Water plant whole-process intelligent optimization scheduling method based on adaptive feedback control

Abstract

The invention discloses a water plant whole-process intelligent optimization scheduling method based on adaptive feedback control. The method comprises the following steps: S1, dividing into a low-demand stage, a medium-demand stage and a high-demand stage; s2, in a low-demand stage, a medium-demand stage and a high-demand stage, correspondingly obtaining real-time monitoring data sets of the water plant; s3, constructing a multi-stage evolutionary game model based on the water plant real-time monitoring data set; s4, generating a local optimal scheduling scheme of the current stage; s5, outputting a local optimal scheduling scheme according to the evolutionary game model by utilizing a self-adaptive feedback control mechanism; s6, performing global optimal scheduling of the system in each stage; s7, optimizing a scheduling scheme in advance; and S8, displaying the optimized scheduling effect through the integrated intelligent scheduling platform, monitoring the operation state of the whole water plant, and generating a scheduling report for reference of management personnel. According to the invention, the predictability and perspectiveness of the scheduling system are effectively improved through the prediction model, and it is ensured that the water plant is always in an efficient and safe operation state.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent water plants, and in particular to a full-process intelligent optimization scheduling method for water plants based on adaptive feedback control. Background Art

[0002] In the prior art, the optimization of water plant dispatching systems mainly relies on traditional rules and preset parameters for control. Fixed operating strategies are usually used to adjust water pump flow, dosage and filtration speed. The control process of traditional systems is often based on historical data and empirical parameters, and lacks sensitivity to real-time changes. As external and internal conditions such as water quality, water demand and energy consumption change, traditional dispatching systems are difficult to respond in a timely manner, which often leads to waste of resources, increased energy consumption, and even affects the overall operating efficiency of the water plant and the rate of water quality compliance.

[0003] Another typical problem with existing technologies is that the scheduling system lacks a dynamic feedback mechanism. In actual operation, the process links of a water plant are complex and changeable, and the status of each link may fluctuate with multiple factors such as the environment, equipment aging, or changes in water quality. Since traditional systems cannot collect and feedback the status changes of each process link in real time, they are unable to adjust strategies in time to cope with these fluctuations. The scheduling plan often lags behind the actual operating status, resulting in low operating efficiency and even safety issues.

[0004] In summary, the existing technology has the following shortcomings in water plant scheduling: First, the traditional scheduling system relies on fixed parameters and preset rules, and cannot make adaptive adjustments based on real-time data, making it difficult to cope with dynamic changes in the external environment and internal conditions of the system; second, it lacks the ability of multi-objective optimization and cannot achieve a balance between multiple optimization goals such as water quality, energy consumption and cost; finally, the feedback mechanism of the existing system is not perfect, making it difficult to form a closed-loop control, resulting in an inability to respond to changes in operation in a timely manner, affecting the overall operational efficiency and safety of the water plant. Summary of the invention

[0005] One purpose of the present invention is to propose a full-process intelligent optimization scheduling method for a water plant based on adaptive feedback control. The present invention effectively improves the predictability and foresight of the scheduling system through a prediction model, ensuring that the water plant is always in an efficient and safe operating state.

[0006] A method for intelligent optimization scheduling of the entire process of a water plant based on adaptive feedback control according to an embodiment of the present invention comprises the following steps:

[0007] S1. Dynamically divide the operation process of the water plant into low demand stage, medium demand stage and high demand stage based on water quality parameters, energy consumption data, water demand and real-time parameters of equipment operation status;

[0008] S2. Obtain real-time monitoring data sets of the water plant in the low demand stage, medium demand stage and high demand stage, including water quality parameters, energy consumption data, equipment status and external environment parameters, and establish a real-time data collection system related to each process link;

[0009] S3. A multi-stage evolutionary game model is constructed based on the real-time monitoring data set of the water plant. Each process link in the water plant is taken as a game participant, and the strategy set and profit function of each game participant are defined. The profit function includes the water quality compliance rate, energy consumption and cost optimization goals;

[0010] S4. In each stage, the evolutionary game model is used to simulate the strategy evolution between each process link, and the strategy combination of each game participant is calculated and optimized by replicating the dynamic equation to generate the local optimal scheduling plan for the current stage;

