Urban short-term water supply optimization scheduling method, equipment and medium based on water-energy balance

Abstract

The present application discloses a method, device and medium for optimizing and scheduling urban short-term water supply based on water-energy balance, which relates to the technical field of water supply scheduling optimization. The method comprises: using dynamic topology combination technology to construct a network diagram of the urban water supply system to be scheduled; extracting water use sequence samples as input features from the historical water use data of the urban water supply system to be scheduled; using a water consumption prediction model and a water consumption prediction error determination model to obtain the next day's water consumption forecast value; inputting the current day's precipitation forecast information into the reservoir inflow prediction model to obtain the next day's reservoir inflow forecast value, determining the next day's reservoir water storage change data, and completing the water use scheduling of the urban water supply system to be scheduled in combination with the scheduling model. By constructing a water consumption prediction model and a scheduling model, the present application can improve the efficiency and rationality of daily urban water supply scheduling.

Description

Technical Field

[0001] The present application relates to the technical field of water supply scheduling optimization, and in particular to a method, equipment and medium for optimizing urban short-term water supply scheduling based on water-energy balance. Background Art

[0002] With the acceleration of urbanization and the continuous growth of urban populations, the demand for water resources is also increasing. However, the supply of water resources is limited by natural conditions and infrastructure capacity, and the imbalance between supply and demand is becoming increasingly prominent. Ensuring the reliability and stability of urban water supply has become a key research topic, especially during droughts and peak water demand periods. At the same time, the energy consumption of water supply systems during the water transmission and distribution process cannot be ignored. Improving the energy efficiency of water supply systems is key to optimizing urban water supply management.

[0003] Traditional urban water supply systems rely primarily on experience and historical data for scheduling and management, lacking scientific forecasting and optimization methods, making them difficult to cope with complex and changing supply and demand situations. Furthermore, traditional methods are often inefficient when dealing with multi-objective optimization problems, failing to achieve optimal energy efficiency while ensuring water supply security. Summary of the Invention

[0004] The purpose of this application is to provide a method, equipment and medium for optimizing the scheduling of urban short-term water supply based on water-energy balance, which can improve the efficiency and rationality of daily urban water supply scheduling.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides a method for optimizing short-term urban water supply scheduling based on water-energy balance, comprising: utilizing dynamic topology combination technology to construct a network diagram of the urban water supply system to be scheduled. The network diagram of the urban water supply system to be scheduled is used to reflect the spatial supply and demand relationships of points, lines, and surfaces within the urban water supply system to be scheduled.

[0007] The water consumption sequence samples are extracted from the historical water consumption data of the urban water supply system to be scheduled as input features.

[0008] The input features are input into the water consumption prediction model to obtain an initial prediction value of the water consumption for the next day.

[0009] The input features are input into a water consumption prediction error determination model to obtain a prediction error. The water consumption prediction model and the water consumption prediction error determination model are obtained by jointly training a bidirectional long short-term memory network and an XGBoost model using historical water consumption data. The bidirectional long short-term memory network includes an attention mechanism.

[0010] The prediction error is used to correct the initial prediction value of the next day's water consumption to obtain the next day's water consumption prediction value.

[0011] Get the current daily precipitation forecast information for the city to be dispatched.

[0012] The current day's precipitation forecast information is input into a reservoir inflow prediction model to obtain a predicted value for the next day's reservoir inflow. The reservoir inflow prediction model is the Xin'an River hydrological model.

[0013] Based on the reservoir inflow for the next day and the reservoir water level data for the current period, the reservoir water storage change data for the next day is determined.

[0014] The water consumption scheduling of the urban water supply system to be scheduled is completed using a scheduling model based on the predicted value of water consumption for the next day, the reservoir water storage capacity change data for the next day and the network diagram of the urban water supply system to be scheduled.

[0015] Optionally, extracting water usage sequence samples from historical water usage data of the urban water supply system to be scheduled as input features includes: obtaining historical water usage data of the urban water supply system to be scheduled.

[0016] The historical water consumption data of the urban water supply system to be scheduled is preprocessed to obtain preprocessed water consumption data. The preprocessing includes accurate data processing and abnormal data processing.

[0017] The maximum information coefficient is used to select water sequence samples from historical water use data as input features.

