Urban inland river system intelligent resource scheduling management system and method based on dynamic data collaboration

Abstract

The invention relates to the technical field of water resource scheduling, and provides an urban inland river system intelligent resource scheduling management system and method based on dynamic data collaboration, and the method comprises the following steps: S1, collecting river hydrological data, water quality indexes, urban water demands and water transfer facility state data in real time, and carrying out the time-space alignment and abnormal value filtering; s2, a hydrodynamic force-water quality coupling model of multiple water sources is established, a multi-objective optimization model is established in combination with constraint conditions of ecological base flow and an engineering upper limit, and the multiple water sources include but not limited to river water, underground water, reservoir water, reclaimed water, rainwater and desalinated seawater; and S3, carrying out global dynamic optimization on the water transfer proportion and the water transfer facility parameters by adopting a model algorithm, and minimizing the water transfer cost and the ecological influence. By quantifying the maximum flow upper limit and the ecological base flow threshold value of each water source and optimizing the water transfer proportion and facility parameters, the water resource utilization efficiency is improved, and it is guaranteed that the water transfer cost and the ecological influence are balanced in the urban river water dispatching process.

Description

Technical Field

[0001] The present invention relates to the technical field of water resource scheduling, and particularly to an intelligent resource scheduling management system and method for urban inland water systems based on dynamic data collaboration. Background Art

[0002] With the acceleration of the urbanization process, urban inland water systems are facing problems such as intensified contradictions between water resource supply and demand, insufficient guarantee of ecological base flow, and poor coordination of multi-source water scheduling. Traditional scheduling methods rely on manual experience and lack the real-time response ability to dynamic hydrological data, water quality indicators, and multi-source water constraints, resulting in over-exploitation of groundwater, disconnection between reservoir scheduling and water use demand, low utilization rate of non-traditional water sources such as reclaimed water, and difficulty in effectively coping with ecological risks such as water quality deterioration and eutrophication.

[0003] At the same time, the single-source water scheduling model cannot meet the multi-objective requirements of urban complex water systems and lacks the capabilities of multi-source data fusion, dynamic modeling, and real-time optimization, resulting in lagging scheduling decisions, rough facility control, and inability to achieve coordinated management of water volume, water quality, and ecology. Summary of the Invention

[0004] Aiming at the problems existing in the prior art, the purpose of the present invention is to provide an intelligent resource scheduling management system and method for urban inland water systems based on dynamic data collaboration to solve the problems raised in the above background art.

[0005] To achieve the above purpose, the present invention provides an intelligent resource scheduling management method for urban inland water systems based on dynamic data collaboration, including the following steps:

[0006] S1. Real-time collect river hydrological data, water quality indicators, urban water use demand, and water transfer facility status data, and perform spatio-temporal alignment and outlier filtering;

[0007] S2. Establish a hydrodynamic-water quality coupling model for multiple water sources, and establish a multi-objective optimization model in combination with the constraint conditions of ecological base flow and engineering upper limit. The multiple water sources include but are not limited to river water, groundwater, reservoir water, reclaimed water, rainwater, and desalinated seawater;

[0008] S3. Use model algorithms to globally and dynamically optimize the water transfer ratio and water transfer facility parameters to minimize the water transfer cost and ecological impact;

[0009] S4. Automatically adjust the equipment in the water transfer facilities in combination with the optimization results to complete real-time dynamic adjustment and scheduling of multiple water sources, and dynamically allocate the water transfer volume through sectional metering;

[0010] S5. Compare the actual water transfer data with the model prediction values, correct the model parameters based on deviation analysis, optimize the ecological base flow threshold in combination with ecological response data, and perform real-time update and optimization of the water transfer plan.

[0011] Preferably, in the step S1, the dynamic data acquisition and multi-source data fusion include the following steps:

[0012] S11. Collect the urban water demand data, establish a water use prediction model, and predict the total water use and hourly water use for the next day. The formula is: , where in the formula, is the predicted daily total water use, is the predicted hourly water use, is the cumulative water use ratio related to time , is the smoothing coefficient, is the predicted value of the water demand in the previous time period, is the actual water use in the previous time period;

[0013] S12. Real-time collect the river water flow data, and predict the average daily river water flow. The formula is: , where in the formula, is the predicted average daily river water flow, is the number of data participating in the average calculation, is the river water flow data collected at the th moment, is the current moment;

[0014] S13. Monitor the river water quality index data, establish a water quality model, and predict the change of the river water quality during the water transfer process. The formula is: , where in the formula, and are the water quality concentration values at the next moment and the current moment respectively, and are the time step and the space step respectively, and are the fluxes at the spatial positions and respectively, is the generation or consumption of water quality substances;

[0015] S14. Eliminate the abnormal data and update the state estimation. The formula is: , where in the formula, and are the state estimation values at the current moment and the previous moment respectively, , and are the state transition matrix, the control matrix, and the observation matrix respectively, is the Kalman gain, is the observed value, is the updated state estimation, and are the particle weight and particle state respectively, is the control matrix for the input at the moment, is the total number of particles, is the starting count for the summation operation;

[0016] S15. Store the processed data in the blockchain and ensure its immutability through a hash function.

