and combine the predicted health status level with the equipment derating factor to construct a predicted anomaly limit to avoid safety risks caused by equipment failure. Among them, the equipment derating factor is used to reduce the operating load of the pump station group equipment. It is generally 1, 0.9, 0.7, and 0.5, corresponding to normal, warning, alert, and fault, respectively. Finally, the flow limits of individual pumping stations, the total flow limits of the pumping station group, and the predicted abnormal limits are summarized and stored together in an empty set to establish flow constraints for the pumping station group. This ensures that the pumping station group does not exceed various capacity limits during operation, thereby ensuring the stability and efficiency of the water supply system.

[0027] Irrigation requirements of crops are obtained, and irrigation priorities are established based on the irrigation requirements.

[0028] In one embodiment, the type of crop requiring irrigation is first determined. Based on the crop type, the current crop stage is retrieved from the crop crop management log. The current crop stage, crop type, and current climate are then combined to retrieve historical irrigation data using Euclidean distance. Historical irrigation data with similarity distances less than or equal to a preset distance are retrieved, and the mean of these historical irrigation data is calculated to determine the crop's irrigation needs, i.e., the current amount of water required. Next, irrigation priorities are established based on these needs. Generally, crops with greater water requirements are given higher priorities, ensuring they receive adequate water supply, thereby improving water resource utilization efficiency and ensuring the stability of agricultural production.

[0029] A multidimensional objective function is established, wherein the multidimensional objective function includes a water shortage minimization objective, a water shortage rate minimization objective, a water loss minimization objective, an irrigation priority adaptation objective, and an energy consumption minimization objective.

[0030] In one embodiment, in order to ensure the efficient use of water resources, a multidimensional objective function is established. This multidimensional objective function includes five parts: water shortage minimization objective, water shortage rate minimization objective, water loss minimization objective, irrigation priority adaptation objective and energy consumption minimization objective. Among them, the water shortage minimization objective aims to minimize the amount of water shortage that occurs and ensure that the water supply can meet the demand as much as possible; the water shortage rate minimization objective is similar to the water shortage minimization objective. The water shortage rate represents the proportion of water shortage to the overall demand. The purpose of this objective is to minimize the proportion of water shortage and ensure that the supply of various water demands is as balanced as possible; the water loss minimization objective aims to reduce evaporation loss, leakage loss and other unavoidable losses during the water transmission process to ensure that more water resources can effectively reach the water use end; the irrigation priority adaptation objective ensures that under limited water resources, irrigation needs can be met according to priority by optimizing the distribution of water, and ensures that important crops (crops in the critical growth period and crops with large water requirements) can get sufficient water supply first; the energy consumption minimization objective aims to reduce the energy consumed in water lifting, transportation and distribution. The multidimensional objective function is constructed by weighting these minimization objectives, as follows

[0031] Among them, Z is the overall optimization performance; w1, w2, w3, w4, w5 are the weight coefficients of each goal, reflecting the importance of different goals; N is the total number of water points; D i is the demand for water use at the i-th point; S i is the actual supply of the i-th water point; M is the number of water transmission paths; L j is the water loss of the j-th water transmission path; p iis the irrigation priority of the i-th water point, P is the total number of pumping stations; Q k is the water delivery flow of the kth pumping station; H k is the head of the k-th pump station; η k is the comprehensive efficiency of the kth pumping station; t k is the operating time of the kth pump station. By establishing this multidimensional objective function, we can minimize water waste, reduce costs, and improve water supply efficiency while ensuring stable water supply, while ensuring that the priorities of different needs are effectively met.

[0032] Adaptive optimization of water scheduling is performed according to the irrigation priority, the scheduling constraints, the real-time status, and the multi-dimensional objective function, and an adaptive optimization result is established.

[0033] In one embodiment, after obtaining irrigation priorities, scheduling constraints, real-time status, and a multidimensional objective function, historical irrigation data is matched based on the irrigation priorities and real-time status to obtain historical irrigation data that meets both the irrigation priority requirements and the real-time status of the current cascade reservoirs. Subsequently, the matched historical irrigation data is secondary screened using the scheduling constraints to eliminate historical data that does not meet the scheduling constraints. The remaining historical irrigation data is then used as the initial population to solve the multidimensional objective function to find the optimal water scheduling solution to ensure optimal water resource allocation. Finally, the optimized water scheduling solution is added to the adaptive optimization results and used for subsequent water scheduling management of the cascade reservoir pump station group to ensure maximum water resource utilization.

