A water quantity optimization configuration method and device based on topological correlation

Abstract

The application discloses a water quantity optimal allocation method and equipment based on topological correlation and belongs to the technical field of water resource management. The existing water distribution scheme does not consider the coupling relationship between water using units and the space-time correlation, so the existing water quantity allocation scheme cannot realize the optimal distribution of water use of each unit. The water quantity optimal allocation method based on topological correlation constructs a topological simulation module, a first water distribution calculation module and an optimization decision module, captures the coupling relationship between water using units and the space-time correlation for an unconstrained scene, dynamically adjusts the distribution strategy and obtains an optimal water distribution scheme. For a constrained scene, the water distribution problem is converted into a multivariable optimization problem to obtain an optimal water distribution scheme. Therefore, the coupling relationship between water using units and the space-time correlation can be fully considered, the optimal distribution of water use of each unit can be realized, the waste of water resources is avoided, the contradiction between supply and demand is reduced, and the sustainable development of water supply economy can be realized.

Description

TECHNICAL FIELD

[0001] The present application relates to a water quantity optimization allocation method and device based on topological correlation, belonging to the technical field of water resources management. BACKGROUND

[0002] A water resources allocation model based on water source-user topological relationship priority scheduling is proposed in Chinese literature (Liu Yisheng, Yan Shaofeng. Research and application of water resources allocation model based on topological relationship [J]. China Rural Water Resources and Hydropower, 2022 (11): 26-31. DOI: 10.12396 / znsd.220096.). The model materializes the water demand of different regions and industries to water users, and through the construction of topological relationship of water source and water user interaction, water user water priority order, and space distribution relationship of series or parallel connection between water sources, etc. The priority scheduling is mainly used to obtain the order of all water supply behaviors of water source-user in the region through unified coding, to meet the requirements of irrigation and water supply design guarantee rate in the region, and to consider the fairness between water users, so as to realize the optimal allocation of water resources.

[0003] The above allocation method mainly allocates water resources according to the priority and weight of each water unit, but does not consider the coupling relationship between water units and the space-time correlation. For example, there may be a negative correlation between an industrial water unit and a residential water unit (industrial water saving may increase the available amount of residential water), so the existing water allocation scheme cannot achieve optimal allocation of water use in each unit.

[0004] In particular, when facing insufficient water quantity, if the internal law of water quantity transmission and allocation between different units is not considered, and the space-time characteristics of water use mode cannot be identified, there will be a large gap between the actual water allocation and the ideal water allocation, which cannot realize the optimal utilization of water resources, causing waste of water resources and aggravation of supply and demand contradiction, and seriously restricting the sustainable development of water supply economy.

[0005] The information disclosed in this background section is only intended to understand the background of the inventive concept, so it can include information that does not constitute prior art. SUMMARY

[0006] In view of the above problems or one of the above problems, the purpose of the present application is to provide a water quantity optimization allocation method and device based on topological correlation, which can capture the coupling relationship between water units and the space-time correlation, and dynamically adjust the allocation strategy to obtain an optimal water allocation scheme for the unconstrained scenario, and can convert the water allocation problem into a multivariable optimization problem to obtain an optimal water allocation scheme for the constrained scenario, so that the coupling relationship between water units and the space-time correlation can be fully considered, and thus the optimal allocation of water use in each unit can be realized.

[0007] In order to solve the above problems or one of the above problems, the second object of the present application is to provide a water quantity optimization configuration method and device based on topological correlation, in the scenario of insufficient water quantity, the internal law of water quantity transmission and distribution among different units is fully considered, the space-time characteristics of water use mode can be accurately identified, the gap between actual water distribution quantity and ideal water distribution quantity is minimized, thereby effectively realizing the optimal utilization of water resources and avoiding the waste of water resources.

[0008] In order to achieve one of the above objects, the first technical solution of the present application is:

[0009] A water quantity optimization configuration method based on topological correlation, comprising the following steps:

[0010] Step one, using a pre-constructed topological simulation module, collecting and processing the water inflow and water demand data of a plurality of water use units, establishing the topological relationship of the water use units, determining the water distribution sequence and performing simulation calculation, and obtaining the basic water distribution information of each water use unit;

[0011] Step two, based on the basic water distribution information, and considering the water distribution rules, performing first water distribution calculation on the water use units according to a pre-constructed first water distribution calculation module, and obtaining a first water distribution scheme;

[0012] According to the first water distribution scheme, the total water demand and the total water inflow are calculated;

[0013] When the total water demand is greater than the total water inflow, step three is executed;

[0014] When the total water demand is less than or equal to the total water inflow, the first water distribution scheme is taken as the optimal water distribution scheme, and step four is executed;

[0015] Step three, using a pre-constructed optimization decision module, for the unconstrained scenario, based on the first water distribution scheme, capturing the coupling relationship and space-time correlation between the water use units, and dynamically adjusting the distribution strategy to obtain an optimal water distribution scheme; for the constrained scenario, converting the water distribution problem into a multivariable optimization problem to obtain an optimal water distribution scheme;

[0016] Step four, outputting the optimal water distribution scheme.

[0017] The present application constructs a topological simulation module, a first water distribution calculation module and an optimization decision module, for the unconstrained scenario, captures the coupling relationship and space-time correlation between the water use units, and dynamically adjusts the distribution strategy to obtain an optimal water distribution scheme; for the constrained scenario, converts the water distribution problem into a multivariable optimization problem to obtain an optimal water distribution scheme, so that the coupling relationship and space-time correlation between the water use units can be fully considered, and thus the optimal distribution of water use of each unit can be realized, the scheme is scientific, reasonable and feasible.

[0018] Further, in the scenario of insufficient water inflow, the present application fully considers the internal law of water transfer and distribution among different units, can accurately identify the spatial and temporal characteristics of water use mode, maximally reduce the gap between actual water distribution and ideal water distribution, thereby effectively realize the optimal utilization of water resources, avoid the waste of water resources, reduce the contradiction between supply and demand, and make the water supply economy sustainable development.

[0019] Still further, according to the scenario difference, different algorithm strategies are selected, for the unconstrained scenario, based on the coupling relationship and the time and space correlation between the water use units, the optimal distribution scheme is obtained, thereby the data processing amount can be effectively reduced, the simulation efficiency is improved, and the water distribution efficiency and accuracy are improved; for the constrained scenario, the water distribution problem is converted into a multivariable optimization problem, thereby the water distribution accuracy can be effectively improved; further, the two distribution algorithms of the present application form a complementary strategy, which can meet various water distribution scenarios, and is especially suitable for super large water distribution system considering multiple scenarios.

[0020] Still further, the unconstrained scenario in the present application is actually a simple constraint scenario, which is not unconstrained, but only constrained by simple constraints, such as the scenario still needs to meet the continuous flow constraint.

[0021] As a preferred technical measure:

[0022] Step one, using the pre-constructed topological simulation module, collecting and processing the water inflow and water demand data of a plurality of water use units, establishing the topological relationship of the water use units, determining the water distribution sequence and performing simulation calculation, the method for obtaining the basic water distribution information of each water use unit is as follows:

[0023] Collecting water inflow and water demand data of a plurality of water use units, water inflow prediction data of each interval basin, and discharge data of each reservoir;

[0024] Reading the structure information of the water use unit, combining the water inflow and water demand data, and constructing the water use unit data set;

[0025] According to the interval basin name and proportion, combining the water inflow prediction data, calculating the water inflow of each water use unit based on rainfall runoff;

[0026] According to the upstream reservoir information corresponding to the water use unit, combining the discharge data of the reservoir, calculating the total discharge of the upstream reservoir of each water use unit;

[0027] Based on the upstream and downstream water demand point relationship of the water use unit, constructing the downstream connection relationship data of each water use unit;

[0028] Updating the in-degree according to the upstream water demand point relationship of the water use unit, obtaining the topological sorting sequence of the water use unit;

[0029] Based on the topological sorting order, downstream connection data, total upstream reservoir discharge, inflow based on rainfall runoff, and data set of water-using units, water allocation simulation is performed for each water-using unit in sequence to generate basic water allocation information for each water-using unit.