[0017] Preferably, in step S2, the construction of the multi-source collaborative scheduling model includes the following steps:

[0018] S21. Describe the river channel flow based on the diffusion wave equation and construct a hydrodynamic model, the formula of which is: , where in the formula, is the river water flow, is the unit time, is the cross-sectional average flow velocity, is the longitudinal coordinate along the river flow direction, is the lateral inflow discharge, is the cross-sectional area of the water passage, is the partial derivative symbol;

[0019] S22. Set multiple objective functions and calculate the weights using the entropy weight method, the formula of which is: , where in the formula, is the minimum value of the objective function, is the entropy value of the th sub-objective function;

[0020] S23. Define the constraint conditions of the ecological base flow constraint and the engineering upper limit constraint, and use hierarchical clustering to divide the effective area and the non-effective area, the formula of which is: , where in the formula, is the Euclidean distance between reservoir and reservoir , and are the characteristic parameters of reservoir and reservoir respectively;

[0021] S24. Construct a hybrid scheduling model and use particle swarm optimization for scheduling, the formula of which is: , where in the formula, and are the particle velocities at the current moment and the previous moment respectively, is the inertia weight, and are the self-learning factor and the social learning factor respectively, and are random numbers between 0 and 1 respectively, and are respectively the optimal position experienced by the -th particle itself and the optimal position experienced by the entire particle swarm, is the -th particle's position at the

[0022] Preferably, in the step S3, the global optimization of the water transfer parameters includes the following steps:

[0023] S31. Based on the water transfer ratio vector and the water transfer facility parameters, deduce the water transfer cost and ecological impact. The formula is: , where in the formula, is the water transfer cost, is the unit water transfer cost of the -th type of water source, is the number of water source types, is the water transfer ratio of the -th type of water source, is the total water transfer demand, is the number of water transfer facilities, is the unit energy consumption cost of the -th facility, is the operation parameter of the -th facility, is the negative impact value of water transfer on the ecological environment, and are ecological weights, and , and are respectively the remaining flow of the river and the ecological base flow of the river after water transfer, and are respectively the measured water quality concentration and the standard water quality concentration;

[0024] S32. Add weights to the water transfer cost and ecological impact, and establish an objective function for minimizing the cost and impact. The formula is: , where in the formula, is the deduced value for minimizing the cost and impact, and are the corresponding weights, is the deviation term;

[0025] S33. Use the roulette wheel selection method to select operations. Calculate the selection probability based on the chromosome fitness value, and find the optimal combination among multiple objective functions for minimizing the cost and impact. The formula is: , where in the formula, is the selection probability of chromosome , is the fitness value;

[0026] S34. By using single-point crossover and Gaussian mutation operations, gene recombination and random perturbation are performed on the chromosomes to increase population diversity and explore the optimal solution of decision variables in a larger search space. The formula is as follows: , where is the new individual obtained after crossover and mutation operations, and are the two parent individuals participating in the crossover operation, is the crossover operator, is the mutation standard deviation, is the standard normal distribution.

[0027] Preferably, in the step S4, the dynamic allocation of water delivery volume includes pump station power, sluice opening, and valve flow control. The pump station power control uses frequency conversion technology to adjust the motor speed to complete the dynamic adjustment of flow rate. The formula is as follows: , where is the pump station flow rate, is the pump efficiency coefficient, is the operating power of the pump station, is the power-flow index. The sluice opening control is based on the optimized sluice opening, and the flow rate through the weir is calculated using the broad-crested weir formula to drive the gate actuator. The formula is as follows: , where is the flow rate through the sluice, is the flow coefficient, is the width of the sluice, is the acceleration due to gravity, is the head above the weir, is the proportionality coefficient, is the sluice opening, is the intercept coefficient. The valve flow control adjusts the pipe network pressure according to the optimized valve opening using the flow coefficient formula to ensure the water volume in the partition. The formula is as follows: , where is the valve flow rate, is the flow coefficient, is the pressure before and after the valve, is the fluid density, is the coefficient of the inherent characteristics of the valve, is to describe the flow coefficient and the sluice opening The exponent of the non-linear relationship between them.