[0034] Furthermore, performing adaptive optimization of water scheduling according to the irrigation priority, the scheduling constraint, the real-time status, and the multi-dimensional objective function, and establishing an adaptive optimization result, includes:

[0035] Calling a historical irrigation database; extracting demand features of the irrigation demand and using the demand features as first matching features; extracting state features of the real-time state and using the state features as second matching features; performing irrigation matching on the historical irrigation database using the first matching features and the second matching features to establish an irrigation matching result; establishing an initial population based on the irrigation matching result and the scheduling constraints to perform adaptive optimization.

[0036] Preferably, first, a historical irrigation database is accessed. This historical irrigation database contains irrigation data from different periods of time, different crop types, and different environmental conditions. By accessing this historical data, detailed information about past irrigation demand and actual water supply can be obtained, providing a reference for subsequent optimization decisions. Subsequently, based on the irrigation priority, demand characteristics of the irrigation demand are extracted and used as the first matching feature for subsequent irrigation demand analysis and matching. These characteristics include crop type, production stage, irrigation water volume, climatic conditions, etc., helping to identify the water demand patterns of different crops. Simultaneously, real-time state characteristics are extracted. These characteristics include reservoir water level, flow rate, pump station operating status, pipeline pressure, etc. This real-time data reflects the current operating status of the water supply system. These real-time state characteristics are used as the second matching feature to compare and match the actual water supply status with the historical data. Next, irrigation matching is performed against the historical irrigation database using the first and second matching features. The similarity distance between the current irrigation demand characteristics and each data item in the historical irrigation database is calculated using Euclidean distance. Historical irrigation data with a similarity distance less than or equal to a preset distance are found and used as the irrigation matching result. Then, based on these irrigation matching results, the system screens the data against scheduling constraints (such as reservoir capacity and pumping station flow limits). Historical irrigation data that doesn't meet the current scheduling constraints is removed. An initial population is then established based on the remaining historical irrigation data from the irrigation matching results. This initial population represents different possible scheduling scenarios, each of which includes a possible water allocation strategy, serving as the starting point for subsequent optimization. Finally, by optimizing the initial population and performing adaptive optimization, the scheduling plan is further adjusted and refined to ensure that water resources achieve optimal water allocation while meeting irrigation needs and constraints.

[0037] Furthermore, establishing an initial population according to the irrigation matching result and the scheduling constraint includes:

[0038] Call the historical element modeling mapped by the irrigation matching result; perform difference analysis based on the modeling result and the historical element modeling to establish difference characteristics; use the difference characteristics to establish iterative attenuation direction constraints for search optimization, and use the iterative attenuation direction constraints to perform adaptive optimization management.

[0039] Optionally, historical factor modeling based on the irrigation matching results is first invoked. This historical factor modeling is obtained by importing historical irrigation data into the aforementioned simulation model, providing data support for subsequent optimization decisions. Subsequently, a discrepancy analysis is performed based on the historical factor modeling and modeling results. By comparing the parameter differences between the historical simulation model and the simulation model, the factors causing the deviation between demand and historical data are analyzed. These differences may be due to factors such as climate change, differences in crop growth conditions, and changes in soil moisture. Based on these differences, differential features are extracted, identifying key factors affecting water demand and supply. Subsequently, the differential features are used to establish iterative attenuation direction constraints for search optimization. This means that during the optimization process, an attenuation direction is determined based on the size and trend of the differential features. This guides the optimization algorithm to gradually reduce unnecessary search areas and focus on areas where efficiency can be improved. The optimization algorithm can be a particle swarm optimization algorithm or a genetic algorithm. When using a genetic algorithm, the optimization process includes steps such as multidimensional objective function evaluation, population selection, population crossover, and population mutation. Finally, adaptive optimization management is performed using iterative attenuation direction constraints and multidimensional objective functions. During the optimization process, the scheduling plan is adjusted through repeated iterations, and the search process is guided according to the attenuation constraints, so that each iteration can more accurately approach the global optimal solution. This process effectively manages the optimization of water scheduling, ensures the rational allocation of resources, and avoids unnecessary energy consumption and water loss.