[0030] As a preferred technical measure:

[0031] The methods for collecting water inflow and demand data for water-using units are as follows:

[0032] Based on the type of water-using unit, clarify the specific needs of each water-using unit and determine the water demand of each water-using unit;

[0033] Or / and, the reservoir's discharge data is obtained through the reservoir's scheduling rules, while the inflow forecast data is obtained through the inflow forecast model or flow meter;

[0034] Alternatively / and, the method for updating the in-degree based on the upstream water demand points of the water-using units to obtain the topological sorting order of the water-using units is as follows:

[0035] Initialize the in-degree of each water-using unit, where the in-degree represents the number of water-using units connected upstream of a given water-using unit;

[0036] Update the in-degree based on the upstream water demand points of the water-using unit, add the water-using unit with an in-degree of 0 to the queue, take the water-using units out of the queue in turn, add them to the topology sorting result set, and reduce the in-degree of its downstream water-using units.

[0037] When the in-degree of a downstream water-using unit becomes 0, it is added to the queue until the queue is empty, thus obtaining the topological sorting order of the water-using units.

[0038] As a preferred technical measure:

[0039] The method for simulating water distribution for each water-using unit and generating basic water distribution information for each water-using unit is as follows:

[0040] Based on the topological sorting order, downstream connection relationship data, total upstream reservoir discharge, inflow based on rainfall runoff, and data set of water use units, the total inflow of each water use unit is calculated.

[0041] Based on the set target water allocation, the smaller value between the target water allocation and the total incoming water volume is taken as the actual water allocation; the remaining water volume is obtained by subtracting the actual water allocation from the total incoming water volume.

[0042] The regional rainfall inflow, reservoir discharge, upstream remaining water volume, total inflow, actual water distribution, theoretical water distribution, and remaining water volume for each water-using unit are recorded in the simulation results set;

[0043] The simulation results set is output as the basic water distribution information for the water-using unit.

[0044] As a preferred technical measure:

[0045] Step two: Based on the pre-built initial water allocation calculation module, and taking into account the basic water allocation information and water allocation rules, the initial water allocation calculation is performed on the water-using units to obtain the initial water allocation scheme. The method is as follows:

[0046] Based on the regional water resources planning scheme, water use policies, and the importance of water use for each water use unit, allocation rules shall be formulated;

[0047] Based on the allocation rules, the water allocation priority information of each water-using unit is determined, including the order of multi-stage allocation, allocation method and corresponding weight;

[0048] Based on the water allocation priority information of each water-using unit, a priority configuration set is constructed;

[0049] Based on the priority configuration set and the water demand data of each water-using unit, and combined with the water inflow of each node, multi-stage water distribution calculation is performed.

[0050] Multi-stage water distribution calculation includes a first calculation stage and a second calculation stage;

[0051] In the first calculation stage, based on the water use type of the water use unit, the water inflow of each node is allocated and calculated to obtain the allocated water volume for each type, thus completing the allocation between types.

[0052] In the second calculation stage, based on the water demand of the water-using unit, the water allocation for each type is calculated and allocated to the specific water-using unit according to the proportion of water demand for that type of water-using unit, so as to obtain the allocated water volume of each water-using unit under each type and complete the allocation between water-using units.

[0053] Finally, the allocation results of each stage are integrated, and the water volume allocated to each type of water user unit is accumulated to generate the initial water allocation plan, which includes the water allocation volume of each water user unit and the total water allocation volume.

[0054] As a preferred technical measure:

[0055] In the first calculation stage, based on the water usage type of the water-using unit, the water inflow to each node is allocated and calculated. The method for obtaining the allocated water volume for each type is as follows, which includes the following:

[0056] Step 21: Based on the water use type of the water use unit, set several allocation weights and normalize each allocation weight to obtain the weight coefficient; calculate the initial water allocation for each water use type based on the weight coefficients.

[0057] Step 22: Based on the water demand of water-using units of the same water type, calculate the water demand of a certain water type, and compare it with the initial allocated water volume to obtain the water type that exceeds the water demand and the over-allocated water volume in the initial allocation, as well as the water type that does not meet the water demand and the water shortage volume.

[0058] For water usage types that exceed demand, the excess water will be recycled and redistributed, resulting in the total recycled water volume:

[0059] Step 23: Redistribute the total recovered water volume according to the weight ratio of the types of water demand that are not met, to obtain the redistributed water volume; and add the redistributed water volume to the initial allocated water volume to obtain the new initial allocated water volume;

[0060] Step 24: Iterative adjustment, repeat steps 22-23 until the remaining water volume is 0 or all types of demand are met, determine the final water volume allocated to each type, the water volume allocated to each type is the sum of the initial water volume allocated in step 21 and the redistributed water volume of several iterations.

[0061] As a preferred technical measure:

[0062] Step 3: Using a pre-built optimization decision module, for unconstrained scenarios, based on the initial water allocation scheme, the spatiotemporal correlation of water-using units is captured, and the allocation strategy is dynamically adjusted to obtain the optimized water allocation scheme as follows:

[0063] Based on the initial water distribution plan, obtain the water distribution volume for each water-using unit;

[0064] Based on the water distribution of the water-using unit, a covariance matrix is ​​constructed, and the mean vector and step size parameters are set.

[0065] The covariance matrix is ​​used to characterize the correlation between water-using units. The off-diagonal elements of the covariance matrix are used to dynamically capture the coupling relationship between water-using units, and the allocation strategy is dynamically adjusted based on the coupling relationship.

[0066] The mean vector represents the ideal water distribution volume;

[0067] The step size parameter is used to adaptively adjust the search range to balance the relationship between global exploration and local development;

[0068] The objective function is the sum of squares of the differences between the actual water distribution and the ideal water distribution.

[0069] Based on the covariance matrix, mean vector, step size parameter, and objective function, repeated iterative calculations are performed until the convergence condition is met, and finally, an optimized water distribution scheme is output.

[0070] As a preferred technical measure:

[0071] Based on the covariance matrix, mean vector, step size parameter, and objective function, repeated iterative calculations are performed until the convergence condition is met, and the final optimized water allocation scheme is output as follows:

[0072] Step 31: Construct a multivariate normal distribution based on the covariance matrix and mean vector; sample from the multivariate normal distribution to generate several candidate solutions, and calculate the objective function value of each candidate solution;

[0073] Step 32: Sort several objective function values ​​in ascending order to obtain a candidate solution array; and select the top μ candidate solutions from the candidate solution array as parents, and assign weights to these parents.

[0074] Step 33: Calculate the new mean vector, step size, and covariance matrix in sequence based on the parent and its weights.

[0075] Step 34: Construct a new multivariate normal distribution based on the new covariance matrix and mean vector; sample from the new multivariate normal distribution to generate several new candidate solutions; calculate the new objective function value based on the new candidate solutions.

