Multi-source prediction scheduling method, device and equipment for salt tide and storage medium

Abstract

A multi-water-source prediction scheduling method, device and equipment for salt tide and a storage medium are disclosed, and belong to the technical field of water source scheduling. The method comprises the following steps: acquiring a first time range of a first region, the first time range being a time range in which salt tide intrusion in the first region is predicted to occur; acquiring a first water demand of the first region in the first time range; establishing a multi-water-source scheduling model, a plurality of objective functions and a plurality of constraint conditions of the multi-water-source scheduling model being determined based on the first time range and the first water demand; based on the plurality of objective functions and the plurality of constraint conditions, a multi-objective genetic algorithm is used to solve the multi-water-source scheduling model to determine a multi-water-source scheduling scheme, the multi-water-source scheduling scheme comprising prediction scheduling of the first region in a second time range, the second time range being used to indicate the length of time of the prediction scheduling, and the second time range being before the first time range. The method can accurately realize prediction scheduling for salt tide.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of water source scheduling, and in particular to a multi-water source prediction scheduling method and device for salt tide, equipment and storage medium. BACKGROUND

[0002] Water source prediction scheduling refers to the behavior of scheduling water in a water source in another region to a region in advance when it is predicted that a certain natural disaster (such as drought, salt tide) will occur in the region. Multi-water source prediction scheduling is to schedule water in multiple water sources to the region.

[0003] In related technologies, the water source scheduling for salt tide is usually salt suppression scheduling. Salt suppression scheduling mainly refers to discharging water stored in a reservoir in a region when salt tide occurs in the region, so as to reduce the salinity of the water source in the region. However, when salt tide invasion occurs, the pumping station cannot take water and can only supply water to the city by the reservoir in the region, and salt suppression scheduling can easily lead to insufficient water storage in the reservoir in the region, and further lead to insufficient water supply in the city. SUMMARY

[0004] The present disclosure provides a multi-water source prediction scheduling method and device for salt tide, equipment and storage medium, which can accurately realize prediction scheduling for salt tide. The technical solution at least includes the following schemes:

[0005] In a first aspect, a multi-water source prediction scheduling method for salt tide is provided, comprising: obtaining a first time range of a first region, the first time range being a predicted time range of salt tide invasion in the first region; obtaining a first water demand of the first region in the first time range; establishing a multi-water source scheduling model, a plurality of objective functions and a plurality of constraint conditions of the multi-water source scheduling model being determined based on the first time range and the first water demand; based on the plurality of objective functions and the plurality of constraint conditions, solving the multi-water source scheduling model by using a multi-objective genetic algorithm to determine a multi-water source scheduling scheme, the multi-water source scheduling scheme comprising prediction scheduling for the first region in a second time range, the second time range being used to indicate the length of time of the prediction scheduling, and the second time range being before the first time range.

[0006] Optionally, the multiple objective functions of the multi-source scheduling model comprise a water supply guarantee rate objective function, a reservoir storage capacity objective function, a water supply damage depth objective function, a green economic benefit objective function, and a seawater desalination utilization objective function; wherein the water supply guarantee rate objective function is used to indicate that the total water supply guarantee rate of a to-be-scheduled region in the first time range and the second time range is maximum, the to-be-scheduled region comprising the first region and other regions supplying water to the first region; the reservoir storage capacity objective function is used to indicate that the storage capacity of a reservoir in the to-be-scheduled region in the first time range and the second time range is maximum; the water supply damage depth objective function is used to indicate that the total water supply damage depth of the to-be-scheduled region in the first time range and the second time range is minimum, the water supply damage depth being used to indicate the degree of water shortage of the to-be-scheduled region; the green economic benefit objective function is used to indicate that the energy consumption of a pump station in the to-be-scheduled region in the first time range and the second time range is minimum; and the seawater desalination utilization objective function is used to indicate that the seawater desalination utilization amount of the to-be-scheduled region in the first time range and the second time range is minimum.

[0007] Optionally, the constraints of the multi-water-source scheduling model comprise: a water balance constraint, a reservoir storage capacity constraint, a recharge flow constraint, a first water demand constraint, a pump station operating water level constraint, a pump station water intake capacity constraint, a pump station operation time constraint, a river water quality and reservoir water quality constraint, and a seawater desalination water supply capacity constraint; the water balance constraint is used to indicate that the water balance principle needs to be satisfied in the multi-water-source scheduling process; the reservoir storage capacity constraint is used to indicate that the storage capacity of a first reservoir needs to satisfy a first storage range, the first storage range is used to indicate the maximum storage capacity and the minimum storage capacity of the first reservoir in a first time period, the first reservoir is any reservoir in the to-be-scheduled region, and the first time period is any time period in the first time range and the second time range; the recharge flow constraint is used to indicate that the recharge flow of the first reservoir in the first time period needs to be less than or equal to the maximum recharge flow of the first reservoir in the first time period; the first water demand constraint is used to indicate that the sum of the first storage capacities of the reservoirs in the to-be-scheduled region is greater than or equal to the first water demand in the first time range, the first storage capacity is used to indicate the water amount for water supply in the reservoir; the pump station operating water level constraint is used to indicate that when any pump station takes water from a water intake port, the river water level of the water intake port needs to be located in the operating water level range of the pump station; the pump station water intake capacity constraint is used to indicate that the water intake flow of any pump station in the first time period needs to be less than or equal to a first water intake flow and a second water intake flow, the first water intake flow is the maximum water intake flow of the pump station, and the second water intake flow is the maximum water intake flow of the river where the water intake port of the pump station is located in the first time period; the pump station operation time constraint is used to indicate that any pump station needs to be turned on for at least a first duration before being turned off, and needs to be turned off for at least the first duration before being turned on again, and the first duration is the minimum operation time of the pump station; the river water quality and reservoir water quality constraint is used to indicate that the water quality of any river used for water intake and the water quality of the reservoir need to satisfy the water quality requirement, and the water quality requirement comprises that the chlorine content is less than or equal to the maximum chlorine content; and the seawater desalination water supply capacity constraint is used to indicate that the water supply amount of the seawater desalination plant is less than or equal to the maximum water purification amount of the seawater desalination plant.

[0008] Optionally, the first time range of the first region is obtained by: obtaining first hydrological data of the first region, the first hydrological data comprising the salinity of water in a river where each water intake port of the first region is located; and inputting the first hydrological data into a salt tide prediction model to obtain the first time range; and the salt tide prediction model is a multi-layer long short-term memory network model.

[0009] Optionally, the obtaining the first water demand of the first region in the first time range comprises: obtaining historical water supply data of a plurality of water plants in the first region; constructing an autoregressive moving average model of each water plant based on the historical water supply data of the water plant; determining the water demand of each water plant in the first time range based on the autoregressive moving average model of the water plant, and then determining the first water demand, the first water demand being the sum of the water demands of the water plants in the first time range.

[0010] Optionally, the solving the multi-source scheduling model based on the plurality of objective functions and the plurality of constraints by using the multi-objective genetic algorithm to determine the multi-source scheduling scheme of the first region comprises: solving the multi-source scheduling model by using the multi-objective genetic algorithm to obtain a non-inferior solution set, the non-inferior solution set including a plurality of non-inferior solutions; performing weighted summation on the plurality of objective functions to obtain a total objective function; and taking an optimal non-inferior solution corresponding to the total objective function as the multi-source scheduling scheme of the first region, the optimal non-inferior solution being in the non-inferior solution set.

[0011] The second aspect also provides a multi-source pre-salt-tide scheduling device for salt tide, comprising: a first obtaining module configured to obtain a first time range of a first region, the first time range being a time range in which salt tide intrusion of the first region is predicted to occur; a second obtaining module configured to obtain a first water demand of the first region in the first time range; a modeling module configured to establish a multi-source scheduling model, a plurality of objective functions and a plurality of constraints of the multi-source scheduling model being determined based on the first time range and the first water demand; and a solving module configured to solve the multi-source scheduling model by using a multi-objective genetic algorithm based on the plurality of objective functions and the plurality of constraints to determine a multi-source scheduling scheme, the multi-source scheduling scheme comprising pre-salt-tide scheduling of the first region in a second time range, the second time range being used to indicate a time length of the pre-salt-tide scheduling, and the second time range being prior to the first time range.

[0012] Optionally, the first obtaining module is further configured to obtain first hydrological data of the first region, the first hydrological data comprising salinity of water in a river where each water intake of the first region is located; and input the first hydrological data into a salt tide prediction model to obtain the first time range, wherein the salt tide prediction model is a multi-layer long short-term memory network model.