[0040] Furthermore, the adaptive optimization management using the iterative attenuation direction constraint includes:

[0041] In the process of performing iterative optimization, the iterative attenuation direction constraint is called according to the current number of iterations to establish a first update reference; the multidimensional objective function is used to perform population adaptation analysis based on real-time status, and a second update reference is established according to the global optimal adaptation fitness, and a third update reference is established according to the historical optimal adaptation fitness of the same family; the first update reference, the second update reference, the third update reference, and the random perturbation factor are used to perform iterative search under the scheduling constraint.

[0042] Optionally, during the iterative optimization process, the previously set iterative attenuation direction constraint is first called according to the current number of iterations. As the number of iterations increases, the effect of the attenuation direction constraint gradually increases, helping to gradually narrow the search space, avoid the search process being too broad and ineffective, and ensure that the optimization process is concentrated in the area where the optimal solution is most likely to be obtained. At this time, a first update reference is established, which reflects the latest optimization result obtained based on the attenuation direction constraint in the current iterative state. Subsequently, a population adaptation analysis is performed using a multidimensional objective function combined with real-time status (such as reservoir water level, flow, pump station operation status, etc.). In this process, the multidimensional objective function is evaluated for each solution in the optimization population, and the evaluation value is subtracted from 1 to obtain the adaptation fitness of each solution. Then, a solution with the best current adaptation fitness is selected as the second update reference. This second update reference takes into account the global optimal adaptation situation and represents the current evaluation result of the global optimal solution, helping to ensure that the search process can continue to approach the global optimal solution. In addition, it is necessary to generate a third update reference based on the historical optimal adaptation fitness of the same family, that is, taking into account the fitness of the previously best-performing solution in the population. This step can prevent the search process from falling into a local optimum, thereby ensuring that the optimization process is more diverse and can explore more possible optimal solutions. Finally, combining these three update references (the first update reference, the second update reference, and the third update reference) and a random perturbation factor, an iterative search is performed. The random perturbation factor introduces a certain degree of randomness, increases the diversity of solutions, and avoids the optimization process being too simple. By continuously adjusting the parameters of each solution, the iterative search will gradually optimize the water resource allocation plan based on the existing scheduling constraints, thereby finding the optimal water scheduling plan.

[0043] Further, the performing the iterative search under the scheduling constraint using the first update reference, the second update reference, the third update reference, and the random perturbation factor includes:

[0044] The iteration status of the same family iterations is recorded, and the taboo round triggering analysis is performed based on the iteration status record result; if the iteration status record result triggers a taboo round, the corresponding solution is added to the taboo list; and iterative search management is performed based on the taboo list.

[0045] Optionally, the iteration status of iterations within the same family (i.e., multiple individuals within the same optimization population) is first recorded. This means that the status and performance of the current solution are recorded at each iteration, including key information such as each solution's fitness value and parameter settings. This record facilitates subsequent analysis of trends and potential improvement directions during the optimization process. Subsequently, based on the iteration status record results, a taboo round triggering analysis is performed. This involves checking whether certain solutions exhibit repetitive patterns or results that are too close to each other during certain iterations. These solutions may have become trapped in local optima and cannot be further optimized. In such cases, a taboo round is triggered, meaning that these solutions are temporarily no longer allowed to be selected or participate in the optimization process. This prevents the algorithm from wandering within the same solution space and enhances optimization diversity. If the iteration status record results trigger a taboo round, the corresponding solution is added to a taboo list, which stores solutions that are temporarily no longer accepted or need to be excluded. These solutions will not be considered or selected for subsequent iterations, preventing the algorithm from being trapped in local optima. Finally, the taboo list is used to manage the iterative search. During the iteration process, the search strategy is adjusted based on the taboo list information to avoid selecting solutions in the taboo list, thereby guiding the search process towards a more diverse solution space. Taboo list management effectively avoids repeated searches and improves search efficiency by limiting the use of certain undesirable solutions, accelerating the process of finding the optimal solution.