[0076] Then calculate the difference between the new objective function value and the previous objective function value, and determine whether the convergence condition is met based on the difference;

[0077] If the convergence condition is not met, execute steps 32 to 34 until the maximum number of iterations is reached.

[0078] When the convergence condition is met, the candidate solution at this point is taken as the final output optimized water distribution scheme.

[0079] Furthermore, when the difference is less than 0.01, it is generally considered that the termination condition is met.

[0080] Furthermore, in step 33, the method for calculating the new mean vector, step size, and covariance matrix sequentially based on the parent and its weights is as follows:

[0081] Step 331: Based on the mean vector, set the mean learning rate and calculate the new mean vector according to the parent and its weights;

[0082] Step 332: Based on the new mean vector, the evolutionary path learning rate, and the quality of effective selection, calculate the first evolutionary path to track long-term changes in the mean vector;

[0083] Based on the new mean vector, step-size evolutionary path learning rate, and decomposition matrix, a second evolutionary path is calculated to track the step-size-related evolutionary direction.

[0084] Step 333: Update the step size based on the magnitude of the second evolutionary path, the step size learning rate, the damping coefficient, and the expected magnitude of the standard normal distribution vector to obtain a new step size;

[0085] Step 334: Update the covariance matrix according to the new step size, the first evolution path and the rank update coefficient to obtain a new covariance matrix.

[0086] As a preferred technical measure:

[0087] For constrained scenarios, the water distribution problem is transformed into a multivariate optimization problem, and the method for obtaining the optimal water distribution scheme is as follows:

[0088] S31, a constrained scenario is a scenario with priority constraints and / or water volume threshold constraints.

[0089] A set of water allocation values ​​is randomly generated to form a water use allocation scheme; each water allocation value represents a potential solution to the water allocation problem.

[0090] S32, based on the ideal water distribution value, evaluates each water allocation value, calculates its fitness, and obtains water consumption deviation information;

[0091] S33, based on the water consumption deviation information, select the water allocation value with the smaller deviation and discard the water allocation value with the larger deviation; and put the selected water allocation value into the matching library;

[0092] Then, a portion of the water allocation values ​​in the pairing library is used as the parent value, and these are replaced and recombined to generate new water allocation values.

[0093] At the same time, some water allocation values ​​in the pairing pool are randomly changed with a small probability of variation to obtain new water allocation values, thus maintaining the diversity of water use configuration schemes;

[0094] S34, calculate the fitness of the new water allocation value to obtain new water consumption deviation information;

[0095] Calculate the difference between the new water consumption deviation information and the previous water consumption deviation information, and determine whether the termination condition is met based on the difference.

[0096] If the termination condition is not met, execute S32 to S34 until the termination condition is met or the maximum number of iterations is reached.

[0097] When the termination condition is met, the several water allocation values ​​at this time are used as the optimized water distribution scheme.

[0098] Furthermore, when the difference is less than 0.01, the termination condition is generally considered met, and the solution at this point is already the optimal solution. The priority constraint sets the user's first priority water consumption; the water consumption threshold constraint is the water inflow to the area, determined based on the inflow.

[0099] To achieve one of the above objectives, the second technical solution of the present invention is as follows:

[0100] A water volume optimization configuration device based on topological correlation, comprising:

[0101] One or more processing units;

[0102] Storage device for storing one or more programs;

[0103] When the one or more programs are executed by the one or more processing units, the one or more processing units implement the above-described method for optimizing water allocation based on topological correlation.

[0104] Compared with existing technical solutions, the present invention has the following beneficial effects:

[0105] This invention constructs a topology simulation module, an initial water allocation calculation module, and an optimization decision module. For unconstrained scenarios, it captures the coupling relationships and spatiotemporal correlations between water-using units and dynamically adjusts the allocation strategy to obtain an optimized water allocation scheme. For constrained scenarios, it transforms the water allocation problem into a multivariate optimization problem to obtain an optimized water allocation scheme. This fully considers the coupling relationships and spatiotemporal correlations between water-using units, thus achieving the optimal allocation of water use for each unit. The scheme is scientific, reasonable, and feasible.

[0106] Furthermore, in scenarios with insufficient water supply, this invention fully considers the inherent laws of water transfer and distribution between different units, accurately identifies the spatiotemporal characteristics of water use patterns, minimizes the gap between actual and ideal water distribution, thereby effectively achieving optimal utilization of water resources, avoiding water waste, reducing supply and demand contradictions, and enabling sustainable development of the water supply economy.

[0107] Furthermore, this invention selects different algorithm strategies based on different scenarios. For unconstrained scenarios, it obtains an optimized allocation scheme based on the coupling relationship and spatiotemporal correlation between water-using units, thereby effectively reducing the amount of data processing, improving simulation efficiency, and increasing the efficiency and accuracy of water allocation. For constrained scenarios, it transforms the water allocation problem into a multivariate optimization problem, thereby effectively improving the accuracy of water allocation. In this way, the two allocation algorithms of this invention form a complementary strategy, which can meet various water allocation scenarios, and is especially suitable for ultra-large water distribution systems that need to consider multiple scenarios. Attached Figure Description

[0108] Figure 1 This is a schematic flowchart of the water quantity optimization allocation method of the present invention;

[0109] Figure 2 This is a schematic diagram of the data flow of the water volume optimization allocation method of the present invention. Detailed Implementation

[0110] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application. This invention covers any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of the invention as defined by the claims.

[0111] like Figure 1 As shown, this is the first specific embodiment of the water quantity optimization allocation method based on topological correlation of the present invention:

[0112] A water allocation optimization method based on topological correlation includes the following steps:

[0113] Step 1: Using a pre-built topology simulation module, collect and process the water inflow and demand data of several water-using units, establish the topology relationship of the water-using units, determine the water distribution sequence, and perform simulation calculations to obtain the basic water distribution information of each water-using unit.

[0114] Step 2: Based on the pre-built initial water allocation calculation module, and taking into account the basic water allocation information and water allocation rules, perform the initial water allocation calculation for the water-using units to obtain the initial water allocation scheme.

[0115] Calculate the total water demand and total water inflow based on the initial water allocation plan;

[0116] When the total water demand exceeds the total water supply, proceed to step three;

[0117] When the total water demand is less than or equal to the total water inflow, the initial water allocation plan will be used as the optimized water allocation plan, and step four will be executed.

[0118] Step 3: Using a pre-built optimization decision module, for unconstrained scenarios, based on the initial water allocation scheme, the coupling relationship and spatiotemporal correlation between water-using units are captured, and the allocation strategy is dynamically adjusted to obtain an optimized water allocation scheme; for constrained scenarios, the water allocation problem is transformed into a multivariate optimization problem to obtain an optimized water allocation scheme.

[0119] Step 4: Output the optimized water distribution plan.

[0120] A second specific embodiment of the water quantity optimization allocation method based on topological correlation of the present invention:

[0121] A water allocation optimization method based on topological correlation includes the following steps:

[0122] Step 1: Through a pre-built data processing module, collect and process water inflow and demand data from several water-using units to generate standardized water distribution data;

[0123] Step 2: Using a pre-built topology simulation module, based on standardized water distribution data, establish the topology relationship of water-using units, determine the water distribution sequence, and perform simulation calculations to obtain the basic water distribution information of each water-using unit, including the inflow, actual distribution, and remaining water volume of each water-using unit.

[0124] Step 3: Based on the pre-built initial water allocation calculation module, and taking into account the basic water allocation information and water allocation rules, perform the initial water allocation calculation for the water-using units to obtain the initial water allocation scheme.