[0013] Optionally, the second obtaining module is further configured to obtain historical water supply data of a plurality of water plants in the first region; construct an autoregressive moving average model of each water plant based on the historical water supply data of the water plant; determine a water demand of each water plant in the first time range based on the autoregressive moving average model of the water plant, and further determine the first water demand, which is a sum of the water demands of the water plants in the first time range.

[0014] Optionally, the solving module is further configured to solve the multi-source scheduling model by using a multi-objective genetic algorithm to obtain a non-inferior solution set, the non-inferior solution set including a plurality of non-inferior solutions; perform weighted summation on the plurality of objective functions to obtain a total objective function; and take an optimal non-inferior solution corresponding to the total objective function as the multi-source scheduling scheme of the first region, the optimal non-inferior solution being in the non-inferior solution set.

[0015] In a third aspect, a computer device is provided, including a memory and a processor, the memory storing at least one computer program, the at least one computer program being loaded and executed by the processor to perform the method for multi-source scheduling of salt tide described in the above embodiments.

[0016] In a fourth aspect, a computer readable storage medium is provided, the computer readable storage medium storing at least one computer program, the at least one computer program being loaded and executed by a processor to perform the method for multi-source scheduling of salt tide described in the above embodiments.

[0017] In a fifth aspect, a computer program product is provided, including computer programs / instructions, which, when executed by a processor, implement the method of the first aspect.

[0018] The technical solutions provided by the embodiments of the present disclosure have at least the following beneficial effects:

[0019] The multi-source scheduling of salt tide first needs to accurately predict a time period of salt tide occurrence in the first region, that is, the first time range, and then supply water to the first region based on the multi-source scheduling scheme before the first time range. The multi-source scheduling of salt tide before the occurrence of salt tide in the first region can make the water storage in the reservoir of the first region sufficient when the salt tide invades, so as to avoid the situation of urban water shortage.

[0020] And the multi-source scheduling can effectively reduce unexpected situations (such as insufficient water supply for scheduling of a single source) in the single-source scheduling, and improve the reliability of the scheduling.

[0021] In the embodiments of the present disclosure, the first time range of the first region is acquired; the first water demand of the first region in the first time range is acquired; a multi-source scheduling model is established, a plurality of objective functions and a plurality of constraint conditions of the multi-source scheduling model are determined based on the first time range and the first water demand; and the multi-source scheduling model is solved by using a multi-objective genetic algorithm to determine a multi-source scheduling scheme, which can improve the accuracy and reliability of the generated multi-source scheduling scheme, and thus can effectively realize multi-source forecast scheduling before the salt tide arrives. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings needed in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present disclosure, and other drawings can be obtained by those skilled in the art without creative effort based on these drawings.

[0023] Figure 1 A schematic diagram of the water supply relationship between various water sources

[0024] Figure 2 A flowchart of a multi-source forecast scheduling method for salt tide provided by an example embodiment of the present disclosure is shown;

[0025] Figure 3 A flowchart of a multi-source forecast scheduling method for salt tide provided by another example embodiment of the present disclosure is shown;

[0026] Figure 4 A flowchart of training a salt tide prediction model is shown;

[0027] Figure 5 A flowchart of acquiring a first water demand is shown;

[0028] Figure 6 A schematic diagram of a multi-source joint complementary scheduling model is shown;

[0029] Figure 7 A flowchart of an NSGA-III algorithm is shown;

[0030] Figure 8 A structural schematic diagram of a multi-source forecast scheduling device for salt tide provided by an example embodiment of the present disclosure is shown;

[0031] Figure 9 A structural schematic diagram of a computer device provided by an example embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0032] Unless otherwise defined, technical terms or scientific terms used herein shall have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. The terms "first", "second", "third", and the like used in the specification and claims of the present disclosure do not necessarily denote any ordinal, quantity or importance, but are used to distinguish different components. Also, the terms "one", "a", or "an" do not denote a quantity of one, but rather denote the presence of at least one. The terms "including", "comprising", and the like used herein are meant to be inclusive, not exclusive, and specify the presence of stated features, integers, steps, components, or combinations thereof, but do not preclude the presence or addition of one or more other features, integers, steps, components, or combinations thereof.

[0033] For the purposes of the present disclosure, the technical solutions and advantages will be more apparent from the following detailed description of the embodiments of the present disclosure, taken in conjunction with the accompanying drawings.

[0034] In order to facilitate the understanding of the embodiments of the present disclosure, the saltwater intrusion and the types of water sources will be described first.

[0035] Saltwater intrusion: In the dry season, seawater flows into the river with the tide, causing the salinity of the river water to rise sharply, exceeding the drinking water treatment standard, and making the water source unable to be taken out of the water at stages. This phenomenon is called saltwater intrusion.

[0036] Water source: The water source includes pump stations, water plants, reservoirs, and seawater desalination plants. Figure 1 For the water supply relationship between various types of water sources, the following will be described in conjunction with Figure 1 The water supply relationship between various types of water sources will be described.

[0037] Pump station 101 is used to take water from river 102 through a water intake, and the water taken out of river 102 by pump station 101 can be provided to water plant 104 and reservoir 103. In addition, any pump station can also supply water to other pump stations with lower elevation information than the pump station. For example, the terrain of pump station A is higher than that of pump station B, so pump station A can supply water to pump station B, while pump station B cannot supply water to pump station A.

[0038] Reservoir 103 is used for water storage, and the water in reservoir 103 can be provided to water plant 104. In addition, any reservoir can also supply water to other reservoirs with lower elevation information than the reservoir. For example, the terrain of reservoir A is higher than that of reservoir B, so reservoir A can supply water to reservoir B, while reservoir B cannot supply water to reservoir A.

[0039] Water plant 104 is used to provide water to city 105, and water plant 104 can also store a certain amount of water.

[0040] Seawater desalination plant 106 is used to take water from seawater 107 and perform desalination treatment, and the water treated by seawater desalination plant 106 can be provided to water plant 104.

[0041] In the embodiments of the present disclosure, the water sources include not only the pump stations, water plants and reservoirs in the first region, but also the pump stations and reservoirs in other regions outside the first region, and the seawater desalination plants are not limited to the region and only need to supply water to the water plants in the first region.

[0042] Figure 2 A flowchart of a method for forecasting and scheduling multiple water sources for salt tide provided by an example embodiment of the present disclosure is shown, which can be executed by a computer device. Referring to FIG. 1, the method comprises the following steps. Figure 2 The method comprises the following steps.

[0043] In step 201, a first time range of the first region is obtained.

[0044] The first time range is a time range in which salt tide intrusion is predicted to occur in the first region. Here, the time range in which salt tide intrusion occurs is also the time range in which the salinity of water in the river channel where the water intake of the first region is located starts to exceed the standard and ends to return to normal, and the salinity of water in the river channel where the water intake is located exceeding the standard means that salt tide has occurred (the salinity standard of the salinity exceeding the standard is not limited in the embodiments of the present disclosure). For example, it is predicted that the salinity of water in the river channel where the water intake of the first region is located exceeds the standard at time A and ends to return to normal at time B after a period of time, where A is before B, and the first time range of the river channel is [A, B].

[0045] Similarly, a first time range can be predicted for each river channel where the water intake of the first region is located, and then a plurality of first time ranges are determined, and in subsequent planning of the multiple water source scheduling model, the plurality of first time ranges are also used for planning.

[0046] Optionally, the first region is a region where the river channel that may have salt tide intrusion is located. For example, a region where the river channel that has a history of salt tide intrusion is located.

[0047] In step 202, a first water demand of the first region in the first time range is obtained.

[0048] In step 203, a multiple water source scheduling model is established.

[0049] The plurality of objective functions and the plurality of constraint conditions of the multiple water source scheduling model are determined based on the first time range and the first water demand.

[0050] In the embodiments of the present disclosure, the multiple water source scheduling model is a mathematical model, which includes the plurality of objective functions and the plurality of constraint conditions.

[0051] In step 204, based on the plurality of objective functions and the plurality of constraint conditions, a multiple objective genetic algorithm is used to solve the multiple water source scheduling model to determine a multiple water source scheduling scheme.

[0052] The multi-source scheduling scheme includes multi-source forecast scheduling of the first region in a second time range, the second time range is used to indicate the length of time of the forecast scheduling, and the second time range is before the first time range. Here, the second time range is the time required for the forecast scheduling, and the length of the second time range is determined according to the multi-source scheduling scheme.