[0028] Preferably, in the step S12, the real-time collection of river water flow data further includes the comprehensive water quality index, the predicted pollutant concentration, and the assessment of eutrophication degree. The comprehensive water quality index is calculated by weighted averaging of various water quality indicators, comprehensively considering different indicators and their importance levels to obtain the comprehensive water quality index. The formula is , where is the comprehensive water quality index, is the index weight, , and are respectively the measured value, the upper limit of the standard value, and the lower limit of the standard value of the water quality index of the th box. The predicted pollutant concentration is obtained by excavating the relationship between relevant historical influencing factors and pollutant concentration and establishing a regression model. The formula is: , where is the predicted concentration of the pollutant at the th moment, is the constant term in the regression model, is the regression coefficient, is the lag variable. The assessment of eutrophication degree is based on the chlorophyll concentration, and the Carlson index is obtained through specific calculations. The formula is: , where is the Carlson index, is the chlorophyll concentration.

[0029] Preferably, in the step S23, the setting of constraint conditions includes the following steps:

[0030] S231. Determine the minimum ecological flow and calculate the ecological base flow threshold. The formula is: , where is the minimum ecological flow, is the proportionality coefficient, is the average flow;

[0031] S232. Calculate the maximum water transfer volume according to the maximum opening of the sluice and the pump station power, and determine the project water conveyance capacity. The formula is: , where is the maximum water transfer volume, is the flow coefficient, is the maximum opening of the sluice, is the opening of the sluice, is the acceleration due to gravity;

[0032] S233. Quantify the substitution relationship of each water source and establish a water source complementary matrix. The formula is: , where is the complementary matrix, is the substitution coefficient, and .

[0033] Preferably, in the step S13, the monitoring of the river water index data includes the following steps:

[0034] S131. Obtain the cross-sectional flow velocity distribution through an acoustic Doppler current profiler to measure the river water flow velocity distribution;

[0035] S132. Calculate the longitudinal diffusion coefficient of the river water, perform spatial difference on the measured concentration data, and solve the concentration gradient. The formula is: , where in the formula, is the longitudinal diffusion coefficient, is the adjustment coefficient, is the substance concentration, is the spatial coordinate, and are respectively the rd and th measured substance concentrations at the th and th monitoring points, is the gradient concentration; Set the diffusion flux threshold corresponding to the environmental capacity. Calculate the diffusion flux based on the calculation results of the longitudinal diffusion coefficient and the gradient concentration, and evaluate the diffusion impact. The formula is. In the formula, is the actual diffusion flux. When the actual diffusion flux exceeds the threshold, that is, the diffusion impact exceeds the environmental carrying capacity, an emergency dispatch needs to be triggered.

[0036] Preferably, in the step S5, the formula for correcting the model parameters is: , where in the formula, and are respectively the parameter values at the th and th iterations, is the learning rate, is the loss function, and , is the sample number of the actual water transfer data, and are respectively the actual water transfer data and the model prediction value.

[0037] An intelligent resource scheduling management system for urban inland water systems based on dynamic data collaboration is applied to the intelligent resource scheduling management method for urban inland water systems based on dynamic data collaboration described in any one of the above. It is characterized by including:

[0038] A data acquisition and processing module, which is used to collect various data of the river in real time and process the data to provide an accurate data basis for subsequent analysis; A model construction module is used to establish a multi-source water transfer model to provide model support for formulating the water transfer plan.

[0039] Optimization calculation module, used to establish a global dynamic optimization model to minimize the water transfer cost and ecological impact;

[0040] Dispatch execution module, used to intelligently control the water transfer facilities and equipment according to the optimization results to ensure the reasonable allocation of water resources;

[0041] Feedback optimization module, used to compare the actual water transfer data, correct the model parameters based on deviation analysis, and ensure the real-time update and continuous optimization of the water transfer plan.

[0042] The intelligent resource scheduling management system and method for urban inland water systems based on dynamic data collaboration provided by the present invention have the following beneficial effects:

[0043] 1. By collecting real-time hydrological, water quality, water use demand, and facility status data, and combining algorithms such as Kalman filtering and particle swarm optimization to construct a hydrodynamic-water quality coupling model and a multi-objective optimization model, the collaborative scheduling of multiple water sources such as river water, groundwater, and reclaimed water is realized. At the same time, the maximum flow upper limit and ecological base flow threshold of each water source are quantified, the water transfer ratio and facility parameters are optimized, the water resource utilization efficiency is improved, and the water transfer cost and ecological impact are balanced during the scheduling process of urban river water.