[0125] Calculate the total water consumption and total water inflow based on the initial water allocation plan;

[0126] When the total water consumption exceeds the total water inflow, proceed to step four;

[0127] When the total water consumption is less than or equal to the total water inflow, the initial water allocation plan will be used as the optimized water allocation plan, and step five will be executed.

[0128] Step 4: Using the pre-built optimization decision module, for unconstrained scenarios, based on the initial water allocation scheme, the spatiotemporal correlation of water-using units is captured, and the allocation strategy is dynamically adjusted to obtain an optimized water allocation scheme; for constrained scenarios, the water allocation problem is transformed into a multivariate optimization problem to obtain an optimized water allocation scheme.

[0129] Step 5: Output the optimized water distribution plan.

[0130] like Figure 2 As shown, this is the third specific embodiment of the water quantity optimization allocation method based on topological correlation of the present invention:

[0131] A water allocation optimization method based on topological correlation includes a data processing module, a topology simulation module, an initial water allocation calculation module, an optimization decision module, and a result output module.

[0132] The data processing module is used to process various types of data required by the model. Short-term water inflow mainly considers reservoir discharge and rainfall runoff. Reservoir discharge flow is obtained through reservoir scheduling rules, while rainfall inflow is obtained through inflow forecasting models or flow meters. Water demand is calculated based on the type of water use unit (agriculture, industry, domestic, ecological) using a water demand model to determine the specific needs of each water use unit. Users formulate allocation rules in medium- and long-term water distribution plans based on factors such as regional water resource planning, water use policies, and the importance of water use in various industries.

[0133] The topology simulation module is used to process and calculate the discharge data of each reservoir, the predicted inflow data of each interval, the water distribution rules, and the weight and priority data of each unit obtained by the data processing module. Specifically:

[0134] First, read the structural information of water use units, and parse the information such as the number, name, water demand point, upstream water demand point, upstream reservoir, name and proportion of the watershed in each water use unit to construct a data set of water use units.

[0135] Based on the names and proportions of the watersheds in the intervals, and combined with the inflow data, the inflow volume of each water-using unit based on rainfall runoff is calculated; based on the upstream reservoir information corresponding to the water-using unit, and combined with the reservoir discharge data, the total discharge volume of the upstream reservoir for each water-using unit is calculated; based on the upstream and downstream water demand points of the water-using units, the downstream connection relationship data of each water-using unit is constructed.

[0136] Then, the in-degree of each water-using unit is initialized. The in-degree represents the number of water-using units connected upstream. The in-degree is updated according to the upstream water demand point relationship of the water-using unit. Water-using units with an in-degree of 0 are added to the queue. Water-using units are taken out from the queue in turn and added to the topology sorting result set. The in-degree of their downstream water-using units is reduced. When the in-degree of a downstream water-using unit becomes 0, it is added to the queue until the queue is empty. The topology sorting order of the water-using units is obtained, which determines the execution order for subsequent water allocation calculations.

[0137] Finally, based on the topological sorting order, water allocation simulations are performed for each water-using unit sequentially, initializing the sets of available water volume, reservoir water volume, remaining water volume records, and simulation result sets for each water-using unit. The total water inflow for each water-using unit is then calculated. It is the amount of available water. Water storage volume With the remaining water volume upstream The sum is:

[0138]

[0139] According to the pre-set target water distribution volume Take the target water allocation volume With total water volume The smaller value in the middle is used as the actual water distribution volume. ,Right now:

[0140]

[0141] The remaining water volume is obtained by subtracting the actual water allocation from the total inflow. Information such as the region, rainfall inflow, reservoir discharge, upstream remaining water volume, total inflow, actual water allocation, theoretical water allocation, and remaining water volume for each water-using unit are recorded in the simulation results set. The final output is a simulation result containing detailed water allocation information for each water-using unit, enabling simulation calculations of water resource allocation and providing a basis for the rational allocation of water resources.

[0142] The initial water allocation calculation module is used to summarize the total water demand of the entire basin according to the water demand data of different water use needs of each water use unit obtained by the data acquisition module, based on the unit and water use type. Combined with the water inflow of each node obtained by the data processing and calculation module, and in accordance with the allocation rules in the medium and long-term water allocation plan, while considering the weight and priority of each unit, the module performs the initial water allocation calculation for the water inflow, as follows:

[0143] First, water demand data for different water use needs of each water-using unit is read, and a water demand data set indexed by date and region is constructed. For a specific date, the water demand of each unit is summarized according to water use type such as domestic, agricultural, industrial, and ecological, forming the total water demand data for each type on that day. At the same time, water allocation priority information in the configuration file is read, including the order of multi-stage allocation, allocation method (such as by demand ratio or weight) and corresponding weights, and a priority configuration set is constructed. Then, multi-stage water allocation is executed based on the priority configuration set.

[0144] In this embodiment, the method of allocating according to demand proportions includes the following:

[0145] If allocated according to demand ratio, the total water inflow will be calculated based on the proportion of water demand for each type. The water volume is allocated to the corresponding type, i.e., the water volume allocated to a certain type k. for:

[0146]

[0147] in: Allocate water volume for type k; Total available water volume; Let be the water demand for type k; n be the total number of water use types.

[0148] When the total water volume is sufficient At that time, the water allocation for each type is equal to its water demand:

[0149]

[0150] In this embodiment, the method of weighted allocation includes the following:

[0151] If water is allocated according to weighted proportions, when the allocated water volume for a certain type exceeds its demand, the excess will be cyclically redistributed to the unmet needs of other types until the remaining water volume falls below a threshold or all demands are met. The specific steps are as follows:

[0152] Step 1: Assign initial weights, the calculation formula is as follows:

[0153]

[0154]

[0155] in: Allocate the initial amount of water for type k; Assign weights to type k. These are the weighting coefficients for weight normalization, converting the original weights into proportions.

[0156] Step 2: Allocate the excess portion. For types where the initial allocation exceeds the water demand, the excess portion will be recovered and redistributed. The calculation formula is as follows:

[0157]

[0158]

[0159] Where: R is the total amount of water recovered; S is the set of types of excess allocation.

[0160] Step 3: Redistribute the recovered water volume R according to the weight ratio of the unmet demand type to obtain the redistributed water volume. Its expression is as follows:

[0161]

[0162] Step 4: Iteratively adjust and repeat steps 2-3 until the remaining water volume is 0 or all types of demand are met. Final water allocation. for:

[0163]

[0164] Where: m is the number of iterations; Let be the amount of water redistributed in the j-th iteration.

[0165] After completing the allocation between water types, the water volume allocated to each type is distributed to specific water-using units according to the proportion of water demand for that type, thus obtaining the allocated water volume for each unit under each type. Finally, the allocation results of each stage are integrated, and the allocated water volumes of each unit for each type are accumulated to generate an initial water allocation plan that includes the water allocation for each type of each water-using unit and the total water allocation. This provides an initial water allocation basis for subsequent optimization decisions, realizing the initial water allocation calculation based on water allocation rules and unit priorities, and initially determining the water allocation for each unit.