[0053] Optionally, the multi-source forecast scheduling of the first region in the second time range includes scheduling the pump stations and reservoirs of multiple regions in the first time range, for example, water in the pump stations and reservoirs of regions other than the first region can be scheduled to the reservoirs of the first region in the second time range to achieve pre-water storage.

[0054] Optionally, the multi-source scheduling scheme further includes multi-source scheduling of the first region in the first time range. The multi-source scheduling of the first region in the first time range includes the following two cases:

[0055] The first case: After the salt tide invasion starts, there are some pump stations in the first region that can still take water. For example, because the saltiness exceeding standard start time of each pump station is different, after the salt tide invasion starts, there may be a case that the saltiness of the location of some pump stations in the first region has not exceeded the standard. Such pump stations can still supply water to the reservoirs and water plants in the first region.

[0056] Therefore, in this case, the multi-source scheduling involved includes: the pump stations in the first region, the pump stations outside the first region, and the reservoirs outside the first region supplying water to the reservoirs in the first region; the pump stations in the first region, the pump stations outside the first region, the reservoirs in the first region, and the reservoirs outside the first region supplying water to the water plants in the first region.

[0057] The second case: After the salt tide invasion starts, all the pump stations in the first region cannot take water. In this case, the water source scheduling involved includes: the pump stations outside the first region, the reservoirs outside the first region supplying water to the reservoirs in the first region; the pump stations outside the first region, the reservoirs in the first region, and the reservoirs outside the first region supplying water to the water plants in the first region.

[0058] In the case of multiple objective functions and multiple constraint conditions, a multi-objective genetic algorithm is used to solve the multi-source scheduling model, and a non-inferior solution set including multiple solutions can be obtained. Then, an optimal solution can be selected from the multiple solutions in the non-inferior solution set, and the optimal solution is the final determined multi-source scheduling scheme of the first region.

[0059] The multi-source pre-dispatching for salt tide first needs to accurately predict the time period of salt tide occurrence in the first region, that is, the first time range, and then supply water for the first region before the first time range based on the multi-source dispatching scheme. The multi-source pre-dispatching before the salt tide occurs in the first region can make the water storage in the reservoir of the first region sufficient when the salt tide invades, so as to avoid the situation of urban water shortage.

[0060] And the multi-source pre-dispatching can effectively reduce the unexpected situation (for example, the insufficient water amount of single-source pre-dispatching) and improve the reliability of pre-dispatching.

[0061] In the embodiment of the present disclosure, the first time range of the first region is obtained, the first water demand of the first region in the first time range is obtained, the multi-source dispatching model is established, and the multiple objective functions and multiple constraint conditions of the multi-source dispatching model are determined based on the first time range and the first water demand. The multi-source dispatching scheme is determined by solving the multi-source dispatching model by using the multi-objective genetic algorithm, which can improve the accuracy and reliability of the generated multi-source dispatching scheme, and effectively realize the multi-source pre-dispatching before the salt tide arrives.

[0062] Figure 3 A flow chart of a method for multi-source pre-dispatching for salt tide provided by another example embodiment of the present disclosure is shown, which can be executed by a computer device. Referring to Figure 3 , the method comprises:

[0063] In step 301, the first hydrological data of the first region is obtained.

[0064] The first hydrological data includes the salinity of water in the river channel where each water intake of the first region is located.

[0065] The related content of the first region is described in the foregoing step 301, and is omitted here.

[0066] Optionally, by detecting the water quality of the water in the river channel where each water intake of the first region is located, the salinity of the water in the river channel where each water intake of the first region is located can be obtained.

[0067] In step 302, the first hydrological data is input into a salt tide prediction model to obtain a first time range.

[0068] The salt tide prediction model is a multi-layer long short-term memory (LSTM) model.

[0069] Optionally, the salt tide prediction model is constructed by the following steps a-c. Figure 4 For the flow chart of training the salt tide prediction model, the following will be described in combination withFigure 4 The steps a-c are described.

[0070] Step a, obtaining a first data set for training a salt tide prediction model.

[0071] The first data set includes historical hydrological data of multiple regions. For any one of the multiple regions, the historical hydrological data includes the following data of the region: basin basic geographic data, salt tide data, pump station operation state data, reservoir operation data, river section key section data, reservoir runoff water quality data, river water intake point measured salinity data, and meteorological hydrological data.

[0072] Optionally, the historical hydrological data of a region is obtained by means of field investigation, literature review, data collation, etc. In some embodiments, the historical hydrological data of the region can also be obtained by means of a discussion, etc. to obtain the working log of the hydrological facilities of the region. The hydrological facilities include river hydrological monitoring sites and river water quality monitoring sites.

[0073] Step b, preprocessing the data in the first data set.

[0074] Optionally, the preprocessing includes outlier rejection processing, interpolation processing, data smoothing processing, and data normalization processing.

[0075] The outlier rejection processing is used to remove obvious abnormal data in the first data set. The interpolation processing is used to reasonably interpolate the missing time point data (such as using the mean of the previous and subsequent time points) to improve the continuity and integrity of the data. The data smoothing processing is used to reduce noise interference; for example, a sliding average filtering algorithm can be used to smooth the time series data in the first data set to reduce noise interference. The data normalization processing is used to convert data of different dimensions to the same scale, so that features of different dimensions can be compared and analyzed on the same scale.

[0076] Step c, training an LSTM model based on the preprocessed first data set to obtain a salt tide prediction model.

[0077] When training the LSTM model, not all data in the first data set is used for training, but part of the data in the first data set is selectively used for training according to a target function.

[0078] In the embodiments of the present disclosure, the target function is the salt degree prediction accuracy, which is represented by minimizing the root mean square value between the salt degree predicted by the LSTM model and the actual salt degree, that is, the smaller the root mean square value, the more accurate the salt degree prediction.

[0079] The part of the data selected from the first data set can be stored in a database for convenient query and analysis.

[0080] Optionally, before training, the part of data selected from the first data set is divided into a training set and a test set, and then the training set is used to train the LSTM model, and the test set is used to verify the performance of the LSTM model. When dividing, for example, a 7:3 ratio is used. Here, attention should be paid to the sample balance problem when dividing, that is, the proportion of data of each class in the finally obtained training set and test set needs to be balanced.

[0081] Optionally, during the training process, hyperparameters can also be set, and the model will be iterated according to the hyperparameters during the training process. The hyperparameters include: input feature dimension input_size, hidden layer state dimension hidden_size, LSTM stack number num_layers, hidden layer bias state bias|, batch setting batch_first, dropout rate, learning rate. Optionally, the Bayesian optimization algorithm is used to optimize the hyperparameters of the LSTM model, that is, the calculation error of the LSTM model is calculated after each iteration, until the error of the LSTM model is less than the error threshold, or the number of iterations reaches the maximum number.

[0082] Optionally, during the model iteration process, there are also learning parameters, which are constantly optimized and updated during the model iteration process, for example, the Adam optimization algorithm is used to optimize and update the learning parameters.

[0083] Optionally, verifying the performance of the LSTM model using the test set includes: inputting the saltiness in the test set into the LSTM model, calculating the target function value (that is, the root mean square between the predicted result and the true saltiness) according to the predicted result output by the LSTM model, and evaluating the performance of the LSTM model according to the target function value. When evaluating the performance of the LSTM model, a prediction accuracy threshold can be preset. When the target function value on the test set is greater than or equal to the prediction accuracy threshold, it means that the LSTM model meets the performance requirement, and the LSTM model can be used as a salt tide prediction model; when the target function value is less than the prediction accuracy threshold, it means that the LSTM model does not meet the performance requirement, and needs to be retrained.

[0084] After the saltiness prediction model is trained, the first hydrological data of the first region can be input into the saltiness prediction model to obtain the first time range.

[0085] In step 303, the first water demand of the first region in the first time range is obtained.

[0086] Optionally, step 303 includes steps d-f. Figure 5 The flowchart for obtaining the first water demand is as follows: Figure 5 Steps d-f are described.

[0087] Step d: Obtain historical water supply data from multiple water plants within the first region.

[0088] When obtaining the first dataset in step a of step 302, the historical hydrological data of the first region can be obtained. Then, in step d, the historical water supply data of multiple water plants in the first region can be obtained based on the first dataset.

[0089] Step e: Based on the historical water supply data of each water plant, construct an autoregressive moving average model for each water plant.