[0018] Optionally, after extracting water usage sequence samples as input features from the historical water usage data of the urban water supply system to be scheduled, the method further includes: constructing a training set. The training set includes multiple data pairs. Each data pair includes the input features corresponding to a historical day and the water consumption on the day following the historical day.

[0019] Optionally, after constructing the training set, the method further includes: inputting multiple input features in the training set into a bidirectional long short-term memory network to obtain an output prediction value corresponding to each input feature.

[0020] The difference between the output prediction and the output corresponding to the same input feature is determined as the actual error.

[0021] Input the input features of the historical day into the XGBoost model to obtain the prediction error.

[0022] The loss function value is determined based on the actual error and the predicted error.

[0023] Based on the loss function value, adjust the parameters of the bidirectional long short-term memory network and the XGBoost model, and return to the step "input multiple input features into the bidirectional long short-term memory network to obtain the output prediction corresponding to each input feature" until the number of iterations reaches the preset number of iterations and the loss function value is less than the loss function value threshold.

[0024] The bidirectional long short-term memory network at the last iteration is determined to be the water consumption prediction model.

[0025] The XGBoost model at the last iteration is determined to be the water consumption prediction error determination model.

[0026] Optionally, based on the next-day water consumption forecast, the next-day reservoir water storage capacity change data, and the network diagram of the city water supply system to be scheduled, using a scheduling model to complete water consumption scheduling of the city water supply system to be scheduled includes: determining the product of an early warning factor and the next-day water consumption forecast as a reservoir capacity determination amount. The early warning factor is a positive number less than 1.

[0027] Determine whether the next day's water storage capacity is lower than the water storage capacity determination amount and obtain a determination result.

[0028] If the judgment result is yes, the emergency dispatch plan is activated.

[0029] If the judgment result is no, the scheduling model is solved using the mixed integer linear programming method based on the current reservoir inflow change data and the network diagram of the urban water supply system to be scheduled to obtain the water resource allocation optimization result, and the scheduling plan is determined based on the water resource allocation optimization result.

[0030] Control the urban water supply system to be scheduled to execute the scheduling plan.

[0031] Optionally, the scheduling model includes a first objective function, a second objective function, a project water diversion capacity constraint, a reservoir storage capacity constraint, and a water balance constraint.

[0032] Optionally, the first objective function is: .

[0033] in, is the first objective function, For the i Reservoir j Water Plant No. t Water supply during the period, For the j The water demand of a water plant in period t.

[0034] The second objective function is: .

[0035] .

[0036] .

[0037] in, is the second objective function, For the i Reservoirs j The energy consumption coefficient of water transmission and distribution of each water plant, For the i Reservoir No. t Time period j The water supply of each water plant, is the number of reservoirs, J is the number of water plants, T is the dispatching period, For the i Reservoirs j The water production energy consumption coefficient of a water plant.

[0038] The water diversion capacity constraints of the project are: .

[0039] in, For the i Reservoirs in Water supply during the period, For the i The maximum water supply capacity of a reservoir.

[0040] The reservoir capacity limits are: .

[0041] in, is the minimum storage capacity of the reservoir, For the i The storage capacity of a reservoir in time period t, The maximum storage capacity of the reservoir.

[0042] The water balance constraint is: .

[0043] in, For the i Reservoirs in t+ Storage capacity for 1 period, For the i Reservoirs t +1 period of inbound traffic, For the i Evaporation and seepage from a reservoir.

[0044] Optionally, the scheduling scheme determination formula is: .

[0045] in, is the actual total water supply of the reservoir, is the total water demand of the water plant, is the total water deficit.

[0046] In a second aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned urban short-term water supply optimization scheduling method based on water-energy balance.

[0047] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned urban short-term water supply optimization scheduling method based on water-energy balance.

[0048] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0049] This application provides a method, device, and medium for optimizing short-term urban water supply scheduling based on water-energy balance. Through precise data collection, feature screening, water demand forecasting, optimal water resource allocation, real-time scheduling, and an early warning and emergency response mechanism, this method achieves efficient management and stable operation of the urban water supply system. By combining a bidirectional long short-term memory network with an attention mechanism and the extreme gradient boosting (XGBoost) algorithm, and using a particle swarm optimization algorithm for hyperparameter optimization, the accuracy and stability of daily water demand forecasts are improved, ensuring that the water supply system can accurately respond to changes in urban water demand. Based on accurate water demand forecasts, a mixed integer linear programming (MILP) model is used to optimize water resource allocation and scheduling. This ensures rational and efficient water resource allocation within different time periods, maximizes resource utilization, minimizes water supply costs and energy consumption, and improves the operational efficiency of the urban water supply system. By monitoring the difference between actual water consumption and predicted values, an early warning threshold is set. When the difference exceeds the set range, an early warning mechanism is triggered, and an emergency scheduling plan is provided. This ensures the stable operation of the water supply system in emergency situations, effectively responds to sudden demand and abnormal conditions, and improves the reliability and safety of the urban water supply system. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0051] Figure 1 This is a flow chart of a method for optimizing and scheduling urban short-term water supply based on water-energy balance in one embodiment of the present application.