[0166] The optimization decision module is used to determine whether the total inflow calculated by the initial water distribution module is greater than or equal to the total water demand. When the inflow is insufficient, the optimization algorithm is activated to optimize the system by minimizing the difference between the actual water distribution and the ideal water distribution for each unit, as detailed below:

[0167] First, based on the initial water allocation scheme obtained from the initial water allocation calculation, and combined with the total inflow data, the water demand data file and configuration file are read to construct a water demand data set and a priority configuration set for each water user unit. Constraints are calculated according to the priority configuration, such as the first priority water consumption being less than or equal to the total inflow, generating constraint values ​​for each water user unit. After converting the initial water allocation scheme into vector form, an appropriate optimization algorithm is selected based on the presence or absence of constraints.

[0168] For unconstrained optimization scenarios, the Covariance Matrix Adaptive Evolutionary Strategy (CMAES) algorithm is adopted. This algorithm can efficiently handle continuous variable optimization problems by dynamically adjusting the covariance matrix of the search distribution. It has strong global search capability and convergence speed for complex objective functions under unconstrained conditions.

[0169] The Covariance Matrix Adaptive Evolutionary Strategy (CMAES) is used for unconstrained continuous variable optimization, with the objective function as the input. Initial search point (n is the dimension of the variable), initial step size Population size (usually taken) The core of this approach involves maintaining the mean vector and other parameters. Covariance matrix and step length To dynamically adjust the search distribution, where Initially, it is the identity matrix. , Initially .

[0170] The algorithm first starts with a mean of m and a covariance of m. Sample generation from multivariate normal distribution candidate solutions Its expression is as follows:

[0171]

[0172] Next, the objective function value of each candidate solution is calculated. Sort in ascending order (assuming it's a minimization problem), and select the top... One optimal solution ( ,generally As the parent generation, weights are assigned to these parents. Its expression is as follows:

[0173]

[0174]

[0175] Where i is the sort index of the parent solution. Then the mean vector is updated. This yields a new mean vector. The formula is:

[0176]

[0177] in It is the mean learning rate. Index for iteration count, Indicates the first The generation in the nth iteration There are several candidate solutions. For the covariance matrix... The update requires first calculating the evolutionary paths at different iteration stages. and ,in The formula used to track long-term changes in the mean is:

[0178]

[0179]

[0180] here It is the evolutionary path learning rate. It is an effective way to select quality; The formula used to track the evolutionary direction related to step size is:

[0181]

[0182] in It is the step size evolutionary path learning rate. The Cholesky decomposition matrix (satisfying) ), It is the covariance matrix at the t-th iteration. Based on The modulus length, for step length Update to obtain the new step size. The formula is:

[0183]

[0184] in It is the step size learning rate; It is the damping coefficient, which is greater than zero; It is the expected value of the magnitude of the vector of the standard normal distribution. It follows a standard normal distribution. The formula for calculating the damping coefficient is as follows:

[0185]

[0186] expect The calculation formula is as follows:

[0187]

[0188] Where n is the dimension of the decision variable x, Finally, update the covariance matrix C to obtain the new covariance matrix. The calculation formula is as follows:

[0189]

[0190]

[0191] in It is a rank-1 update coefficient. It is rank Update the coefficients. Repeat the above sampling, selection, and update process until the convergence condition (such as step size) is met. (If the value is less than the threshold, the function value change is less than the threshold, or the maximum number of iterations is reached), the final output is the optimal solution obtained through optimization. (Usually the candidate solution with the smallest objective function value in the last generation) and the corresponding objective function value .

[0192] When constraints such as priority and water volume thresholds exist, the Genetic Algorithm (GA) is employed. Based on the characteristics of natural selection and genetic mechanisms, it uses selection, crossover, and mutation operations to heuristically search the solution space. This effectively balances local development with global exploration while satisfying constraints, ensuring that the water allocation scheme approaches the optimal solution within the constraints. The flowchart of the basic principle of the Genetic Algorithm is as follows, including the following steps:

[0193] S1, Initialize the population: Randomly generate a set of individuals (water allocation values), called the population (water allocation scheme). Each individual represents a potential solution to the problem.

[0194] S2, Fitness Evaluation: Evaluate each individual and calculate its fitness (such as minimizing the water consumption deviation). The fitness function can be defined according to the requirements of the problem and is usually an indicator to measure the quality of an individual's problem-solving.

[0195] The expression of the fitness function is as follows:

[0196]

[0197] Where, is the water consumption deviation information, is the actual water allocation volume, is the ideal water allocation value, which amplifies the influence of the deviation through the square difference.

[0198] S3, Perform selection, crossover, and mutation operations, which include the following:

[0199] The selection operation refers to the operation of selecting excellent individuals from the population and eliminating inferior individuals. It is based on the fitness evaluation. The greater the fitness of an individual (i.e., the closer the water allocation scheme is to the ideal value), the greater the possibility of being selected, and the more offspring it will have in the next generation. The selected individuals are placed in the pairing pool. Currently, the commonly used selection methods include the roulette wheel method, the best individual retention method, the expected value method, the ranking selection method, the competition method, and the linear normalization method.

[0200] The crossover operation refers to the operation of replacing and recombining part of the structures of two parent individuals to generate new individuals. The purpose of crossover is to produce new individuals in the next generation. The crossover operation is randomly performed on two individuals selected from the matching pool according to a certain crossover probability, and the crossover position is also random. The crossover probability is generally set relatively large, between 0.6 - 0.9. The crossover operation can be carried out through single-point crossover, multi-point crossover, uniform crossover, etc.

[0201] The mutation operation is to randomly change the values of some genes of individuals in the population with a very small mutation probability Pm. The basic process of the mutation operation is: generate a random number rand between [0,1]. If rand < Pm, then perform the mutation operation. The mutation operation itself is a local random search. Combined with the selection operator and the crossover operator, it can avoid some permanent loss of information caused by the selection and crossover operators, ensure the effectiveness of the genetic algorithm, endow the genetic algorithm with local random search ability, and at the same time enable the genetic algorithm to maintain the diversity of the population to prevent premature convergence. In the mutation operation, the mutation probability should not be set too large. If Pm > 0.5, the genetic algorithm degenerates into a random search. The mutation operation can change the value of one or more genes.

[0202] S4 repeats steps S2 through S3 until a termination condition is met. The termination condition can be reaching the maximum number of iterations, finding a satisfactory solution, etc. Through continuous selection, crossover, and mutation operations, individuals in the population gradually tend towards better solutions. Ultimately, the genetic algorithm converges to one or more superior solutions with high fitness.

[0203] The selection strategy for the two algorithms balances optimization efficiency and constraint compatibility. The objective function of the optimization problem is the actual water distribution volume. With ideal water distribution The sum of squares of the differences is calculated using the following formula:

[0204]

[0205] in, For water distribution vector, Let be the number of water-using units, and the constraint is that the water allocation for each water-using unit must meet priority constraints. The expression for the first priority water consumption constraint is as follows:

[0206]

[0207] in, This is the set of water-using units with the highest priority.

[0208] The optimal water allocation vector is obtained through iterative optimization algorithms and converted into an optimization scheme in dictionary format. Then, the topology simulation module calculates detailed information such as the actual water allocation volume and the water allocation volume of each type for each water-using unit. If no feasible solution is found, the water allocation scheme with the minimum constraint violation is returned. Finally, the optimization results containing the actual water allocation volume of each unit, the water demand of each type, and the remaining water volume are output to ensure the water demand of high-priority units and realize the optimization and adjustment of the water allocation scheme.

[0209] The result output module is used to output the water distribution plan, which includes the ideal / actual water distribution volume of each unit, details of water demand for each type, total water demand, and actual total water distribution volume.