[0090] After obtaining historical water supply data from multiple water plants in the first region, before constructing the Auto-Regression and Moving Average (ARMA) model, it is necessary to preprocess the historical water supply data from multiple water plants in the first region, including: removing and replacing outliers and removing noise.

[0091] The outlier removal and replacement process is used to remove outliers from historical water supply data from multiple water plants within a first region. After removing outliers, it's necessary to fill the positions of the outliers with other data, such as neighboring data, to replace them. For example, consider a set of data: 2, 2, 2, 7, 3… In this set, the outlier is 7. We need to remove the outlier 7 and insert data adjacent to it. The data adjacent to the outlier 7 are those adjacent to it, such as 2 or 3, or the average of the two (2.5). After this outlier removal and replacement process, the data set can be 2, 2, 2, 2, 3…

[0092] Noise removal can be achieved using the Kalman filter method, which removes noise from the data, and the remaining data is the filtered data. There are many related technologies that describe the implementation of the Kalman filter method, so a detailed description is omitted here.

[0093] Optionally, within the first region, the first The ARMA model of a water plant is represented by formula (1).

[0094] (1)

[0095] In formula (1), Indicates the first region The water plant Water supply during a given time period. Used to indicate the The number of autoregressive terms for each water plant represents the lag period of the time series data itself; the total number of autoregressive terms is [number missing]. One autoregressive term. is the number of moving average terms of the i th water plant. is the coefficient of the autoregressive term of the i th water plant, is the coefficient of the moving average term of the i th water plant, represents the random error term of the i th water plant in the first region in the time period. In constructing the ARMA model of the i th water plant, first, an initial value of p and q is set, and then other parameters (including the coefficient of the autoregressive term, the coefficient of the moving average term, the random error term, etc.) in the ARMA model of the i th water plant are estimated by using the least square method.

[0096] In the case that the initial values of p and q are different, the estimated other parameters are also different, and a suitable initial value of p and q needs to be selected. In the embodiments of the present disclosure, the AIC score under each combination of p and q is calculated by using the Akaike information criterion (AIC), the combination of p and q with the minimum AIC score and the other parameters estimated based on the combination of p and q are taken as the values of the respective parameters in the ARMA model.

[0097] In the case that the initial values of p and q are different, the estimated other parameters are also different, and a suitable initial value of p and q needs to be selected. In the embodiments of the present disclosure, the AIC score under each combination of p and q is calculated by using the Akaike information criterion (AIC), the combination of p and q with the minimum AIC score and the other parameters estimated based on the combination of p and q are taken as the values of the respective parameters in the ARMA model.

[0098] By taking the values of the parameters in the combination of p and q with the minimum AIC score as the values of the respective parameters in the ARMA model, the performance of the ARMA model can be optimized. For the other water plants in the first region except the i th water plant, the ARMA model can also be constructed by using the above method, and the detailed description is omitted here.

[0099]

[0100] Step f, based on the autoregressive moving average model of each water plant, the water demand of each water plant in the first time range is determined, and then the first water demand is determined.

[0101] ​​​​​​​​​​​​​​​​​​​​​​​​After the ARMA model of each water plant in the first region is determined, the water demand of each water plant in the first region in the first time range can be predicted based on the ARMA model of each water plant in the first region, and the first water demand is the sum of the water demands of the water plants in the first time range.

[0102] In step 304, a multi-water source scheduling model is established.

[0103] The multi-water source scheduling model includes a plurality of objective functions and a plurality of constraint conditions, and the plurality of objective functions and the plurality of constraint conditions of the multi-water source scheduling model are determined based on the first time range and the first water demand.

[0104] Optionally, the objective functions of the multi-water source scheduling model include a water supply guarantee rate objective function, a reservoir water storage quantity objective function, a water supply damage depth objective function, a green economic benefit objective function, and a seawater desalination utilization objective function. Figure 6 For the schematic diagram of the multi-water source joint complementary scheduling model, the following will be described in combination with Figure 6 The five objective functions will be described respectively.

[0105] The water supply guarantee rate objective function is used to indicate that the total water supply guarantee rate of the to-be-scheduled region in the first time range and the second time range is maximum.

[0106] The water supply guarantee rate is used to indicate whether the water demand of the to-be-scheduled region is met in a certain time period. The greater the water supply guarantee rate, the more the water demand of the to-be-scheduled region is met, and the smaller the water supply guarantee rate, the less the water demand of the to-be-scheduled region is met. Therefore, the greater the water supply guarantee rate is, the better.

[0107] The first time range and the second time range are continuous, and the first time range and the second time range can be divided into a plurality of time periods for planning by time period. For example, the starting time of the second time range is T1 moment, and the ending time of the second time range is T2 moment, that is, the salt tide invasion of the first region occurs at T2 moment, and the salt tide invasion of the first region ends at T3 moment. The first time range is the time range from T2 moment to T3 moment. When planning, T1 moment to T3 moment can be divided into a plurality of time periods, and the time length of each time period is equal, so that there is a water demand of the to-be-scheduled region in each time period.

[0108] In this case, the water supply guarantee rate objective function can be represented by formula (2).

[0109] (2)

[0110] In formula (2), is the water supply guarantee rate objective function, is the total number of time periods, indicates whether the water demand of the to-be-dispatched area is met in the first time period, if the water demand of the to-be-dispatched area is met in the first time period, the value is 1, if the water demand of the to-be-dispatched area is not met in the first time period, the value is 0. is an integer, is greater than 0 and less than .

[0111] The water supply guarantee rate target function is to maximize the water supply in the normal water supply period and the salt tide invasion period, to meet the water demand of the to-be-dispatched area in the second time range and the first time range.

[0112] The reservoir storage capacity target function is used to indicate that the reservoirs in the to-be-dispatched area have the maximum storage capacity in the first time range and the second time range.

[0113] The reservoirs in the to-be-dispatched area include reservoirs in the first region and reservoirs outside the first region.

[0114] Optionally, the reservoir storage capacity target function is represented by formula (3).

[0115] (3)

[0116] In formula (3), is the reservoir storage capacity target function, is the total number of reservoirs in the to-be-dispatched area, is the storage capacity of the i-th reservoir in the to-be-dispatched area at the beginning of the j-th time period, is the weight of the i-th reservoir in the to-be-dispatched area. is an integer, is greater than 0 and less than . Other parameters in formula (3) are the same as in formula (2), and details are omitted here. The weight of the i-th reservoir can be set according to experience, and the embodiments of the present disclosure do not limit this. The reservoir storage capacity target function is to fully utilize the available storage capacity of each reservoir in the to-be-dispatched area, to maximize the storage capacity of each reservoir before the salt tide arrives, to meet the water demand of the first region in the case that the river water pumping station in the first region cannot work normally during the salt tide invasion period.

[0117]

[0118] The reservoir storage capacity target function is to fully utilize the available storage capacity of each reservoir in the to-be-dispatched area, to maximize the storage capacity of each reservoir before the salt tide arrives, to meet the water demand of the first region in the case that the river water pumping station in the first region cannot work normally during the salt tide invasion period.

[0119] ​​​​​​​​The water supply damage depth objective function is used to indicate that the total water supply damage depth of the to-be-scheduled region in the first time range and the second time range is minimum. The water supply damage depth is used to indicate the degree of water shortage of the to-be-scheduled region, and the greater the water supply damage depth, the more water shortage the users of the to-be-scheduled region. Here, the water shortage of the users of the to-be-scheduled region means that the total water supply amount of each water source for water supply in a certain time period is less than the total water demand amount of each water plant, so that the smaller the water supply damage depth is in the planning, the better.

[0120] Optionally, the water supply damage depth objective function is represented by formula (4).

[0121] (4)

[0122] In formula (4), is the water supply damage depth objective function, is the actual water supply amount of the to-be-scheduled region in the i-th time period, is the actual water demand amount of the to-be-scheduled region in the i-th time period. Other parameters in formula (4) are the same as those in formula (2), and details are omitted here. The actual water supply amount is the sum of the first water storage amount of each water source of the to-be-scheduled region, and the first water storage amount is used to indicate the water amount for water supply in the reservoir. Assuming that the water storage amount in a certain reservoir is 10L, but in order to ensure the normal operation of the reservoir, a certain amount of water (for example, 1L) needs to be reserved in the reservoir, so the first water storage amount of the reservoir is 9L. For any water source in the to-be-scheduled region, the first water storage amount in the i-th time period can be determined, and then the actual water supply amount can be obtained.