[0052] Figure 2This is a schematic diagram of a method for optimizing the short-term urban water supply scheduling based on water-energy balance in one embodiment of the present application.

[0053] Figure 3 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0054] The following will be combined with the 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 part of the embodiments of this application, not all of the embodiments. 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.

[0055] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0056] In an exemplary embodiment, Figure 1 and Figure 2 As shown in FIG, a method for optimizing the short-term urban water supply scheduling based on water-energy balance is provided, including:

[0057] Step 101: Using dynamic topology combination technology, construct a network diagram of the urban water supply system to be scheduled. The network diagram of the urban water supply system to be scheduled is used to reflect the spatial supply and demand relationship of points, lines and surfaces in the urban water supply system to be scheduled.

[0058] Step 102: Extract water usage sequence samples from the historical water usage data of the urban water supply system to be scheduled as input features.

[0059] Step 103: Input the input features into the water consumption prediction model to obtain an initial prediction value of the water consumption for the next day.

[0060] Step 104: Input the input features into a water consumption prediction error determination model to obtain a prediction error. The water consumption prediction model and the water consumption prediction error determination model are obtained by jointly training a bidirectional long short-term memory network and an XGBoost model using historical water consumption data. The bidirectional long short-term memory network includes an attention mechanism.

[0061] Step 105: Correct the initial predicted value of the next day's water consumption using the prediction error to obtain the next day's water consumption prediction value.

[0062] Step 106: Obtain the current daily precipitation forecast information for the city to be scheduled.

[0063] Step 107: Input the current day's precipitation forecast information into the reservoir inflow prediction model to obtain the next day's reservoir inflow prediction value. The reservoir inflow prediction model is the Xin'an River hydrological model.

[0064] The Xin'an River hydrological model is used to predict reservoir inflows. The Xin'an River model uses rainfall, evaporation, and other hydrological data to make predictions.

[0065]

[0066] is the predicted reservoir inflow for the next day, For the Xin'an River model, is the rainfall on day t, is the evaporation on day t, is the hydrological data such as initial soil moisture content.

[0067] Step 108: Based on the reservoir inflow for the next day and the reservoir water level data for the current period, determine the reservoir water storage capacity change data for the next day.

[0068] Step 109: Based on the next day's water consumption forecast value, the next day's reservoir water storage capacity change data and the network diagram of the city's water supply system to be scheduled, the scheduling model is used to complete the water consumption scheduling of the city's water supply system to be scheduled.

[0069] Prior to step 102, the method further includes obtaining historical water usage data for the city water supply system to be scheduled. The historical water usage data for the city water supply system to be scheduled is preprocessed to obtain preprocessed water usage data. The preprocessing includes both accurate data processing and abnormal data processing. A water usage sequence sample is selected from the historical water usage data using the maximum information coefficient as an input feature.

[0070] Collect historical water consumption data from the city's water supply system, primarily daily. Use the Maximum Information Coefficient (MIC) method to select highly relevant input features. Clean the collected data, address missing values ​​and outliers, and convert the data into a supervised learning format using a sliding window approach.

[0071] Construct a training set. The training set includes multiple data pairs. Any of the data pairs includes input features corresponding to the historical day, and water consumption the day after the corresponding historical day. Input multiple input features in the training set into the bidirectional long short-term memory network to obtain the output prediction amount corresponding to each input feature. Determine the difference between the output prediction amount and the output amount corresponding to the same input feature as the actual error. Input the input features of the historical day into the XGBoost model to obtain the prediction error. Determine the loss function value based on the actual error and the prediction error. Based on the loss function value, adjust the parameters of the bidirectional long short-term memory network and the XGBoost model, and return to the step "input multiple input features into the bidirectional long short-term memory network to obtain the output prediction amount corresponding to each input feature" until the number of iterations reaches the preset number of iterations and the loss function value is less than the loss function value threshold. Determine that the bidirectional long short-term memory network at the last iteration is the water consumption prediction model. Determine that the XGBoost model at the last iteration is the water consumption prediction error determination model.