[0210] The short-term water allocation optimization model provided by this invention constructs a topological sorting model for water-using units, strictly following the physical laws of water transfer to avoid downstream flow interruption caused by excessive upstream water extraction. This ensures that water resource allocation within the basin conforms to hydrological logic, improving the rationality and reliability of the allocation scheme. Simultaneously, based on the weights and priorities of water-using units and combined with multi-stage water allocation rules, the water allocation scheme is dynamically adjusted through optimization algorithms when water inflow is insufficient, ensuring that high-priority water needs such as domestic and ecological water use are prioritized, effectively alleviating supply and demand imbalances. Furthermore, by establishing a mathematical model with the objective of minimizing the difference between the actual and ideal water allocation, and combining the Covariance Matrix Adaptive Evolutionary Strategy (CMAES) and the Genetic Algorithm (GA), precise quantitative allocation of water resources is achieved, significantly improving water resource utilization efficiency and reducing waste. Moreover, a modular design is adopted, with each functional module (data processing, simulated water allocation, initial water allocation, optimization decision-making, and result output) working independently yet collaboratively. This supports flexible integration with water resource planning, water use policies, and water demand models in different regions, exhibiting broad applicability and scalability. Furthermore, by integrating water inflow data from multiple sources such as reservoir discharge and rainfall runoff, as well as water demand data from various sectors including agriculture, industry, and domestic use, we can achieve full-process digital management from data collection and simulation calculation to optimization decision-making, providing a scientific and reliable basis for water resource management.

[0211] The fourth specific embodiment of the water quantity optimization allocation method based on topological correlation of the present invention:

[0212] The water allocation optimization method based on topological correlation of this invention is a short-term water allocation optimization model based on upstream and downstream topological relationships and optimization algorithms. The method for achieving scientific allocation of water resources is as follows:

[0213] The data processing module collects and processes inflow and demand data to generate standardized inputs.

[0214] The topology simulation module constructs the topological relationships of water-using units, determines the water distribution sequence, and performs simulation calculations.

[0215] The initial water allocation calculation module performs an initial allocation based on demand ratios or weights, generating an initial water allocation plan.

[0216] The optimization decision-making module activates an intelligent algorithm when water is insufficient, and optimizes the water distribution plan based on constraints.

[0217] The results output module outputs water distribution results in a formatted manner, providing a basis for management decisions.

[0218] In this embodiment, the data processing module processes the data as follows:

[0219] Step 11: Obtain the outflow from each reservoir through reservoir scheduling rules, for example, by reading daily outflow data from the reservoir management system in real time to form a reservoir water release sequence. Process meteorological data (rainfall, evaporation, etc.) using an inflow forecasting model to calculate the rainfall inflow in each watershed, for example, by dividing the watershed into watershed units using a distributed hydrological model and allocating rainfall runoff according to area proportions.

[0220] Step 12: Calculate water demand based on the type of water use unit (agriculture, industry, domestic, and ecology) using the corresponding water demand model: Agriculture: Calculated based on crop type, planting area, and irrigation quota; Industry: Calculated based on output value, water use efficiency coefficient, and production scale; Domestic: Calculated based on population size, per capita water use quota, and water use habits; Ecology: Determined based on planning indicators such as river ecological base flow and wetland water replenishment demand.

[0221] Step 13: Store the water demand by date-region-type index to form a water demand data set. For example, on a certain day, the agricultural water demand of a certain region is X ten thousand cubic meters, and the industrial water demand is Y ten thousand cubic meters.

[0222] Step 14: Users define medium- and long-term water allocation rules through the interactive interface or configuration file, including the priority of each water use type (e.g., domestic water use has the highest priority), allocation method (demand ratio or weight) and corresponding weight parameters (e.g., ecological water use weight is 0.3).

[0223] In this embodiment, the steps for processing the data using the topology simulation module are as follows:

[0224] Step 21: Read the water use unit structure data, including unit number, name, upstream water demand point, associated reservoir and watershed ratio. For example, the upstream of unit A is unit B and reservoir C, and the rainfall inflow of watershed D accounts for 20%.

[0225] Based on the upstream reservoir and water release data associated with each unit, the inflow of water to that unit is calculated cumulatively. Rainfall runoff is allocated according to the proportion of the watershed within the interval; for example, if the total rainfall inflow to the watershed is 1 million cubic meters, and the unit accounts for 20%, then the rainfall inflow is 200,000 cubic meters. A downstream connection table is established for each unit to record the downstream water-using units of each unit, forming a complete watershed water network topology.

[0226] Step 22: Count the number of upstream connected units for each unit as its in-degree. For example, if unit C has 2 upstream units, its in-degree is 2. Add units with an in-degree of 0 (no upstream units) to the queue; sequentially remove units from the queue and add them to the topology sorting result set; decrease the in-degree of all downstream units of that unit. If the in-degree of a downstream unit becomes 0, add it to the queue; repeat until the queue is empty to obtain the water distribution order according to the water flow direction (e.g., upstream farmland → midstream factory → downstream town).

[0227] Step 23, the calculation method for water distribution simulation is as follows:

[0228] Calculate the total water inflow for each unit. The formula is:

[0229]

[0230] Available water volume Including water inflow from the area; water storage capacity The unit's own reservoir's available water supply; the remaining water supply upstream. This refers to the remaining water volume after water is distributed from the upstream unit.

[0231] Based on the target water distribution volume Total water inflow Calculate the actual water distribution volume To ensure that the water distribution volume does not exceed the available water volume, the formula is as follows:

[0232]

[0233] The total incoming water volume is subtracted from the actual distributed water volume to obtain the remaining water volume, which is then transferred to downstream units for subsequent water distribution calculations. Simultaneously, information such as the incoming water volume, distributed water volume, and remaining water volume for each unit is stored to form a set of simulated water distribution results.

[0234] In this embodiment, the data processing procedure based on the initial water distribution calculation module is as follows:

[0235] Step 31 involves summarizing water demand and configuring priorities, which includes the following:

[0236] The total water demand of the entire basin is summarized according to the types of domestic, agricultural, industrial, and ecological needs. For example, if the daily domestic water demand is 1.2 million cubic meters and the agricultural water demand is 2.8 million cubic meters, the total water demand data for each type is generated.

[0237] Read the water allocation priority information in the configuration file, including the multi-stage allocation order (e.g., domestic → ecological → industrial → agricultural), allocation method (demand ratio or weight) and corresponding weight (e.g., industrial water weight 0.2).

[0238] Step 32: Allocate according to demand ratio, including the following:

[0239] When water is allocated proportionally to demand, the allocated water volume for a certain type k for:

[0240]

[0241] in, For type k, water demand This represents the total water demand for the entire basin. In a sufficient water supply scenario: if the total inflow is greater than or equal to the total water demand, then the allocated water volume for each water type equals its water demand. It directly meets all needs.

[0242] Step 33, assign weights, including the following:

[0243] According to the set weight Calculate the initial water allocation Its expression is as follows:

[0244]

[0245] For example, with a domestic water weight of 0.4, the initial allocation is 1.2 million cubic meters when the total water inflow is 3 million cubic meters;

[0246] For types of water allocation exceeding demand, the excess water will be recovered. The calculation formula is as follows:

[0247]

[0248] Where S is the set of excess types.

[0249] The recovered water volume R is redistributed according to the weighted proportions of the unmet demand types, and this process is repeated until the remaining water volume is below the threshold or all demands are met, resulting in the final water allocation. for:

[0250]

[0251] The water allocation is determined by the proportion of water demand for that type in a unit and then allocated to that specific unit. For example, if the industrial type is allocated 1 million cubic meters of water, and unit A has an industrial water demand of 30%, then 300,000 cubic meters will be allocated to that unit.