[0123] The actual water demand amount is the sum of the water demand amounts of each water plant (including the water plants in the first region and the water plants outside the first region) in the to-be-scheduled region. For the water demand amount of the water plant in the first region in the i-th time period, the method in step 303 can be used to determine; and for the water demand amount of the water plant outside the first region in the i-th time period, a method similar to that in step 303 can also be used to establish an ARMA model and use the ARMA model to predict the water demand amount. In this way, the sum of the water demand amounts of each water plant in the to-be-scheduled region can be determined.

[0124] The actual water demand amount is the sum of the water demand amounts of each water plant (including the water plants in the first region and the water plants outside the first region) in the to-be-scheduled region. For the water demand amount of the water plant in the first region in the i-th time period, the method in step 303 can be used to determine; and for the water demand amount of the water plant outside the first region in the i-th time period, a method similar to that in step 303 can also be used to establish an ARMA model and use the ARMA model to predict the water demand amount. In this way, the sum of the water demand amounts of each water plant in the to-be-scheduled region can be determined. The water supply damage depth objective function is used to ensure that the water demand amount of the to-be-scheduled region in any time period is satisfied as much as possible.

[0125] The water supply damage depth objective function is used to ensure that the water demand amount of the to-be-scheduled region in any time period is satisfied as much as possible.

[0126] ​​​Green economic benefit objective function: The green economic benefit objective function is used to indicate the minimum energy consumption of the pumping stations in the area to be dispatched within the first and second time ranges.

[0127] The pumping stations in the area to be dispatched include those within the first area and those outside the first area.

[0128] Here, the energy consumption of the pumping station is related to the pumping head; the smaller the pumping head, the less energy the pumping station consumes.

[0129] Pump head refers to the height to which a pumping station can lift water. The head of a given pump equals the net head plus the head loss. Head loss is a fixed value and cannot be changed. Net head refers to the elevation difference between the pump's suction point and the highest control point. For example, if a pump draws water from a river and delivers it to a higher water tank, the net head is simply the elevation difference between the river's surface and the higher water tank.

[0130] It can be seen that the higher the river level, the smaller the net head of the pumping station, and the lower the pumping head, the less energy is consumed. Therefore, any pumping station should draw water when the river water level at its intake is high, thereby reducing the energy consumption of the pumping station.

[0131] Optionally, the objective function for green economic benefits can be expressed by formula (5).

[0132] (5)

[0133] In formula (5), The objective function is the green economic benefit; The average water level at the intake of each pumping station in the area to be dispatched during the start-up period; This represents the total number of pump stations within the area to be dispatched; For the first The pump station is at the first On / off status within a time period It can be 1 or 0. Taking 1 indicates the first... The pump station is at the first The device is powered on during a specific time period. When the value is 0, it indicates the first... The pump station is at the first The device was switched off for a certain period of time. In the first Within the time period, the first The river water level at the water intake of each pumping station; It is an integer. Greater than 0 and less than The other parameters in formula (5) are the same as those in formula (2), and will not be detailed here.

[0134] The green economic benefit objective function is used to ensure that the average water level at the intake of each pumping station in the area to be dispatched is the highest during the start-up period. This is equivalent to maximizing the water intake of each pumping station during periods of high water level, thereby reducing energy consumption and achieving the goal of green economic benefits.

[0135] Objective function for seawater desalination utilization: The objective function for seawater desalination utilization is used to indicate the minimum amount of seawater desalination utilization in the area to be scheduled within the first and second time ranges.

[0136] Optionally, the objective function for seawater desalination is expressed by formula (6).

[0137] (6)

[0138] In formula (6), This represents the total number of seawater desalination plants within the area to be dispatched. In the first Within the time period, the first The water supply of a seawater desalination plant. It is an integer. Greater than 0 and less than The other parameters in formula (6) are the same as those in formula (2), and will not be described in detail here.

[0139] Seawater desalination is a relatively expensive way to obtain fresh water. Under the premise of maximizing water supply reliability, existing water supply projects such as rivers and reservoirs should be fully utilized, with priority given to direct supply from freshwater sources. Seawater desalination should only be used when freshwater sources cannot guarantee normal water supply. This will minimize the amount of water supplied through desalination and reduce water supply costs.

[0140] Optionally, the constraints of this multi-source water scheduling model include: water balance constraints, reservoir storage constraints, replenishment flow constraints, primary water demand constraints, pump station operating water level constraints, pump station water intake capacity constraints, pump station operating time constraints, river water quality and reservoir water quality constraints, and seawater desalination supply capacity constraints. The following is a combination of... Figure 6 Each of these nine constraints will be explained.

[0141] Water balance constraints: Water balance constraints are used to indicate the water balance principle that must be met during the scheduling of multiple water sources.

[0142] The principle of water balance means that the total amount of water in the entire water supply system will not change.

[0143] In the process of multi-source water allocation, the water storage capacity of each reservoir, the water replenishment flow from other water sources, and the water discharge flow from the reservoir itself must meet the principle of water balance.

[0144] For the entire multi-source water dispatching process, the principle of water balance must also be met between the incoming water, the stored water volume, and the water consumption.

[0145] Optionally, the water balance constraint is expressed by formulas (7) to (8). Formula (7) is used to express the water balance constraint for a single reservoir, and formula (8) is used to express the water balance constraint in the entire multi-source water dispatching process.

[0146] (7)

[0147] In formula (7), In the first At the beginning of the first time period (that is, the first time period) (End of the time period) The water storage capacity of each reservoir In the first Within the time period, the first The inflow rate of each reservoir, In the first Within the time period, the first The water supply flow of each reservoir. In the first Within the time period, the first The discharge flow of a reservoir is used to indicate the water flow that meets the requirements of functions other than water supply, such as downstream ecological flow. In the first Within the time period, the first The sum of the replenishment flow that each reservoir receives from other water sources. In the first Within the time period, the first The reservoir discharges water to the first... The sum of the discharge from all reservoirs other than the one mentioned above. The length of the time period is calculated in units. The other parameters in formula (7) have the same meaning as the parameters in formulas (2), (3) and (5), and are omitted here in detail.

[0148] (8)

[0149] In formula (8), In the first Within the time period, the first The water intake flow rate of each pumping station, and the other parameters in formula (8) have the same meaning as the parameters in formula (7) and formula (4), which are omitted here.

[0150] Reservoir storage capacity constraint: The reservoir storage capacity constraint is used to indicate that the water storage capacity in the first reservoir must meet the first storage range. Among them, the first storage range is used to indicate the maximum and minimum water storage capacity of the first reservoir within the first time period. The first reservoir is any reservoir within the area to be dispatched, and the first time period is any time period within the first time range and the second time range.

[0151] Here, the first water storage range refers to the boundary conditions of each reservoir. The boundary conditions of each reservoir are different at different times, and the water storage volume of the first reservoir at any time must meet the boundary conditions of that time period. For example, for a water supply reservoir that also has flood control functions, the maximum water storage volume of the reservoir during the flood season cannot exceed the reservoir capacity corresponding to the flood limit water level, while during the non-flood season, it can be stored up to the maximum beneficial storage capacity.

[0152] Optionally, the water balance constraint is expressed by formula (9).

[0153] (9)

[0154] In formula (9), In the first Within the time period, the first The minimum water storage capacity of each reservoir, which is also the lower limit of the first water storage range, In the first Within the time period, the first The maximum water storage capacity of a reservoir is also the upper limit of the first water storage range. The meanings of the other parameters in formula (9) are the same as those in formulas (2) and (3), and are omitted here.

[0155] Replenishment flow constraint: The replenishment flow constraint is used to indicate that the replenishment flow of the first reservoir in the first time period must be less than or equal to the maximum replenishment flow of the first reservoir in the first time period.

[0156] The replenishment flow between different reservoirs at different times is constrained by the maximum flow capacity of the water diversion pipes or channels, as well as other factors such as policies and regulations. For example, if two reservoirs belong to two different administrative regions, and the policies of these two regions set upper limits on the replenishment flow between them, then the replenishment flow between different reservoirs at different times should be less than or equal to the maximum allowable replenishment flow for that time period.

[0157] Optionally, the supply flow constraint is expressed by formula (10).

[0158] (10)

[0159] In formula (10) Indicates the first Within the time period, the first The maximum recharge flow of each reservoir. The meanings of the other parameters in formula (10) are the same as those in formula (7), and are omitted here.