[0044] 2. Automatically adjust the pump station power, sluice opening, and valve flow according to the optimization results to achieve dynamic and precise control of the flow; through the "monitoring-modeling-optimization-execution-feedback" closed loop, use the gradient descent algorithm to correct the model parameters, and dynamically update the ecological base flow threshold in combination with ecological response data. At the same time, evaluate the comprehensive water quality index, pollutant diffusion risk, and eutrophication degree in real time, trigger the emergency scheduling mechanism, strengthen ecological security and system resilience, and improve the intelligence and reliability of urban river water resource management. Description of the Drawings

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0047] Figure 1 It is a schematic diagram of the method steps of the intelligent resource scheduling management system and method for urban inland water systems based on dynamic data collaboration provided by this application;

[0048] Figure 2 It is a schematic diagram of the system modules of the intelligent resource scheduling management system and method for urban inland water systems based on dynamic data collaboration provided by this application. Detailed Embodiments

[0050] The following further describes in detail the specific implementation manners of the present invention in conjunction with the accompanying drawings of the specification and embodiments. The following embodiments are only used to illustrate the present invention, but cannot be used to limit the scope of the present invention.

[0051] As Figure 1 - Figure 2 shown, this embodiment proposes an intelligent resource scheduling management method for urban inland water systems based on dynamic data collaboration, including the following steps:

[0052] S1. Real-time collect river hydrological data, water quality indicators, urban water demand, and the status data of water diversion facilities, and perform spatio-temporal alignment and outlier filtering;

[0053] S2. Establish a hydrodynamic-water quality coupling model for multiple water sources, and establish a multi-objective optimization model in combination with the constraints of ecological base flow and engineering upper limits. The multiple water sources include, but are not limited to, river water, groundwater, reservoir water, reclaimed water, rainwater, and seawater desalination water;

[0054] S3. Use model algorithms to globally optimize the water diversion ratio and the parameters of water diversion facilities, and minimize the water diversion cost and ecological impact;

[0055] S4. Automatically adjust the equipment in the water diversion facilities in combination with the optimization results to complete the real-time dynamic regulation and scheduling of multiple water sources, and dynamically allocate the water conveyance volume through sectional metering;

[0056] S5. Compare the actual water diversion data with the model prediction values, correct the model parameters based on deviation analysis, optimize the ecological base flow threshold in combination with the ecological response data, and perform real-time update and optimization of the water diversion plan.

[0057] In this embodiment, in step S1, the dynamic data collection and multi-source data fusion include the following steps:

[0058] S11. Collect the data of urban water demand, establish a water use prediction model, and predict the total daily water use and the hourly water use in the future day. The formula is: , where is the predicted total daily water use, is the predicted hourly water use, is the cumulative water use ratio related to time , is the smoothing coefficient, is the predicted value of the water demand in the previous time period, is the actual water use in the previous time period;

[0059] S12. Real-time collect the river flow data, and predict the average daily river flow. The formula is: , where is the predicted average daily river flow, is the number of data involved in the average calculation, For the The river flow data collected at each moment, for the current moment;

[0060] S13. Monitor river water index data, establish a water quality model, and predict changes in river water quality during water transfer. The formula is: , where and are the water quality concentration values ​​at the next moment and the current moment respectively, and are the time step and space step respectively, and In spatial position and The flux at The generation or consumption of water quality substances;

[0061] S14, remove abnormal data and update the state estimate, the formula is: , where and are the estimated values ​​of the state at the current moment and the previous moment respectively, , and are the state transfer matrix, control matrix and observation matrix respectively, is the Kalman gain, is the observed value, is the updated state estimate, and are particle weight and particle state respectively, for The control matrix that controls the input at all times, is the total number of particles, is the starting count for the sum operation;

[0062] S15. Store the processed data in the blockchain and use a hash function to ensure that it cannot be tampered with.

[0063] In this embodiment, in step S12, the real-time collection of river flow data also includes a comprehensive water quality index, predicted pollutant concentrations, and an assessment of the degree of eutrophication. The comprehensive water quality index is calculated by weighted average calculation of various water quality indicators, comprehensively considering different indicators and their importance, and obtaining a comprehensive water quality index. The formula is: , where is the comprehensive water quality index, is the indicator weight, , as well as Respectively The measured values, upper limits of standard values, and lower limits of standard values of the water quality indicators in the box. The predicted pollutant concentration is obtained by mining the relationship between relevant influencing factors and pollutant concentration in history and establishing a regression model. The formula is: , where is the predicted concentration of the pollutant at time the constant term in the regression model, the regression coefficient, the lag variable. The assessment of the eutrophication degree is based on the chlorophyll concentration, and the Carlson index is obtained through specific calculations. The formula is: , where is the Carlson index, is the chlorophyll concentration.