[0072] A bidirectional long-short-term memory (Attention-BiLSTM) network with an attention mechanism was constructed to capture the bidirectional characteristics of water use data and the impact of important time points. A particle swarm optimization algorithm was used to optimize the model's hyperparameters and improve prediction accuracy.

[0073] The prediction formula is as follows:

[0074] .

[0075] The predicted water consumption for the next day. It is an attention bidirectional long short-term memory network model. is the current input feature, For the historical implicit state, is the optimized hyperparameter, is the residual.

[0076] The historical water use series samples (input features) are used as input, and the extreme gradient boosting (XGBoost) algorithm is used to perform residual correction to improve the prediction accuracy.

[0077]

[0078] Step 109 includes: determining the product of the early warning factor and the next day's water consumption forecast value as the reservoir capacity determination value. The early warning factor is a positive number less than 1.

[0079] Determine whether the next day's water storage capacity is lower than the water storage capacity determination amount and obtain a determination result.

[0080] If the judgment result is yes, the emergency dispatch plan is activated. When the emergency dispatch plan is activated, an emergency dispatch plan warning signal is issued and the backup water source is activated. When the emergency dispatch plan is activated, domestic water use has a higher priority than industrial water use and service water use.

[0081] When the emergency dispatch plan is activated: 1. Issue an early warning: To ensure the safety and stability of the city's water supply system, the system will immediately issue an early warning when it detects potential risks to the water supply. 2. Prioritize domestic water use and moderately reduce industrial and service water use: During the early warning state, prioritize domestic water use for residents while maintaining overall water supply balance by reducing non-emergency industrial and service water demand. 3. Activate backup water sources to supplement water supply: Activate backup water sources, including groundwater sources, emergency reservoirs, and other alternative water sources, to fill water supply gaps and ensure the stable operation of the water supply system.

[0082] If the judgment result is no, the scheduling model is solved using the mixed integer linear programming method based on the current reservoir inflow change data and the network diagram of the urban water supply system to be scheduled to obtain the water resource allocation optimization result, and the scheduling plan is determined based on the water resource allocation optimization result.

[0083] Control the urban water supply system to be scheduled to execute the scheduling plan.

[0084] Based on the short-term water demand forecast and reservoir inflow forecast results, the mixed integer linear programming (MILP) method is used to allocate water resources to ensure the balance between water resource allocation and water supply energy consumption in different time periods.

[0085] Among them, the scheduling model includes the first objective function, the second objective function, the project water diversion capacity constraint, the reservoir storage capacity limit and the water balance constraint.

[0086] The first objective function is: .

[0087] in, is the first objective function, For the i Reservoir j Water Plant No. t Water supply during the period, For the j The water demand of a water plant in period t.

[0088] The second objective function is: .

[0089] .

[0090] .

[0091] in, is the second objective function, For the i Reservoirs j The energy consumption coefficient of water transmission and distribution of each water plant, For the i Reservoir No. t Time period j The water supply of each water plant, is the number of reservoirs, J is the number of water plants, T is the dispatching period, For the i Reservoirs j The water production energy consumption coefficient of a water plant.

[0092] The water diversion capacity constraints of the project are: .

[0093] in, For the i Reservoirs in Water supply during the period, For the i The maximum water supply capacity of a reservoir.

[0094] Reservoir capacity limits are: .

[0095] in, is the minimum storage capacity of the reservoir, For the i The storage capacity of a reservoir in time period t, The maximum storage capacity of the reservoir.

[0096] The water balance constraint is: .

[0097] in, For the i Reservoirs in t+ Storage capacity for 1 period, For the i The inflow of each reservoir during period t+1 is: For the i Evaporation and seepage from a reservoir.

[0098] The scheduling scheme determination formula is: .

[0099] in, is the actual total water supply of the reservoir, is the total water demand of the water plant, is the total water deficit.