[0160] First water demand constraint: The first water demand constraint indicates that within a first time period, the sum of the first water storage volumes of all reservoirs in the area to be dispatched is greater than or equal to the first water demand. Here, the sum of the first water storage volumes of all reservoirs is the actual water storage volume in the objective function for the depth of water supply disruption.

[0161] If the pumping stations in the first area are unable to draw water within the first time frame, the actual water storage capacity of each reservoir in the waiting dispatch area should meet the first water demand so that the first area does not suffer from water shortage.

[0162] Optionally, the first water demand constraint is expressed by formula (11).

[0163] (11)

[0164] In formula (11), This refers to the start time of the first time range. The duration of the first time range, that is... It equals the end time of the first time range. For the first The reservoir is Water storage at any given time For the first The dead storage capacity of a reservoir, that is, the dead storage capacity of the reservoir. The lowest water level that is allowed to drop to under normal operating conditions. For safety reasons, , For the area to be dispatched in the first The water demand over a given time period. The meanings of the other parameters in formula (11) are the same as those in formula (2), and will not be detailed here.

[0165] Pump station operating water level constraint: The pump station operating water level constraint is used to indicate that when any pump station draws water from its water intake, the river water level at the water intake must be within the operating water level range of the pump station.

[0166] The pumping station can only operate normally when the river water level at the water intake is within the working water level range. Optionally, the working water level constraint of the pumping station is expressed by formula (12).

[0167] (12)

[0168] In formula (12), Indicates the first Within the time period, the first The river water level at the water intake where each pumping station is located Indicates the first The lower limit of the operating water level of each pumping station. Indicates the first The upper limit of the working water level of each pumping station. The meanings of the other parameters in formula (12) are the same as those in formulas (2) and (5), and are omitted here in detail.

[0169] Pump station water intake capacity constraint: The pump station water intake capacity constraint is used to indicate that the water intake flow of any pump station in the first time period must be less than or equal to the first water intake flow and the second water intake flow. The first water intake flow is the maximum water intake flow of the pump station, and the second water intake flow is the maximum water intake flow of the river channel where the pump station's water intake is located in the first time period.

[0170] Due to limitations in pump station output, operational safety, and permissible water intake flow in the river, as well as other engineering practical and boundary conditions, the pump station flow rate cannot be arbitrarily increased, resulting in constraints on the pump station's water intake capacity.

[0171] Optionally, the water intake capacity constraint of the pumping station is expressed by formula (13).

[0172] (13)

[0173] In formula (13), For the first The first water intake flow of each pumping station For the first The second water intake flow rate of the pumping station indicates the first The water intake of the pumping station is located in the river channel of the [missing information]. The maximum water intake flow rate within a time period. The meanings of the other parameters in formula (13) are the same as those in formula (8), and are omitted here.

[0174] Pump station operating time constraints: Pump station operating time constraints are used to indicate that any pump station must wait at least a first duration after being turned on before it can be turned off, and any pump station must wait at least a first duration after being turned off before it can be turned on again. The first duration is the minimum operating time of the pump station.

[0175] In actual scheduling, due to the difficulty and workload of manual operation and the characteristics of pump station engineering, it is impossible for the pump station to be turned on and off at any time according to the external water level. That is, after the pump station enters a certain state (such as on or off), it needs to maintain it for a period of time before it can switch to another state.

[0176] Optionally, the pump station operating time constraint is expressed by formula (14).

[0177] (14)

[0178] In formula (14), For the first duration, for The duration for which a pump station maintains operation in a certain state (e.g., on or off) after entering that state.

[0179] River and reservoir water quality constraints: These constraints indicate that the water quality of any river or reservoir used for water intake must meet specific requirements. These requirements include a chlorine content that is less than or equal to the maximum chlorine content.

[0180] The water quality of rivers and reservoirs must meet the required standards. If the chlorine content in the water is too high, exceeding the maximum permissible chlorine content for water plant intake or industrial use, water cannot be drawn from the river or reservoir. In other special circumstances, such as a sudden pollution incident upstream of a river or reservoir causing severe deterioration of the water quality, exceeding the permissible raw water quality requirements for water plant intake or industrial use, water intake will also be prohibited. Therefore, during the scheduling period, the water quality at river intake points and in reservoirs must meet the required standards.

[0181] Optionally, the water quality requirements also include that the concentration of any water quality indicator in the water body must be less than or equal to the concentration threshold corresponding to that water quality indicator. Water quality indicators include, for example, COD (Chemical Oxygen Demand) and NH3-N (an ammonia nitrogen content indicator). The concentration threshold corresponding to any water quality indicator is an empirical value, and this embodiment does not limit the value of the concentration threshold.

[0182] Optionally, the water quality of the river and the water quality constraints of the reservoir are expressed by formulas (15) and (16).

[0183] (15)

[0184] In formula (15), In the first Within the time period, the first Chlorine content of the reservoir In the first Within the time period, the first The chlorine content of the river water at the water intake where each pumping station is located The maximum chlorine content. The meanings of the other parameters in formula (15) are the same as those in formulas (2), (3) and (5), and are omitted here.

[0185] (16)

[0186] In formula (15), In the first Within the time period, the first For any water quality indicator of a reservoir, In the first Within the time period, the first Any water quality index of the river water body at the water intake of each pumping station This is the concentration threshold corresponding to the water quality index. The meanings of the other parameters in formula (16) are the same as those in formulas (2), (3) and (5), and are omitted here.

[0187] Seawater desalination supply capacity constraints: Seawater desalination supply capacity constraints are used to indicate that the water supply of a seawater desalination plant is less than or equal to the maximum net water capacity of the seawater desalination plant.

[0188] Seawater desalination plants have an upper limit to their operating capacity (maximum net water volume), therefore the water supply of a seawater desalination plant should be less than or equal to this upper limit.

[0189] Optionally, the seawater desalination water supply capacity constraint is expressed by formula (17).

[0190] (17)

[0191] In formula (17), In the first Within the time period, the first The maximum net water output of a seawater desalination plant. The meanings of the other parameters in formula (17) are the same as those in formula (6), and are omitted here.

[0192] After determining the above-mentioned multiple objective functions and multiple constraints, a multi-source scheduling model can be established based on the multiple objective functions and multiple constraints.

[0193] In step 305, a multi-objective genetic algorithm is used to solve the multi-water source scheduling model based on multiple objective functions and multiple constraints to determine the multi-water source scheduling scheme.

[0194] Optionally, the multi-objective genetic algorithm used in this embodiment is the NSGA-Ⅲ (Non-dominated Sorting Genetic Algorithm III) algorithm.

[0195] Optionally, step 305 includes the following step gi.

[0196] Step g: Use a multi-objective genetic algorithm to solve the multi-water source scheduling model and obtain a set of non-dominated solutions.

[0197] The set of non-dominated solutions includes multiple non-dominated solutions.

[0198] Optionally, when the multi-objective genetic algorithm is the NSGA-Ⅲ algorithm, step g includes the following five steps. Figure 7The flowchart of the NSGA-III algorithm is shown in the following Figure 7 The step g is described. In the step g, gen is used to count the number of iteration evolutions, and gen equal to 1 indicates that the first iteration evolution is performed, and gen equal to 2 indicates that the second iteration evolution is performed. Figure 7

[0199] In the first step, the algorithm parameters of the NSGA-III algorithm are set.

[0200] Optionally, the algorithm parameters of the NSGA-III algorithm include the population size, the maximum iteration number, and the probabilities of selection, crossover and mutation. The setting mode of the algorithm parameters is more in the related art, and thus is omitted here.

[0201] In the second step, based on the multi-source scheduling model, an initial parent population is randomly generated, the objective function values of all individuals in the parent population are calculated according to the set algorithm parameters, and all individuals in the parent population are sorted according to the non-dominated relationship according to different objective functions. In the sorting, all individuals in the parent population are selected, crossed and mutated according to the set fitness value to obtain a first generation of offspring population.

[0202] Here, the initial parent population includes multiple individuals, the multi-source scheduling model includes five objective functions, and thus five objective function values can be calculated for each individual. The initial parent population and the first generation of offspring population constitute the first iteration evolution.

[0203] In the third step, the parent population and the offspring population are combined, the combined population is quickly non-dominated sorted, the crowding degree is calculated, and the appropriate individuals are selected according to the crowding degree value to form a new parent population.