[0064] In this embodiment, in step S13, the monitoring of the river water index data includes the following steps:

[0065] S131. Obtain the cross-section velocity distribution through an acoustic Doppler current profiler and measure the river water velocity distribution;

[0066] S132. Calculate the longitudinal diffusion coefficient of the river water, perform spatial difference on the measured concentration data, and solve the concentration gradient. The formula is: , where is the longitudinal diffusion coefficient, is the adjustment coefficient, is the substance concentration, is the spatial coordinate, and are respectively the and the measured substance concentrations at the th and th monitoring points, is the distance between two adjacent monitoring points,

[0067] S133. Set the diffusion flux threshold corresponding to the environmental capacity, calculate the diffusion flux based on the calculation results of the longitudinal diffusion coefficient and the gradient concentration, and evaluate the diffusion impact. The formula is: , where is the actual diffusion flux. When the actual diffusion flux exceeds the threshold, that is, the diffusion impact exceeds the environmental carrying capacity, and emergency scheduling needs to be triggered.

[0068] In this embodiment, in step S2, the construction of the multi-source collaborative scheduling model includes the following steps:

[0069] S21. Describe the river channel flow based on the diffusion wave equation and construct a hydrodynamic model. The formula is: , where is the river flow rate, is the unit time, is the cross-sectional average velocity, is the longitudinal coordinate along the river flow direction, is the lateral inflow rate, is the cross-sectional area of flow, is the partial derivative symbol;

[0070] S22. Set multiple objective functions and calculate the weights using the entropy weight method. The formula is: , where is the minimum value of the objective function, is the entropy value of the th objective, is the

[0071] S23. Define the constraint conditions of the ecological base flow constraint and the engineering upper limit constraint, and use hierarchical clustering to divide the effective area and non-effective area. The formula is: , where is the reservoir and the reservoir Euclidean distance, and are the characteristic parameters of the reservoir and the reservoir respectively;

[0072] S24. Build a hybrid scheduling model and use particle swarm optimization for scheduling. The formula is: , where and are the particle velocities at the current moment and the previous moment respectively, is the inertia weight, and are the self-learning factor and the social learning factor respectively, and are random numbers between 0 and 1 respectively, and are the optimal positions experienced by the th particle itself and the optimal position experienced by the entire particle swarm respectively, is the th particle's position at the th generation.

[0073] In this embodiment, in step S23, the setting of the constraint conditions includes the following steps:

[0074] S231. Determine the minimum ecological flow rate and calculate the ecological base flow threshold. The formula is: , where is the minimum ecological flow rate, is the proportionality coefficient, is the average flow rate;

[0075] S232. Calculate the maximum water transfer volume based on the maximum opening of the sluice and the power of the pumping station, and determine the water conveyance capacity of the project. The formula is: , where in the formula, is the maximum water transfer volume, is the flow coefficient, is the maximum opening of the sluice, is the opening of the sluice, is the acceleration due to gravity;

[0076] S233. Quantify the substitution relationship of each water source and establish a water source complementary matrix. The formula is: , where in the formula, is the complementary matrix, is the substitution coefficient, and .

[0077] Specifically, by collecting real-time data on hydrology, water quality, water use demand, and facility status, combined with algorithms such as Kalman filtering and particle swarm optimization, a hydrodynamic-water quality coupling model and a multi-objective optimization model are constructed to achieve the coordinated scheduling of multiple water sources such as river water, groundwater, and reclaimed water. At the same time, the maximum flow rate upper limit and ecological base flow threshold of each water source are quantified, the water transfer ratio and facility parameters are optimized, the water resource utilization efficiency is improved, and the balance between water transfer cost and ecological impact is ensured during the scheduling of river water in the urban area.

[0078] In this embodiment, in step S3, the global optimization of water transfer parameters includes the following steps:

[0079] S31. Deduce the water transfer cost and ecological impact based on the water transfer ratio vector and water transfer facility parameters. The formula is: , where in the formula, is the water transfer cost, is the unit water transfer cost of the th type of water source, is the number of water source types, is the water transfer ratio of the th type of water source, is the total water transfer demand, is the number of water transfer facilities, is the unit energy consumption cost of the th facility, is the operation parameter of the th facility, is the negative impact value of water transfer on the ecological environment, and are ecological weights, and , and They are the remaining flow of the river and the ecological base flow of the river after water transfer respectively. and They are the measured water quality concentration and the standard water quality concentration respectively.

[0080] S32. Add weights to the water transfer cost and ecological impact, and establish an objective function to minimize the cost and impact. The formula is: , where is the deduced value to minimize the cost and impact, and are the corresponding weights, is the deviation term;

[0081] S33. Use the roulette wheel method to select operations. Calculate the selection probability based on the chromosome fitness value, and find the optimal combination among multiple objective functions to minimize the cost and impact. The formula is: , where is the chromosome 's selection probability, is the fitness value;

[0082] S34. Use single-point crossover and Gaussian mutation operations. Recombine genes and perform random perturbations on the chromosomes to increase population diversity and explore the optimal solution of decision variables in a larger search space. The formula is: , where is the new individual obtained after crossover and mutation operations, and are the two parent individuals participating in the crossover operation, is the crossover operator, is the mutation standard deviation, is the standard normal distribution.