[0100] Based on the optimized water resource allocation results, a water supply scheduling plan is developed. This plan takes into account real-time monitoring data of the water supply system, historical scheduling records, predicted water demand, and reservoir inflow forecasts. The water supply scheduling plan is adjusted in real time, using an adaptive dynamic scheduling algorithm to dynamically optimize the scheduling scheme based on changes in real-time monitoring data and forecast data. The current reservoir capacity is calculated based on the predicted inflow. When the predicted reservoir capacity falls below 80% of the water demand, an early warning mechanism is triggered. Emergency scheduling plans are provided to ensure stable operation of the water system.

[0101] This application uses advanced feature screening and data processing methods to accurately extract key factors influencing water demand during water demand forecasting. Feature engineering using historical water supply data improves the model's prediction accuracy and stability. By coupling the attention mechanism and a bidirectional long short-term memory (BiLSTM) network with the XGBoost algorithm, a high-precision daily-scale water demand forecasting model is constructed. The attention mechanism effectively captures important information in time series data, the BiLSTM network better handles the dependencies between time series, and the XGBoost algorithm further optimizes model performance through gradient boosting. A water resource allocation method is proposed that aims to minimize the difference between water supply and demand and minimize energy consumption in the water supply system. Solved using a mixed integer linear programming (MILP) model, it comprehensively considers multiple constraints, including water supply and demand balance, energy efficiency optimization, and evaporation and leakage losses, achieving efficient configuration of the urban water supply system and balanced water supply energy consumption. Regarding water resource scheduling, this solution optimizes reservoir water level profiles to ensure rational and safe reservoir operation under varying water inflow conditions. At the same time, an early warning emergency mechanism has been set up. When the predicted water inflow is significantly lower than the water demand, an early warning will be triggered and emergency measures will be taken in a timely manner to ensure the safety of the city's water supply.

[0102] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 3As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. The processor, memory and input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for optimizing the scheduling of urban short-term water supply based on water-energy balance is implemented.

[0103] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present application and does not constitute a limitation on the computer device to which the solution of the present application is applied. A specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps of the above-mentioned method embodiments when executing the computer program.

[0104] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0105] In an exemplary embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0106] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0107] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0108] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.

[0109] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0110] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for optimizing urban short-term water supply scheduling based on water-energy balance, characterized in that: include: Use dynamic topology combination technology to construct the network diagram of the urban water supply system to be dispatched; The network diagram of the urban water supply system to be scheduled is used to reflect the spatial supply and demand relationship of points, lines and surfaces within the urban water supply system to be scheduled; Extract water consumption sequence samples from the historical water consumption data of the urban water supply system to be scheduled as input features; Inputting the input features into a water consumption prediction model to obtain an initial prediction value of water consumption for the next day; Inputting the input features into a water consumption prediction error determination model to obtain a prediction error; the water consumption prediction model and the water consumption prediction error determination model are obtained by jointly training a bidirectional long short-term memory network and an XGBoost model using historical water consumption data; the bidirectional long short-term memory network includes an attention mechanism; Correcting the initial forecast value of the next day's water consumption using the forecast error to obtain the next day's water consumption forecast value; Obtain the current daily precipitation forecast information for the city to be dispatched; Inputting the current day's precipitation forecast information into a reservoir inflow prediction model to obtain a predicted value of the next day's reservoir inflow; the reservoir inflow prediction model is the Xin'an River hydrological model; Determine the next day's reservoir water storage change data based on the next day's reservoir inflow and the current period's reservoir water level data; Based on the predicted water consumption value for the next day, the reservoir water storage capacity change data for the next day, and the network diagram of the urban water supply system to be scheduled, the scheduling model is used to complete the water consumption scheduling of the urban water supply system to be scheduled; Based on the next day's water consumption forecast value, the next day's reservoir water storage capacity change data, and the network diagram of the city water supply system to be scheduled, a scheduling model is used to complete water scheduling of the city water supply system to be scheduled, including: Determine the product of the early warning factor and the next day's water consumption forecast value as the reservoir capacity determination amount; the early warning factor is a positive number less than 1; Determine whether the next day's water storage capacity is lower than the water storage capacity determination amount, and obtain a determination result; If the judgment result is yes, the emergency dispatch plan is activated; If the judgment result is no, then according to the current reservoir inflow change data and the network diagram of the urban water supply system to be scheduled, the mixed integer linear programming method is used to solve the scheduling model to obtain the water resource allocation optimization result, and the scheduling plan is determined based on the water resource allocation optimization result; Controlling the urban water supply system to be scheduled to execute the scheduling plan; The prediction formula for the initial predicted value of the next day's water consumption is: in, is the predicted water consumption for the next day, f Attention-Bilstm is the attention bidirectional long short-term memory network model, x t is the current input feature, h t is the historical implicit state, θ * is the optimized hyperparameter, ε t is the residual; The expression of the reservoir inflow flow for the next day is: in, is the predicted reservoir inflow for the next day, f XAJ For the Xinanjiang model, P t is the rainfall on day t, E t is the evaporation on day t, S0 is the initial soil moisture content; The method of extracting water consumption sequence samples from the historical water consumption data of the urban water supply system to be scheduled as input features includes: Obtain historical water consumption data of the urban water supply system to be dispatched; Preprocessing the historical water consumption data of the urban water supply system to be dispatched to obtain preprocessed water consumption data; the preprocessing includes accurate data processing and abnormal data processing; The maximum information coefficient is used to extract water consumption sequence samples from historical water consumption data as input features; The scheduling model includes a first objective function, a second objective function, a project water diversion capacity constraint, a reservoir storage capacity constraint and a water balance constraint.