[0046] Furthermore, performing iterative search management according to the taboo list includes:

[0047] Set the taboo selection similarity. When any solution is in the taboo list, use the taboo selection similarity to perform solution similarity matching in the iterative population. If the similarity matching result is a pass result, the corresponding solution is prohibited from being selected. If the period of any solution in the taboo list meets the taboo period, the unbanning operation of the corresponding solution is performed.

[0048] Optionally, first, a taboo selection similarity is set. This defines a similarity metric for solutions in the taboo list, used to determine the degree of similarity between the current solution and existing solutions in the taboo list. If the current solution is too similar to a solution in the taboo list, it may be in a local optimum, and repeated selection of it will not lead to new optimizations. Subsequently, similar matching of solutions in the iterative population is performed. This involves calculating the similarity between each solution in the current population and the solutions in the taboo list using methods such as Euclidean distance and cosine similarity. If the current solution has a high similarity to a solution in the taboo list, it is considered a taboo solution. In this case, the similarity selection mechanism prohibits the selection of the corresponding solution, preventing the optimization algorithm from selecting it again and maintaining search diversity. Furthermore, if a solution already exists in the taboo list and meets the taboo period conditions, it can be unbanned. In other words, the taboo state is not permanent, but has a periodic limit after which the solution can re-enter the search process. Through this management method, the taboo selection similarity and taboo period work together to ensure that the optimization algorithm will not fall into the local optimal solution. At the same time, it also increases the diversity of understanding, avoids excessive dependence on similar solutions, and helps the algorithm explore the solution space more comprehensively.

[0049] The adaptive optimization result is used to carry out water dispatching management of the cascade reservoir pump station group.

[0050] In one embodiment, after obtaining the adaptive optimization results, the operating parameters of each reservoir and pump station are obtained from the adaptive optimization results, including the water output of the reservoir, the working status of the pump station, the flow rate of the water pipeline, etc. This ensures that water resources can meet the needs of each region while minimizing water losses, avoiding water shortages, and achieving energy efficiency optimization. Subsequently, by implementing the scheduling plan in the optimization results, the operating status of the reservoir and pump station is monitored in real time to ensure that each link operates according to the set goals. Any deviations or abnormalities in operation can be automatically adjusted by the water scheduling system to ensure the stability and efficiency of the water supply system. Through such water scheduling management, it is possible to ensure the rational allocation of water resources under constraints and goals, and improve the efficiency and reliability of the water supply system of the cascade reservoir pump station group.

[0051] Furthermore, the water dispatching management of the cascade reservoir pump station group based on the adaptive optimization result also includes:

[0052] Perform scheduling monitoring on water quantity scheduling management to generate a real scheduling state; generate a consistent scheduling anomaly using the real scheduling state and the adaptive optimization result; and perform scheduling anomaly management based on the consistent scheduling anomaly.

[0053] Preferably, first, the water scheduling management is monitored, that is, through real-time monitoring of the sensor network, the operating status of the entire water supply system is continuously tracked, which includes monitoring the actual flow, water level, pressure and other key indicators of each reservoir, pump station and pipeline, so as to generate the real scheduling status, that is, the actual operation of the current water supply system. Subsequently, the real scheduling status is compared with the previously obtained adaptation optimization result to detect whether there is a consistent scheduling anomaly. The consistent scheduling anomaly refers to the existence of a large difference between the current actual scheduling and the optimized optimal scheduling result (exceeding the preset tolerance threshold), for example, the actual water distribution deviates from the predetermined target, the pump station does not operate as planned, etc. By comparing these deviations, potential problems or anomalies can be identified in a timely manner. Once a consistent scheduling anomaly is detected, scheduling anomaly management is immediately implemented. This involves executing pre-set response measures based on the location and anomaly parameters of the consistent scheduling anomaly. For example, while satisfying various constraints, the pump station speed or valve opening is dynamically adjusted to return actual water output to the optimized target, correcting any inconsistencies in water distribution. If a device fails, a backup device can be activated to supplement the water supply, ensuring that water resources are allocated according to the optimal plan and minimizing potential water waste or unstable water supply. This anomaly management mechanism responds to various changes in real time during the operation of the water scheduling system, ensuring the stability and efficiency of the water supply system.