[0083] In this embodiment, in step S4, the dynamic allocation of the water transfer volume includes pump station power, sluice opening, and valve flow control. The pump station power control uses frequency conversion technology to adjust the motor speed to complete the dynamic adjustment of the flow rate. The formula is: , where is the pump station flow rate, is the pump efficiency coefficient, is the pump station operating power, is the power-flow index. The sluice opening control is based on the optimized sluice opening. The flow rate through the weir is calculated using the broad-crested weir formula, and the gate actuator is driven. The formula is: , where is the flow rate through the sluice, is the flow coefficient, is the weir width, is the acceleration due to gravity, is the water head above the weir, is the proportionality coefficient. is the opening of the sluice gate, is the intercept coefficient. The valve flow control adjusts the pipe network pressure according to the optimized valve opening and uses the flow coefficient formula to ensure the water volume in the partition. The formula is: , where, is the valve flow rate, is the flow coefficient, is the pressure before and after the valve, is the fluid density, is the coefficient of the inherent characteristics of the valve, is used to describe the flow coefficient and the opening of the sluice gate is the exponent of the non-linear relationship between them.

[0084] In this embodiment, in step S5, the formula for correcting the model parameters is: , where, and are the parameter values at the th and th iterations respectively, is the learning rate, is the loss function, and , is the sample number of the actual water transfer data, and are the actual water transfer data and the model prediction values respectively.

[0085] An intelligent resource scheduling management system for urban inland water systems based on dynamic data collaboration, which is applied to the intelligent resource scheduling management method for urban inland water systems based on dynamic data collaboration in any one of the above, is characterized by including:

[0086] A data acquisition and processing module, which is used to collect various data of the river in real time and process the data to provide an accurate data basis for subsequent analysis;

[0087] A model construction module, which is used to establish a multi-source scheduling model to provide model support for formulating water transfer plans;

[0088] An optimization calculation module, which is used to establish a global dynamic optimization model to minimize water transfer costs and ecological impacts;

[0089] A scheduling execution module, which is used to intelligently control water transfer facilities and equipment according to the optimization results to ensure the reasonable allocation of water resources;

[0090] A feedback optimization module, which is used to compare the actual water transfer data and correct the model parameters based on deviation analysis to ensure that the water transfer plan is updated in real time and continuously optimized.

[0091] Specifically, the pump station power, sluice opening and valve flow are automatically adjusted according to the optimization results to achieve dynamic and precise flow control; through the "monitoring-modeling-optimization-execution-feedback" closed loop, the gradient descent algorithm is used to correct the model parameters, and the ecological base flow threshold is dynamically updated in combination with the ecological response data. At the same time, the comprehensive water quality index, pollutant diffusion risk and eutrophication degree are evaluated in real time, the emergency dispatch mechanism is triggered, the ecological safety and system resilience are strengthened, and the intelligence and reliability of urban river water resources management are improved.

[0092] The above embodiments are only used to illustrate the present invention, rather than to limit the present invention. Although the present invention has been described in detail with reference to the embodiments, those of ordinary skill in the art should understand that various combinations, modifications or equivalent replacements of the technical solutions of the present invention do not depart from the spirit and scope of the technical solutions of the present invention, and should all be covered by the scope of the claims of the present invention.

Claims

1. An intelligent resource scheduling and management method for urban inland water systems based on dynamic data collaboration, characterized in that, It includes the following steps: S1. Real-time collect river hydrological data, water quality indicators, urban water demand, and the status data of water transfer facilities, and perform spatio-temporal alignment and outlier filtering; S2. Establish a hydrodynamic-water quality coupling model for multiple water sources, and combine the constraint conditions of ecological base flow and engineering upper limits to establish a multi-objective optimization model. The multiple water sources include, but are not limited to, river water, groundwater, reservoir water, reclaimed water, rainwater, and desalinated seawater; S3. Use model algorithms to globally and dynamically optimize the water transfer ratio and water transfer facility parameters to minimize the water transfer cost and ecological impact; S4. Automatically adjust the equipment in the water transfer facilities in combination with the optimization results to complete the real-time dynamic regulation and dispatching of multiple water sources, and dynamically allocate the water transfer volume through zonal metering; S5. Compare the actual water transfer data with the model prediction values, correct the model parameters based on deviation analysis, optimize the ecological base flow threshold in combination with the ecological response data, and update and optimize the water transfer plan in real time.