[0252] In this embodiment, the data processing procedure using the optimization decision module is as follows:

[0253] Step 41, construct the constraints, including the following:

[0254] Ensure that the total water consumption of high priority does not exceed the total water inflow. For example, the total water consumption of the first priority (domestic + ecological) should be less than or equal to the total water inflow. Set a lower limit for the basic water demand of each unit (e.g., the ecological water demand should not be lower than the river base flow) to form constraints on water allocation.

[0255] Step 42, select the optimization algorithm, including the following:

[0256] For unconstrained scenarios: The Covariance Matrix Adaptive Evolutionary Strategy (CMAES) algorithm is adopted. It optimizes the search direction by dynamically adjusting the covariance matrix to minimize the objective function. This approach is suitable for continuous variable optimization without explicit constraints. The expression is as follows:

[0257]

[0258] in, This represents the actual water distribution volume in the i-th iteration. Let be the ideal water distribution volume for the i-th iteration.

[0259] For constrained scenarios: Genetic Algorithm (GA) is used to handle priority and threshold constraints through selection, crossover, and mutation operations, ensuring that the water allocation scheme approaches the optimal solution within the feasible region, such as prioritizing the needs of domestic water use while not exceeding the reservoir's discharge capacity.

[0260] Step 43, optimize the process, including the following:

[0261] Convert the initial water distribution scheme into a water distribution vector (e.g.) The initial solution is used as the initial solution; the optimal water allocation vector is solved iteratively by combining the constraints and the objective function; if no feasible solution is found, the solution with the minimum constraint violation is returned to ensure that the water demand of high priority units is maximized.

[0262] In this embodiment, the result output module can output information such as unit details, statistical information, and remaining water volume.

[0263] The unit details include the ideal water allocation, actual water allocation, and water allocation for each type (domestic / agricultural, etc.) for each water-using unit; the statistical information includes the total water demand of the entire basin, the actual total water allocation, details of water demand for each type, and the satisfaction rate; the surplus water includes the surplus water of each unit and the entire basin, which is used to evaluate the efficiency of water resource utilization.

[0264] Through the specific embodiments described above, this invention realizes a complete water allocation process from data acquisition to optimization decision-making. Each module works independently yet collaboratively, supporting flexible configuration across different regions and ensuring the scientific, efficient, and fair allocation of water resources in the short term, thus possessing significant engineering application value.

[0265] A system embodiment applying the method of the present invention:

[0266] A water volume optimization configuration device based on topological correlation, comprising:

[0267] One or more processing units;

[0268] Storage device for storing one or more programs;

[0269] When the one or more programs are executed by the one or more processing units, the one or more processing units implement the above-described method for optimizing water allocation based on topological correlation.

[0270] The storage device can be internal memory, external memory, cache memory, or other special memory. The processing unit has signal processing capabilities and can be a general-purpose processor, a digital signal processor, an application-specific integrated circuit, an off-the-shelf programmable gate array, or other programmable logic device.

[0271] An embodiment of a device applying the method of the present invention:

[0272] An electronic device is provided with a computer-readable storage medium on which a computer program is stored. When the program is executed by a processing unit, it implements the above-described method for optimizing water allocation based on topological correlation.

[0273] Computer-readable storage media refers to physical carriers capable of storing computer-recognizable data, instructions, or programs. These media must meet the core characteristics of being readable by a computer (i.e., the data exists in the form of electrical, magnetic, or optical signals and can be converted into binary information that a computer can process through appropriate devices). The physical carrier can be a magnetic storage medium, optical storage medium, semiconductor storage medium, or other storage media.

[0274] The module in this application is an object that uses physical or virtual representation to form an objective description of form and structure. The object is not the same as a physical object, and is not limited to physical or virtual. It can be a data processing function, software program, processing mode, usage method, operation mode, workflow, application process, electronic hardware, circuit module, processing system, system imitation or simulation object.

[0275] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features within the scope of the technology disclosed in the present invention; and these modifications or substitutions will not cause the substance of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. Any modifications or equivalent substitutions that do not deviate from the spirit and scope of the present invention should be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A water quantity optimization allocation method based on topological correlation, characterized in that: Includes the following steps: Step 1: Using a pre-built topology simulation module, collect and process the water inflow and demand data of several water-using units, establish the topology relationship of the water-using units, determine the water distribution sequence, and perform simulation calculations to obtain the basic water distribution information of each water-using unit. Step 2: Based on the pre-built initial water allocation calculation module, and taking into account the basic water allocation information and water allocation rules, perform the initial water allocation calculation for the water-using units to obtain the initial water allocation scheme. Calculate the total water demand and total water inflow based on the initial water allocation plan; When the total water demand exceeds the total water supply, proceed to step three; When the total water demand is less than or equal to the total water inflow, the initial water allocation plan will be used as the optimized water allocation plan, and step four will be executed. Step 3: Using a pre-built optimization decision module, for unconstrained scenarios, based on the initial water allocation scheme, the coupling relationship and spatiotemporal correlation between water-using units are captured, and the allocation strategy is dynamically adjusted to obtain an optimized water allocation scheme. The method is as follows: Based on the initial water distribution plan, obtain the water distribution volume for each water-using unit; Based on the water distribution of the water-using unit, a covariance matrix is ​​constructed, and the mean vector and step size parameters are set. The covariance matrix is ​​used to characterize the correlation between water-using units. The off-diagonal elements of the covariance matrix are used to dynamically capture the coupling relationship between water-using units, and the allocation strategy is dynamically adjusted based on the coupling relationship. The mean vector represents the ideal water distribution volume; The step size parameter is used to adaptively adjust the search range to balance the relationship between global exploration and local development; The objective function is the sum of squares of the differences between the actual water distribution and the ideal water distribution. Based on the covariance matrix, mean vector, step size parameter and objective function, repeated iterative calculations are performed until the convergence condition is met, and finally the optimized water distribution scheme is output. For constrained scenarios, the water distribution problem is transformed into a multivariate optimization problem, resulting in an optimized water distribution scheme; Step 4: Output the optimized water distribution plan.

2. The water quantity optimization allocation method based on topological correlation as described in claim 1, characterized in that: Step 1: Using a pre-built topology simulation module, collect and process the inflow and demand data of several water-using units, establish the topological relationship between the water-using units, determine the water distribution sequence, and perform simulation calculations to obtain the basic water distribution information for each water-using unit. The method is as follows: Collect water inflow and demand data from several water-using units, water inflow forecast data for each watershed, and discharge data from each reservoir; Read the structural information of the water-using unit, and combine it with the water inflow and water demand data to construct a data set of the water-using unit; Based on the names and proportions of the watersheds in the intervals, and combined with the predicted water inflow data, the water inflow of each water-using unit based on rainfall runoff is calculated. Based on the upstream reservoir information corresponding to each water user unit, and combined with the reservoir discharge data, calculate the total water discharge from the upstream reservoirs of each water user unit. Based on the upstream and downstream water demand points of each water-using unit, downstream connection data of each water-using unit is constructed. The in-degree is updated based on the relationship between the upstream water demand points of the water-using unit to obtain the topological sorting order of the water-using units; Based on the topological sorting order, downstream connection data, total upstream reservoir discharge, inflow based on rainfall runoff, and data set of water-using units, water allocation simulation is performed for each water-using unit in sequence to generate basic water allocation information for each water-using unit.