[0054] In summary, the embodiments of the present application have at least the following technical effects:

[0055] The embodiment of the present application performs modeling of water supply system elements of cascade reservoirs, activates the sensor network deployed in the cascade reservoirs, and establishes the real-time status of the cascade reservoirs based on the monitoring data and modeling results; then, the modeling results are used to establish scheduling constraints, and the scheduling constraints include reservoir capacity constraints, pump station group flow constraints, pipeline pressure flow constraints, and water flow path constraints; thereafter, the irrigation needs of crops are obtained, and irrigation priorities are established based on the irrigation needs; further, a multidimensional objective function is established, and the multidimensional objective function includes a water shortage minimization target, a water shortage rate minimization target, a water loss minimization target, an irrigation priority adaptation target, and an energy consumption minimization target; then, based on the irrigation priority, the scheduling constraints, the real-time status, and the multidimensional objective function, adaptive optimization of water scheduling is performed to establish an adaptive optimization result; finally, the adaptive optimization result is used to perform water scheduling management of the cascade reservoir pump station group. These technical effects jointly solve the technical problems of low water resource allocation efficiency and high water shortage risk caused by imbalance in water supply and demand and complex scheduling constraints during the water scheduling of cascade reservoir pump station groups. They achieve the technical effect of achieving accuracy and flexibility in water scheduling and reducing water losses by introducing multi-dimensional objective functions and real-time status monitoring.

[0056] The second embodiment is based on the same inventive concept as the water dispatching method of the cascade reservoir pump station group in the above embodiment. Figure 2 As shown, the present application provides a water scheduling system for a cascade reservoir pump station group, the system comprising: a real-time status monitoring module 11: executing modeling of the water supply system elements of the cascade reservoir, activating the sensor network deployed in the cascade reservoir, and establishing the real-time status of the cascade reservoir according to the monitoring data and the modeling results; a constraint establishment module 12: establishing scheduling constraints using the modeling results, the scheduling constraints including reservoir capacity constraints, pump station group flow constraints, pipeline pressure flow constraints, and water flow path constraints; a priority establishment module 13: obtaining the irrigation needs of crops, and establishing irrigation priorities according to the irrigation needs; a function establishment module 14: establishing a multidimensional objective function, the multidimensional objective function including a water shortage minimization target, a water shortage rate minimization target, a water loss minimization target, an irrigation priority adaptation target, and an energy consumption minimization target; an adaptive optimization module 15: performing adaptive optimization of water scheduling according to the irrigation priority, the scheduling constraints, the real-time status, and the multidimensional objective function, and establishing an adaptive optimization result; a scheduling management module 16: performing water scheduling management of the cascade reservoir pump station group based on the adaptive optimization result.

[0057] Furthermore, the constraint establishing module 12 is further configured to execute the following method:

[0058] A water flow direction diagram is generated based on the modeling results, wherein the nodes in the water flow direction diagram represent reservoirs, pumping stations, and pipelines, and the edges in the water flow direction diagram represent water flow paths; a fitting analysis of the path feasibility is performed based on the water flow direction diagram, and water flow path constraints are established based on the fitting analysis results.

[0059] Furthermore, the constraint establishing module 12 is further configured to execute the following method:

[0060] Obtain the flow limit of a single pumping station in a pumping station group; perform a pumping station linkage analysis on the pumping station group, and establish a total flow limit for the pumping station group based on the pumping station linkage analysis results; perform equipment status prediction within the pumping station group, and generate predicted abnormality limits based on the equipment status prediction results; establish a pumping station group flow constraint based on the single pumping station flow limit, the total flow limit for the pumping station group, and the predicted abnormality limits.

[0061] Furthermore, the adaptation and optimization module 15 is further configured to execute the following method:

[0062] Calling a historical irrigation database; extracting demand features of the irrigation demand and using the demand features as first matching features; extracting state features of the real-time state and using the state features as second matching features; performing irrigation matching on the historical irrigation database using the first matching features and the second matching features to establish an irrigation matching result; establishing an initial population based on the irrigation matching result and the scheduling constraints to perform adaptive optimization.