2. The intelligent resource scheduling and management method for urban inland water systems based on dynamic data collaboration according to claim 1, wherein In the step S1, the dynamic data collection and multi-source data fusion include the following steps: S11. Collect the data of urban water demand, establish a water use prediction model, and predict the total water use and hourly water use for the next day. The formula is: , where is the predicted total daily water use, is the predicted hourly water use, is the cumulative water use ratio related to time , is the smoothing coefficient, is the predicted value of water demand in the previous time period, is the actual water use in the previous time period; S12. Real-time collect the river water flow data and predict the average daily river water flow. The formula is as follows: , where is the predicted average daily river water flow, is the number of data participating in the average calculation, is at the th moment, the river water flow data collected, is the current moment; S13. Monitor the river water index data, establish a water quality model, and predict the changes in river water quality during the water transfer process. The formula is: , where and are the water quality concentration values at the next moment and the current moment respectively, and are the time step and the space step respectively, and are the fluxes at the spatial positions and respectively, is the generation or consumption of water quality substances; S14. Reject abnormal data and update the state estimation. The formula is: , where and are the state estimation values at the current moment and the previous moment respectively, , and are the state transition matrix, the control matrix and the observation matrix respectively, is the Kalman gain, is the observed value, is the updated state estimation, and are the particle weight and the particle state respectively, is the control matrix of the control input at time is the total number of particles, is the starting count of the summation operation; S15. Store the processed data in the blockchain and ensure its immutability through a hash function.

3. The intelligent resource scheduling and management method for urban inland water systems based on dynamic data collaboration according to claim 2, wherein In the step S2, the construction of the multi-source collaborative scheduling model includes the following steps: S21. Describe the river channel flow based on the diffusion wave equation and construct a hydrodynamic model. The formula is as follows: , where is the river water flow rate, is the unit time, is the cross-sectional average flow velocity, is the longitudinal coordinate along the river flow direction, is the lateral inflow rate, is the cross-sectional area of the water passage, is the partial derivative symbol; S22. Set multiple objective functions and calculate the weights using the entropy weight method. The formula is: , where is the minimum value of the objective function, is the entropy value of the th objective, is the th sub-objective function; S23. Define the constraint conditions of ecological base flow constraint and engineering upper limit constraint, and use hierarchical clustering to divide the effective area and non-effective area. The formula is: , where is the Euclidean distance between reservoir and reservoir . and are the characteristic parameters of reservoir and reservoir respectively. S24. Build a hybrid scheduling model and adopt particle swarm optimization scheduling. The formula is as follows: , where and are the particle velocities at the current moment and the previous moment respectively, is the inertia weight, and are the self-learning factor and the social learning factor respectively, and are random numbers between 0 and 1 respectively, and are the optimal positions experienced by the -th particle itself and the optimal position experienced by the entire particle swarm respectively, is the position of the -th particle at the -th generation.

4. The intelligent resource scheduling and management method for urban inland water systems based on dynamic data collaboration according to claim 3, wherein In the step S3, the global optimization of water transfer parameters includes the following steps: S31. Deduce the water transfer cost and ecological impact based on the water transfer ratio vector and water transfer facility parameters. The formula is: , where, is the water transfer cost, is the water transfer cost per unit of the type of water source, is the number of water source types, is the water transfer ratio of the type of water source, is the total water transfer demand, is the number of water transfer facilities, is the unit energy consumption cost of the facility, is the operation parameter of the facility, is the negative impact value of water transfer on the ecological environment, and are ecological weights, and , and are the remaining river flow and river ecological base flow after water transfer respectively, and are the measured water quality concentration and the standard water quality concentration respectively; S32. Add weights to the water transfer cost and ecological impact, and establish an objective function to minimize the cost and impact. The formula is: , where is the deduction value for minimizing the cost and impact, and are the corresponding weights, is the deviation term; S33. The roulette wheel selection operation calculates the selection probability based on the chromosome fitness value to find the optimal combination among multiple objective functions that minimize costs and impacts. The formula is as follows: , where is the chromosome 's selection probability, and is the fitness value; S34. By using single-point crossover and Gaussian mutation operations to perform gene recombination and random perturbation on chromosomes, the population diversity is increased to explore the optimal solution of decision variables in a larger search space. The formula is as follows: , where is the new individual obtained after crossover and mutation operations, and are the two parent individuals participating in the crossover operation, is the crossover operator, is the mutation standard deviation, is the standard normal distribution.

5. The intelligent resource scheduling and management method for urban inland water systems based on dynamic data collaboration according to claim 3, wherein In step S4, the dynamic allocation of water delivery volume includes pump station power, sluice opening, and valve flow control. The pump station power control uses frequency conversion technology to adjust the motor speed to complete the dynamic adjustment of flow rate. The formula is: , where is the pump station flow rate, is the pump efficiency coefficient, is the pump station operating power, is the power-flow index. The sluice opening control is based on the optimized sluice opening. The broad-crested weir formula is used to calculate the flow rate through the sluice, and the gate actuator is driven. The formula is: , where is the flow rate through the sluice, is the flow coefficient, is the sluice width, is the acceleration due to gravity, is the head above the weir, is the proportionality coefficient, is the sluice opening, is the intercept coefficient. The valve flow control adjusts the pipe network pressure according to the optimized valve opening using the flow coefficient formula to ensure the water volume in each zone. The formula is: , where is the valve flow rate, is the flow coefficient, is the pressure before and after the valve, is the fluid density, is the coefficient of the inherent characteristics of the valve, is to describe the flow coefficient and the sluice opening The exponent of the nonlinear relationship between them.