2. The urban short-term water supply optimization scheduling method based on water-energy balance according to claim 1 is characterized in that: After extracting water consumption sequence samples from the historical water consumption data of the urban water supply system to be scheduled as input features, it also includes: Construct a training set; the training set includes multiple data pairs; any of the data pairs includes input features corresponding to a historical day and water consumption on the day following the historical day.

3. The urban short-term water supply optimization scheduling method based on water-energy balance according to claim 2 is characterized in that: After building the training set, also include: Input multiple input features in the training set into the bidirectional long short-term memory network to obtain the output prediction corresponding to each input feature; Determine the difference between the output prediction and the output corresponding to the same input feature as the actual error; Input the historical daily input features into the XGBoost model to obtain the prediction error; Determine the loss function value based on the actual error and the predicted error; Based on the loss function value, adjust the parameters of the bidirectional long short-term memory network and XGBoost model, and return to step "input multiple input features into the bidirectional long short-term memory network to obtain the output prediction corresponding to each input feature" until the number of iterations reaches the preset number of iterations and the loss function value is less than the loss function value threshold; Determine that the bidirectional long short-term memory network at the last iteration is the water consumption prediction model; The XGBoost model at the last iteration is determined to be the water consumption prediction error determination model.

4. The urban short-term water supply optimization scheduling method based on water-energy balance according to claim 1 is characterized in that: The first objective function is: Among them, f1 is the first objective function, Q ijt is the water supply from the i-th reservoir to the j-th water plant in the t-th period, D jt is the water demand of the jth water plant in the tth period; The second objective function is: From ijt =α ij ·Q ijt ; P ijt =β ij ·Q ijt ; Among them, f2 is the second objective function, α ij is the energy consumption coefficient of water distribution from the i-th reservoir to the j-th water plant, Q ijt is the water supply from the i-th reservoir to the j-th water plant in the t-th period, I is the number of reservoirs, J is the number of water plants, T is the scheduling period, β ij is the water production energy consumption coefficient of the i-th reservoir to the j-th water plant; The water diversion capacity constraints of the project are: Among them, Q i,t is the water supply of the i-th reservoir in period t, Q i,max is the maximum water supply capacity of the i-th reservoir; The reservoir capacity limits are: In min ≤V i,t ≤V max ; Among them, V min is the minimum storage capacity of the reservoir, V i,t is the storage capacity of the i-th reservoir in period t, V max is the maximum storage capacity of the reservoir; The water balance constraint is: Among them, V i,t+1 is the storage capacity of the i-th reservoir in the t+1 period, is the inflow of the i-th reservoir in the t+1 period, L i,t is the evaporation and leakage of the i-th reservoir.

5. The urban short-term water supply optimization scheduling method based on water-energy balance according to claim 4 is characterized in that: The scheduling scheme determination formula is: Q T =D T -ΔQ T ; Among them, Q T is the actual total water supply of the reservoir, D T is the total water demand of the water plant, ΔQ t is the total water deficit.

6. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the urban short-term water supply optimization scheduling method based on water-energy balance according to any one of claims 1 to 5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for optimizing and scheduling urban short-term water supply based on water-energy balance according to any one of claims 1 to 5 is implemented.

Images