[0063] Furthermore, the adaptation and optimization module 15 is further configured to execute the following method:

[0064] Call the historical element modeling mapped by the irrigation matching result; perform difference analysis based on the modeling result and the historical element modeling to establish difference characteristics; use the difference characteristics to establish iterative attenuation direction constraints for search optimization, and use the iterative attenuation direction constraints to perform adaptive optimization management.

[0065] Furthermore, the adaptation and optimization module 15 is further configured to execute the following method:

[0066] In the process of performing iterative optimization, the iterative attenuation direction constraint is called according to the current number of iterations to establish a first update reference; the multidimensional objective function is used to perform population adaptation analysis based on real-time status, and a second update reference is established according to the global optimal adaptation fitness, and a third update reference is established according to the historical optimal adaptation fitness of the same family; the first update reference, the second update reference, the third update reference, and the random perturbation factor are used to perform iterative search under the scheduling constraint.

[0067] Furthermore, the adaptation and optimization module 15 is further configured to execute the following method:

[0068] The iteration status of the same family iterations is recorded, and the taboo round triggering analysis is performed based on the iteration status record result; if the iteration status record result triggers a taboo round, the corresponding solution is added to the taboo list; and iterative search management is performed based on the taboo list.

[0069] Furthermore, the adaptation and optimization module 15 is further configured to execute the following method:

[0070] Set the taboo selection similarity. When any solution is in the taboo list, use the taboo selection similarity to perform solution similarity matching in the iterative population. If the similarity matching result is a pass result, the corresponding solution is prohibited from being selected. If the period of any solution in the taboo list meets the taboo period, the unbanning operation of the corresponding solution is performed.

[0071] Furthermore, the scheduling management module 16 is further configured to execute the following method:

[0072] Perform scheduling monitoring on water quantity scheduling management to generate a real scheduling state; generate a consistent scheduling anomaly using the real scheduling state and the adaptive optimization result; and perform scheduling anomaly management based on the consistent scheduling anomaly.

[0073] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0074] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0075] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A water dispatching method for a cascade reservoir pump station group, characterized in that: The method comprises: Modeling of water supply system elements of cascade reservoirs is performed, and sensor networks deployed in cascade reservoirs are activated to establish the real-time status of cascade reservoirs based on monitoring data and modeling results; Establishing scheduling constraints using the modeling results, wherein the scheduling constraints include reservoir capacity constraints, pump station group flow constraints, pipeline pressure flow constraints, and water flow path constraints; obtaining irrigation requirements of crops and establishing irrigation priorities based on the irrigation requirements; Establishing a multidimensional objective function, wherein the multidimensional objective function includes a water shortage minimization objective, a water shortage rate minimization objective, a water loss minimization objective, an irrigation priority adaptation objective, and an energy consumption minimization objective; Performing adaptive optimization of water scheduling according to the irrigation priority, the scheduling constraints, the real-time status, and the multidimensional objective function, and establishing an adaptive optimization result; The adaptive optimization result is used to carry out water dispatching management of the cascade reservoir pump station group.

2. The water dispatching method for a cascade reservoir pump station group according to claim 1, characterized in that: The adaptive optimization of water scheduling is performed according to the irrigation priority, the scheduling constraint, the real-time status, and the multi-dimensional objective function, and the adaptive optimization result is established, including: Call historical irrigation database; Extracting a demand feature of the irrigation demand, and using the demand feature as a first matching feature; Extracting a state feature of the real-time state, and using the state feature as a second matching feature; Performing irrigation matching in a historical irrigation database using the first matching feature and the second matching feature to establish an irrigation matching result; An initial population is established according to the irrigation matching result and the scheduling constraints to perform adaptive optimization.

3. The water dispatching method for a cascade reservoir pump station group according to claim 2, characterized in that: The establishing of the initial population according to the irrigation matching result and the scheduling constraint includes: Calling the historical element modeling of the irrigation matching result mapping; Perform difference analysis based on the modeling results and the historical element modeling to establish difference characteristics; The difference characteristics are used to establish iterative attenuation direction constraints for search optimization, and the iterative attenuation direction constraints are used to perform adaptive optimization management.