[0204] The process of the second iteration evolution is briefly described as follows: firstly, the initial parent population and the first generation of offspring population are combined, the combined population is quickly non-dominated sorted, the crowding degree is calculated, and the appropriate individuals are selected according to the crowding degree value to form a new parent population. Then, the fourth step is executed to obtain the offspring population of the new parent population (i.e., the second generation of offspring population). The new parent population and the second generation of offspring population constitute the second iteration evolution.

[0205] Each iteration evolution after the second iteration evolution is performed in a similar manner to the second iteration evolution, that is, the third step and the fourth step are repeatedly executed, and thus the detailed description is omitted here.

[0206] In the fourth step, the offspring population is generated by selection, crossover and mutation operators.

[0207] In the fifth step, the third step and the fourth step are performed until the iteration number reaches the maximum iteration number, and a non-inferior solution set is obtained.

[0208] ​The implementation of each step of the NSGA-III algorithm is known in the related art, and thus is not described in detail.

[0209] Step h: performing weighted summation on the plurality of objective functions to obtain a total objective function.

[0210] In the weighted summation of the plurality of objective functions, the weight coefficient of each objective function needs to be determined.

[0211] By way of example, the weight coefficient of each objective function can be determined by investigating water conservancy, water affairs, environmental, municipal and other stakeholders in the region to be dispatched, and by expert consultation.

[0212] Optionally, the total objective function is represented by formula (18).

[0213] (18)

[0214] In formula (18), are weight coefficients of , and . is the total objective function. The meanings of other parameters in formula (18) are the same as those in formulas (2) to (6), and thus are not described in detail.

[0215] Step i: taking the optimal non-inferior solution corresponding to the total objective function as the multi-source dispatching scheme of the first region.

[0216] The optimal non-inferior solution corresponding to the total objective function is in the non-inferior solution set.

[0217] The essence of step i is to find an optimal non-inferior solution from the non-inferior solution set by using the total objective function. Here, after the total objective function is determined, each non-inferior solution in the non-inferior solution set can be used to calculate a total objective function value according to the total objective function, and the non-inferior solution with the largest total objective function value is the optimal non-inferior solution corresponding to the total objective function. The optimal non-inferior solution is the multi-source dispatching scheme of the first region.

[0218] In the optimal non-inferior solution, both the multi-source forecast dispatching of the first region in the second time range and the multi-source dispatching of the first region in the first time range are included.

[0219] The multi-source forecast dispatching for salt tide first needs to accurately predict the time period of salt tide occurrence in the first region, that is, the first time range, and then supply water to the first region based on the multi-source dispatching scheme before the first time range. The multi-source forecast dispatching is performed before the salt tide occurs in the first region, so that the water storage in the reservoir of the first region is sufficient when the salt tide invades, and the situation of urban water shortage does not occur.

[0220] Using multiple water sources for forecasting and scheduling can effectively reduce unexpected situations that may occur when using a single water source for forecasting and scheduling (such as insufficient water available for scheduling from a single source), and improve the reliability of forecasting and scheduling.

[0221] In this embodiment of the disclosure, by obtaining a first time range of a first region; obtaining a first water demand of the first region within the first time range; establishing a multi-source water scheduling model, wherein multiple objective functions and multiple constraints of the multi-source water scheduling model are determined based on the first time range and the first water demand; and using a multi-objective genetic algorithm to solve the multi-source water scheduling model and thus determine the multi-source water scheduling scheme, the accuracy and reliability of the generated multi-source water scheduling scheme can be improved, thereby effectively realizing multi-source water forecasting and scheduling before the arrival of saltwater intrusion, as well as multi-source water scheduling during the saltwater intrusion process.

[0222] The following are device embodiments of this application. For details not described in detail in the device embodiments, please refer to the above method embodiments.

[0223] Figure 8 A schematic diagram of a multi-source water source forecasting and scheduling device for saltwater intrusion provided in an exemplary embodiment of this disclosure is shown. See also Figure 8 The multi-source water forecasting and scheduling device 800 for saltwater intrusion includes: a first acquisition module 801, a second acquisition module 802, a modeling module 803, and a solution module 804.

[0224] The first acquisition module 801 is used to acquire the first time range of the first region, which is the predicted time range for saltwater intrusion in the first region.

[0225] The second acquisition module 802 is used to acquire the first water demand of the first region within the first time range.

[0226] Modeling module 803 is used to establish a multi-source water scheduling model. The multiple objective functions and multiple constraints of the multi-source water scheduling model are determined based on the first time range and the first water demand.

[0227] The solution module 804 is used to solve the multi-water source scheduling model based on multiple objective functions and multiple constraints using a multi-objective genetic algorithm to determine the multi-water source scheduling scheme. The multi-water source scheduling scheme includes forecasting scheduling of the first region within a second time range. The second time range is used to indicate the duration of the forecast scheduling and is prior to the first time range.

[0228] Optionally, the first acquisition module 801 is further configured to acquire first hydrological data of the first region, the first hydrological data including the salinity of the water in the river where each water intake is located in the first region; input the first hydrological data into the saltwater intrusion prediction model to obtain a first time range; wherein, the saltwater intrusion prediction model is a multilayer long short-term memory network model.

[0229] Optionally, the second obtaining module 802 is further configured to obtain historical water supply data of a plurality of water plants in the first region; construct an autoregressive moving average model of each water plant based on the historical water supply data of each water plant; determine the water demand of each water plant in the first time range based on the autoregressive moving average model of each water plant, and further determine the first water demand, which is the sum of the water demand of each water plant in the first time range.

[0230] Optionally, the solving module 804 is further configured to solve the multi-source scheduling model by using a multi-objective genetic algorithm to obtain a non-inferior solution set, the non-inferior solution set including a plurality of non-inferior solutions; perform weighted summation on the plurality of objective functions to obtain a total objective function; and take an optimal non-inferior solution corresponding to the total objective function as the multi-source scheduling scheme of the first region, the optimal non-inferior solution being in the non-inferior solution set.

[0231] It should be noted that: the multi-source prediction and scheduling device for salt tide provided in the above embodiments is used for multi-source prediction and scheduling, and only the division of the above functional modules is used as an example for illustration. In actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the multi-source prediction and scheduling device for salt tide provided in the above embodiments and the multi-source prediction and scheduling method for salt tide belong to the same concept, and the specific implementation process is detailed in the method embodiments, which will not be repeated here.

[0232] The division of the modules in the embodiments of the present disclosure is illustrative, and is only a logical function division. In actual implementation, another division manner can be used. In addition, each functional module in each embodiment of the present disclosure can be integrated in one processor, or can be a separate physical existence, or two or more modules can be integrated into one module. The integrated module can be realized in the form of hardware or in the form of a software functional module.

[0233] The integrated module, if implemented in the form of a software function module and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present disclosure, essentially or in other words, the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling an end device (which can be a personal computer, a mobile phone, or a communication device, etc.) or a processor (processor) to perform all or part of the steps of the methods according to the embodiments of the present disclosure. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (read-only memory, ROM), random access memory (random access memory, RAM), magnetic disk or optical disk, and various other media that can store program codes.

[0234] Figure 9 is a structural schematic diagram of a computer device provided by an embodiment of the present disclosure. As shown in Figure 9 the computer device 900 includes a processor 901 and a memory 902.

[0235] The processor 901 can include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 901 can be implemented in at least one of the hardware forms of a DSP (Digital Signal Processing), an FPGA (Field-Programmable Gate Array), and a PLA (Programmable Logic Array). The processor 901 can also include a main processor and a coprocessor. The main processor is a processor for processing data in an awake state, also known as a CPU (Central Processing Unit). The coprocessor is a low-power processor for processing data in a standby state. In some embodiments, the processor 901 can be integrated with a GPU (Graphics Processing Unit). The GPU is responsible for rendering and drawing the content required to be displayed by the display screen. In some embodiments, the processor 901 can also include an AI (Artificial Intelligence) processor. The AI processor is used to process machine learning-related computing operations.

[0236] The memory 902 can include one or more computer-readable storage media. The memory 902 can also include high-speed random access memory and non-volatile, computer-readable storage media such as one or more magnetic disk storage devices, optical storage devices, flash memory devices, solid-state storage devices, or any other non-transitory computer- readable storage media that stores data. In some embodiments, the non-transitory computer- readable storage medium in the memory 902 stores at least one instruction for execution by the processor 901 to implement the method for multi-water source forecast scheduling of salt tide provided in the embodiments of the present disclosure.