6. The intelligent resource scheduling management method for urban inland water systems based on dynamic data collaboration according to claim 2, wherein In the step S12, the real-time collection of river water flow data further includes the comprehensive water quality index, the predicted pollutant concentration, and the assessment of the eutrophication degree. The comprehensive water quality index is calculated by weighted averaging of various water quality indicators, comprehensively considering different indicators and their importance levels to obtain the comprehensive water quality index. The formula is , where is the comprehensive water quality index, is the index weight, , and are respectively the measured value, the upper limit of the standard value, and the lower limit of the standard value of the water quality index of the th box. The predicted pollutant concentration is obtained by excavating the relationship between relevant historical influencing factors and pollutant concentrations and establishing a regression model. The formula is: , where is the predicted concentration of the pollutant at time, is the constant term in the regression model, is the regression coefficient, is the lag variable. The assessment of the eutrophication degree is based on the chlorophyll concentration, and the Carlson index is obtained through specific calculations. The formula is: , where is the Carlson index, is the chlorophyll concentration.

7. The intelligent resource scheduling and management method for urban inland water systems based on dynamic data collaboration according to claim 5, wherein In the step S23, the setting of constraint conditions includes the following steps: S231. Determine the minimum ecological flow rate and calculate the ecological base flow threshold. The formula is: , where is the minimum ecological flow rate, is the proportionality coefficient, is the average flow rate; S232. Calculate the maximum water transfer volume according to the maximum opening of the sluice and the power of the pumping station, and determine the water conveyance capacity of the project. The formula is as follows: , where is the maximum water transfer volume, is the discharge coefficient, is the maximum opening of the sluice, is the opening of the sluice, is the acceleration of gravity; S233. Quantify the substitution relationships of each water source and establish a water source complementary matrix. The formula is as follows: , where in the formula, is the complementary matrix, is the substitution coefficient, and .

8. The intelligent resource scheduling and management method for urban inland water systems based on dynamic data collaboration according to claim 2, wherein In the step S13, the monitoring of river water index data includes the following steps: S131. Obtain the cross-sectional flow velocity distribution through an acoustic Doppler velocimeter and measure the river flow velocity distribution; S132. Calculate the longitudinal diffusion coefficient of the river water, perform spatial differentiation on the measured concentration data, and solve the concentration gradient. The formula is: , where is the longitudinal diffusion coefficient, is the adjustment coefficient, is the substance concentration, is the spatial coordinate, and are respectively the measured substance concentrations at the -th and the -th monitoring points, is the distance between two adjacent monitoring points, is the variable index of the river water flow velocity, is the gradient concentration; S133. Set the diffusion flux threshold corresponding to the environmental capacity. Calculate the diffusion flux based on the calculation results of the longitudinal diffusion coefficient and the gradient concentration, and evaluate the diffusion impact. The formula is: , where is the actual diffusion flux. When the actual diffusion flux exceeds the threshold, that is, the diffusion impact exceeds the environmental carrying capacity, and emergency dispatching needs to be triggered.

9. The intelligent resource scheduling and management method for urban inland water systems based on dynamic data collaboration according to claim 1, characterized in that, In the step S5, the formula for correcting the model parameters is as follows: , where and are the parameter values at the -th and -th iterations respectively, is the learning rate, is the loss function, and , is the number of samples of the actual water transfer data, and are the actual water transfer data and the model prediction value respectively.

10. An intelligent resource scheduling and management system for urban inland water systems based on dynamic data collaboration, which is applied to the intelligent resource scheduling and management method for urban inland water systems based on dynamic data collaboration according to any one of claims 1-9, characterized in that, It includes: A data collection and processing module, which is used to collect various data of the river in real time and process the data to provide an accurate data basis for subsequent analysis; A model construction module, which is used to establish a multi-source scheduling model to provide model support for formulating a water transfer plan; An optimization calculation module, which is used to establish a global dynamic optimization model to minimize the water transfer cost and ecological impact; A scheduling execution module, which is used to intelligently control the water transfer facility equipment according to the optimization results to ensure the reasonable allocation of water resources; A feedback optimization module, which is used to compare the actual water transfer data, correct the model parameters based on deviation analysis, and ensure the real-time update and continuous optimization of the water transfer plan.

Images