4. The water dispatching method for a cascade reservoir pump station group according to claim 3, characterized in that: The adaptive optimization management using the iterative attenuation direction constraint includes: During the iterative optimization process, the iterative attenuation direction constraint is called according to the current number of iterations to establish a first update reference; Performing population adaptation analysis based on real-time status using the multidimensional objective function, establishing a second update reference based on the global optimal adaptation fitness, and establishing a third update reference based on the historical optimal adaptation fitness of the same family; An iterative search under the scheduling constraint is performed using the first update reference, the second update reference, the third update reference, and a random perturbation factor.

5. The water dispatching method for a cascade reservoir pump station group according to claim 4, characterized in that: The performing the iterative search under the scheduling constraint by using the first update reference, the second update reference, the third update reference, and a random perturbation factor includes: Record the iteration status of the same family iterations and use the iteration status record results to perform taboo round trigger analysis; If the result of the iterative state record triggers a taboo round, the corresponding solution is added to the taboo list; Iterative search management is performed according to the taboo list.

6. The water dispatching method for a cascade reservoir pump station group according to claim 5, characterized in that: The iterative search management according to the taboo list includes: Setting a taboo selection similarity, and when any solution is in the taboo list, performing solution similarity matching in the iterative population using the taboo selection similarity; If the similarity matching result is a passing result, the corresponding solution is prohibited from being selected; If the period of any solution in the taboo list meets the taboo period, the unbanning operation of the corresponding solution is executed.

7. The water dispatching method for a cascade reservoir pump station group according to claim 1, characterized in that: The water flow path constraint is constructed as follows: Generate a water flow diagram based on the modeling results, wherein the nodes in the water flow diagram represent reservoirs, pumping stations, and pipelines, and the edges in the water flow diagram represent water flow paths; A fitting analysis of the path feasibility is performed based on the water flow direction diagram, and a water flow path constraint is established based on the fitting analysis results.

8. The water dispatching method for a cascade reservoir pump station group according to claim 1, characterized in that: The flow constraint of the pump station group is constructed as follows: Get the flow limit of a single pumping station in a pumping station group; Conduct pump station linkage analysis on the pump station group and establish total flow limit for the pump station group based on the pump station linkage analysis results; Perform equipment status prediction within the pump station group and generate prediction abnormality limits based on the equipment status prediction results; A pump station group flow constraint is established based on the single pump station flow limit, the pump station group total flow limit, and the predicted abnormality limit.

9. The water dispatching method for a cascade reservoir pump station group according to claim 1, characterized in that: The water dispatching management of the cascade reservoir pump station group based on the adaptive optimization result also includes: Conduct scheduling monitoring for water scheduling management and generate real scheduling status; Generate a consistent scheduling exception using the real scheduling state and the adaptive optimization result; Scheduling exception management is performed according to the consistency scheduling exception.

10. The water dispatching system of the cascade reservoir pump station group is characterized by: The water dispatching system is used to execute the water dispatching method for the cascade reservoir pump station group according to any one of claims 1 to 9, comprising: Real-time status monitoring module: Executes modeling of the water supply system elements of the cascade reservoirs, activates the sensor network deployed in the cascade reservoirs, and establishes the real-time status of the cascade reservoirs based on monitoring data and modeling results; Constraint establishment module: using the modeling results to establish scheduling constraints, the scheduling constraints include reservoir capacity constraints, pump station group flow constraints, pipeline pressure flow constraints, and water flow path constraints; Priority establishment module: obtaining the irrigation requirements of crops and establishing irrigation priorities according to the irrigation requirements; Function establishment module: establishing a multidimensional objective function, wherein the multidimensional objective function includes a water shortage minimization objective, a water shortage rate minimization objective, a water loss minimization objective, an irrigation priority adaptation objective, and an energy consumption minimization objective; Adaptive optimization module: performs adaptive optimization of water scheduling according to the irrigation priority, the scheduling constraints, the real-time status, and the multi-dimensional objective function, and establishes adaptive optimization results; Scheduling management module: uses the adaptive optimization results to perform water scheduling management of the cascade reservoir pump station group.

Images