3. The water quantity optimization allocation method based on topological correlation as described in claim 2, characterized in that: The methods for collecting water inflow and demand data for water-using units are as follows: Based on the type of water-using unit, clarify the specific needs of each water-using unit and determine the water demand of each water-using unit; Or / and, the reservoir's discharge data is obtained through the reservoir's scheduling rules, while the inflow forecast data is obtained through the inflow forecast model or flow meter; Alternatively / and, the method for updating the in-degree based on the upstream water demand points of the water-using units to obtain the topological sorting order of the water-using units is as follows: Initialize the in-degree of each water-using unit, where the in-degree represents the number of water-using units connected upstream of a given water-using unit; Update the in-degree based on the upstream water demand points of the water-using unit, add the water-using unit with an in-degree of 0 to the queue, take the water-using units out of the queue in turn, add them to the topology sorting result set, and reduce the in-degree of its downstream water-using units. When the in-degree of a downstream water-using unit becomes 0, it is added to the queue until the queue is empty, thus obtaining the topological sorting order of the water-using units.

4. The water quantity optimization allocation method based on topological correlation as described in claim 3, characterized in that: The method for simulating water distribution for each water-using unit and generating basic water distribution information for each water-using unit is as follows: Based on the topological sorting order, downstream connection relationship data, total upstream reservoir discharge, inflow based on rainfall runoff, and data set of water use units, the total inflow of each water use unit is calculated. Based on the set target water allocation, the smaller value between the target water allocation and the total incoming water volume is taken as the actual water allocation; the remaining water volume is obtained by subtracting the actual water allocation from the total incoming water volume. The regional rainfall inflow, reservoir discharge, upstream remaining water volume, total inflow, actual water distribution, theoretical water distribution, and remaining water volume for each water-using unit are recorded in the simulation results set; The simulation results set is output as the basic water distribution information for the water-using unit.

5. The water quantity optimization allocation method based on topological correlation as described in claim 1, characterized in that: Step two: Based on the pre-built initial water allocation calculation module, and taking into account the basic water allocation information and water allocation rules, the initial water allocation calculation is performed on the water-using units to obtain the initial water allocation scheme. The method is as follows: Based on the regional water resources planning scheme, water use policies, and the importance of water use for each water use unit, allocation rules shall be formulated; Based on the allocation rules, the water allocation priority information of each water-using unit is determined, including the order of multi-stage allocation, allocation method and corresponding weight; Based on the water allocation priority information of each water-using unit, a priority configuration set is constructed; Based on the priority configuration set and the water demand data of each water-using unit, and combined with the water inflow of each node, multi-stage water distribution calculation is performed. Multi-stage water distribution calculation includes a first calculation stage and a second calculation stage; In the first calculation stage, based on the water use type of the water use unit, the water inflow of each node is allocated and calculated to obtain the allocated water volume for each type, thus completing the allocation between types. In the second calculation stage, based on the water demand of the water-using unit, the water allocation for each type is calculated and allocated to the specific water-using unit according to the proportion of water demand for that type of water-using unit, so as to obtain the allocated water volume of each water-using unit under each type and complete the allocation between water-using units. Finally, the allocation results of each stage are integrated, and the water volume allocated to each type of water user unit is accumulated to generate the initial water allocation plan, which includes the water volume allocated to each water user unit and the total water volume.

6. The water quantity optimization allocation method based on topological correlation as described in claim 5, characterized in that: In the first calculation stage, based on the water usage type of the water-using unit, the water inflow to each node is allocated and calculated. The method for obtaining the allocated water volume for each type is as follows, which includes the following: Step 21: Based on the water use type of the water use unit, set several allocation weights and normalize each allocation weight to obtain the weight coefficient; calculate the initial water allocation for each water use type based on the weight coefficients. Step 22: Based on the water demand of water-using units of the same water type, calculate the water demand of a certain water type, and compare it with the initial allocated water volume to obtain the water type that exceeds the water demand and the over-allocated water volume in the initial allocation, as well as the water type that does not meet the water demand and the water shortage volume. For water usage types that exceed demand, the excess water will be recycled and redistributed, resulting in the total recycled water volume: Step 23: Redistribute the total recovered water volume according to the weight ratio of the types of water demand that are not met, to obtain the redistributed water volume; and add the redistributed water volume to the initial allocated water volume to obtain the new initial allocated water volume; Step 24: Iterative adjustment, repeat steps 22-23 until the remaining water volume is 0 or all types of demand are met, determine the final water volume allocated to each type, the water volume allocated to each type is the sum of the initial water volume allocated in step 21 and the redistributed water volume of several iterations.

7. The water quantity optimization allocation method based on topological correlation as described in claim 1, characterized in that: Based on the covariance matrix, mean vector, step size parameter, and objective function, repeated iterative calculations are performed until the convergence condition is met. The final output of the optimized water allocation scheme is as follows: Step 31: Construct a multivariate normal distribution based on the covariance matrix and mean vector; sample from the multivariate normal distribution to generate several candidate solutions, and calculate the objective function value of each candidate solution; Step 32: Sort several objective function values ​​in ascending order to obtain a candidate solution array; and select the top μ candidate solutions from the candidate solution array as parents, and assign weights to these parents. Step 33: Calculate the new mean vector, step size, and covariance matrix in sequence based on the parent and its weights. Step 34: Construct a new multivariate normal distribution based on the new covariance matrix and mean vector; sample from the new multivariate normal distribution to generate several new candidate solutions; calculate the new objective function value based on the new candidate solutions. Then calculate the difference between the new objective function value and the previous objective function value, and determine whether the convergence condition is met based on the difference; If the convergence condition is not met, execute steps 32 to 34 until the maximum number of iterations is reached. When the convergence condition is met, the candidate solution at this point is taken as the final output optimized water distribution scheme.

8. The water quantity optimization allocation method based on topological correlation as described in claim 1, characterized in that: For constrained scenarios, the water distribution problem is transformed into a multivariate optimization problem, and the method for obtaining the optimal water distribution scheme is as follows: S31, a constrained scenario is a scenario with priority constraints and / or water volume threshold constraints. A set of water allocation values ​​is randomly generated to form a water use allocation scheme; each water allocation value represents a potential solution to the water allocation problem. S32, based on the ideal water distribution value, evaluates each water allocation value, calculates its fitness, and obtains water consumption deviation information; S33, based on the water consumption deviation information, select the water allocation value with the smaller deviation and discard the water allocation value with the larger deviation; and put the selected water allocation value into the matching library; Then, a portion of the water allocation values ​​in the pairing library is used as the parent value, and these are replaced and recombined to generate new water allocation values. At the same time, some water allocation values ​​in the pairing pool are randomly changed with a small probability of variation to obtain new water allocation values, thus maintaining the diversity of water use configuration schemes; S34, calculate the fitness of the new water allocation value to obtain new water consumption deviation information; Calculate the difference between the new water consumption deviation information and the previous water consumption deviation information, and determine whether the termination condition is met based on the difference. If the termination condition is not met, execute S32 to S34 until the termination condition is met or the maximum number of iterations is reached. When the termination condition is met, the several water allocation values ​​at this time are used as the optimized water distribution scheme.

9. A water quantity optimization configuration device based on topological correlation, characterized in that: It includes: One or more processing units; Storage device for storing one or more programs; When the one or more programs are executed by the one or more processing units, the one or more processing units implement a water quantity optimization configuration method based on topological correlation as described in any one of claims 1-8.

Images