[0237] Those skilled in the art can understand that the structure shown in the above Figure 9 The structure shown in the above does not constitute a limitation on the computer device 900, and can include more or less components than shown, or combine certain components, or arrange the components differently.

[0238] The embodiments of the present disclosure also provide a non-transitory computer- readable storage medium, when instructions in the storage medium are executed by a processor of a computer device, the computer device is enabled to perform the method for multi-water source forecast scheduling of salt tide provided in the embodiments of the present disclosure.

[0239] The embodiments of the present disclosure also provide a computer program product, including computer programs / instructions, when the computer programs / instructions are executed by a processor, the method for multi-water source forecast scheduling of salt tide provided in the embodiments of the present disclosure is implemented.

[0240] The above only describes optional embodiments of the present disclosure, and does not limit the present disclosure. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A multi-source water source forecasting and scheduling method for salinity intrusion, characterized in that, The method includes: Obtain the first time range of the first region, which is the predicted time range for saltwater intrusion in the first region; Obtain the first water demand of the first region within the first time range, wherein the first water demand is the sum of the water demand of each water plant in the first region within the first time range; A multi-source water scheduling model is established. The multiple objective functions and multiple constraints of the multi-source water scheduling model are determined based on the first time range and the first water demand. The water sources in the multi-source water scheduling model include pumping stations, water plants, reservoirs, and seawater desalination plants. The multiple objective functions of the multi-source water scheduling model include: water supply guarantee rate objective function, reservoir storage capacity objective function, water supply damage depth objective function, green economic benefit objective function, and seawater desalination utilization objective function. Based on the multiple objective functions and multiple constraints, a multi-objective genetic algorithm is used to solve the multi-water source scheduling model to determine a multi-water source scheduling scheme. The multi-water source scheduling scheme includes forecasting scheduling of the first region within a second time range. The second time range is used to indicate the duration of the forecasted scheduling and is prior to the first time range. The water supply guarantee rate objective function is used to indicate that the total water supply guarantee rate of the area to be dispatched is maximized within the first time range and the second time range. The area to be dispatched includes the first region and other regions that supply water to the first region. The objective function for reservoir storage capacity is used to indicate the maximum water storage capacity of the reservoirs in the area to be dispatched within the first time range and the second time range. The objective function for water supply disruption depth is used to indicate that the total water supply disruption depth of the area to be dispatched is minimized within the first time range and the second time range, and the water supply disruption depth is used to indicate the degree of water shortage in the area to be dispatched. The green economic benefit objective function is used to indicate that the pumping stations in the area to be dispatched consume the least amount of energy in the first time range and the second time range; The objective function for seawater desalination utilization is used to indicate the minimum amount of seawater desalination utilization in the area to be scheduled within the first time range and the second time range.

2. The method according to claim 1, characterized in that, The constraints of the multi-source water dispatch model include: water balance constraints, reservoir storage constraints, replenishment flow constraints, first water demand constraints, pump station operating water level constraints, pump station water intake capacity constraints, pump station operating time constraints, river water quality and reservoir water quality constraints, and seawater desalination supply capacity constraints. The water balance constraint is used to indicate that the water balance principle must be met during the multi-water source scheduling process; The reservoir storage capacity constraint is used to indicate that the storage capacity in the first reservoir must meet the first storage range. The first storage range is used to indicate the maximum and minimum storage capacity of the first reservoir within a first time period. The first reservoir is any reservoir within the area to be dispatched, and the first time period is any time period within the first time range and the second time range. The replenishment flow constraint is used to indicate that the replenishment flow of the first reservoir during the first time period must be less than or equal to the maximum replenishment flow of the first reservoir during the first time period. The first water demand constraint is used to indicate that within the first time range, the sum of the first water storage volumes of each reservoir in the area to be dispatched is greater than or equal to the first water demand, and the first water storage volume is used to indicate the amount of water in the reservoir used for water supply. The pump station operating water level constraint is used to indicate that when any pump station draws water from its water intake, the river water level at the water intake must be within the operating water level range of the pump station. The pump station water intake capacity constraint is used to indicate that the water intake flow of any pump station in the first time period must be less than or equal to the first water intake flow and the second water intake flow. The first water intake flow is the maximum water intake flow of the pump station, and the second water intake flow is the maximum water intake flow of the river where the pump station's water intake is located in the first time period. The pump station operating time constraint is used to indicate that any pump station must wait at least a first duration after being turned on before it can be turned off, and any pump station must wait at least the first duration after being turned off before it can be turned on again, where the first duration is the shortest operating time of the pump station. The river and reservoir water quality constraints are used to indicate that the water quality of any river or reservoir used for water intake must meet the water quality requirements, which include a chlorine content less than or equal to the maximum chlorine content. The seawater desalination supply capacity constraint is used to indicate that the water supply of the seawater desalination plant is less than or equal to the maximum net water volume of the seawater desalination plant.

3. The method according to claim 1 or 2, characterized in that, The acquisition of the first time range of the first region includes: Obtain first hydrological data for the first region, including the salinity of the water in the river channels where each water intake is located in the first region; The first hydrological data is input into the saltwater intrusion prediction model to obtain the first time range; The saltwater intrusion prediction model is a multilayer long short-term memory network model.

4. The method according to claim 1 or 2, characterized in that, The step of obtaining the first water demand of the first region within the first time range includes: Obtain historical water supply data from multiple water plants within the first region; Based on the historical water supply data of each water plant, an autoregressive moving average model is constructed for each water plant. Based on the autoregressive moving average model of each water plant, the water demand of each water plant within the first time range is determined, and thus the first water demand is determined.

5. The method according to claim 1 or 2, characterized in that, The process of solving the multi-water source scheduling model using a multi-objective genetic algorithm based on the multiple objective functions and multiple constraints to determine the multi-water source scheduling scheme for the first region includes: A multi-objective genetic algorithm is used to solve the multi-water source scheduling model to obtain a set of non-dominated solutions, which includes multiple non-dominated solutions. The total objective function is obtained by weighted summation of the multiple objective functions. The optimal non-dominated solution corresponding to the overall objective function is taken as the multi-water source scheduling scheme for the first region, and the optimal non-dominated solution corresponding to the overall objective function is in the set of non-dominated solutions.

6. A multi-source water source forecasting and scheduling device for salinity intrusion, characterized in that, The device includes: The first acquisition module is used to acquire the first time range of the first region, wherein the first time range is the predicted time range of the first region where saltwater intrusion will occur; The second acquisition module is used to acquire the first water demand of the first region within the first time range, wherein the first water demand is the sum of the water demand of each water plant in the first region within the first time range; The modeling module is used to establish a multi-source water scheduling model. The multiple objective functions and multiple constraints of the multi-source water scheduling model are determined based on the first time range and the first water demand. The water sources in the multi-source water scheduling model include pumping stations, water plants, reservoirs, and seawater desalination plants. The multiple objective functions of the multi-source water scheduling model include: water supply guarantee rate objective function, reservoir storage capacity objective function, water supply damage depth objective function, green economic benefit objective function, and seawater desalination utilization objective function. The solution module is used to solve the multi-water source scheduling model using a multi-objective genetic algorithm based on the multiple objective functions and the multiple constraints, so as to determine the multi-water source scheduling scheme. The multi-water source scheduling scheme includes forecasting scheduling of the first region within a second time range. The second time range is used to indicate the duration of the forecasted scheduling and is prior to the first time range. The water supply guarantee rate objective function is used to indicate that the total water supply guarantee rate of the area to be dispatched is maximized within the first time range and the second time range. The area to be dispatched includes the first region and other regions that supply water to the first region. The objective function for reservoir storage capacity is used to indicate the maximum water storage capacity of the reservoirs in the area to be dispatched within the first time range and the second time range. The objective function for water supply disruption depth is used to indicate that the total water supply disruption depth of the area to be dispatched is minimized within the first time range and the second time range, and the water supply disruption depth is used to indicate the degree of water shortage in the area to be dispatched. The green economic benefit objective function is used to indicate that the pumping stations in the area to be dispatched consume the least amount of energy in the first time range and the second time range; The objective function for seawater desalination utilization is used to indicate the minimum amount of seawater desalination utilization in the area to be scheduled within the first time range and the second time range.

7. A computer device, characterized in that, The computer device includes a memory and a processor, wherein the memory stores at least one computer program, which is loaded and executed by the processor to implement the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one computer program, which is loaded and executed by a processor to implement the method of any one of claims 1 to 5.

9. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the method described in any one of claims 1 to 5.

Images