[0011] S5. Using the adaptive feedback control mechanism, according to the local optimal scheduling plan output by the evolutionary game model, the pump flow, dosage and filtration speed of each process link of the water plant are adjusted in real time, and new real-time monitoring data are continuously collected during the adjustment process and fed back to the control system to form a closed-loop feedback regulation;

[0012] S6. When the operating conditions or external environment of the water plant change, the operation stage is dynamically adjusted according to the local optimal scheduling scheme of the evolutionary game model and the adaptive feedback control mechanism, and the strategy combination of each process link is switched to achieve the global optimal scheduling of the system at each stage;

[0013] S7. Combine historical monitoring data with real-time monitoring data to predict the possible system change trend of the water plant in the future and optimize the dispatching plan in advance;

[0014] S8. The optimized scheduling effect is displayed through the integrated intelligent scheduling platform, the operating status of the entire water plant is monitored and a scheduling report is generated for the reference of managers.

[0015] Optionally, the S1 includes:

[0016] S11. Obtain water quality parameters P, energy consumption data E, water demand D and equipment operation status parameters S of the water plant operation, and use the data as input to form a real-time data set X(t) of the water plant;

[0017] S12. Based on the real-time data set X(t), the demand stage division standard of the water plant is set as low demand stage L, medium demand stage M and high demand stage H. The demand stage division standard is defined by the threshold range of each parameter:

[0018] The interval of water quality parameter P is [P low ,P high ], the energy consumption data E is in the range of [E low ,E high], the interval of water demand D is [D low ,D high ], the interval of the equipment state parameter S is [S low ,S high ];

[0019] S13. Determine the demand stage to which each parameter belongs at a certain time t by calculating the real-time data set X(t), and use the classification function F(X(t)) to classify the operation status of the water plant:

[0020] L, if P(t)∈[P low ,P mid ]and E(t)∈[E low ,E mid ]and D(t)∈[D low ,D mid ]and S(t)∈[S low ,S mid ]F(X(t))={M,ifP(t)∈[P mid ,P high ]andE(t)∈[E mid ,E high ]andD(t)∈[D mid ,D high ]andS(t)∈[S mid ,S high ];

[0021] H, if P(t)>P high or E(t)>E high or D(t)>D high or S(t)>S high

[0022] S14. According to the real-time data set X(t) and the classification function F(X(t)), the operation process of the water plant is dynamically divided into a low demand stage, a medium demand stage and a high demand stage.

[0023] Optionally, S2 includes:

[0024] S21. Obtain real-time monitoring data sets of each process link of the water plant during the low demand stage, medium demand stage and high demand stage of the water plant, including water quality parameters P(t), energy consumption data E(t), equipment status S(t) and external environment parameters W(t):

[0025] P(t) represents the relevant parameters of the water quality of the water plant at a certain time t, including turbidity, pH value and chemical oxygen demand (COD);

[0026] E(t) represents the energy consumption data of the water plant at time t, including total energy consumption and sub-item energy consumption of each process link;

[0027] S(t) represents the operating status parameters of the main equipment in the water plant, including the pump pressure, filter status, and the working status of the dosing system;

[0028] W(t) represents the external environmental parameters, including temperature, humidity and water quality parameters of the external water source;

[0029] S22. Based on the real-time monitoring of water quality parameters, energy consumption data, equipment status and external environmental parameters of each process link, a real-time data acquisition system DCS(t) is constructed. The real-time data acquisition system independently monitors and collects data for each process link in different demand stages, and sets the acquisition frequency f of the real-time data acquisition system. Different real-time data acquisition frequencies f are set in the low demand stage, medium demand stage and high demand stage respectively. L ,f M ,f H :

[0030] f L The frequency of data collection for the low demand phase;

[0031] f M The frequency of data collection for the medium demand stage;

[0032] f H The frequency of data collection for the high-demand phase;

[0033] S24. Dynamically adjust the parameter settings of the real-time data acquisition system DCS(t) according to the monitoring requirements of each stage to build a real-time monitoring data set for the water plant:

[0034] X(t)={P(t),E(t),S(t),W(t)}.

[0035] Optionally, the S3 includes:

[0036] S31. A multi-stage evolutionary game model is constructed based on the real-time monitoring data set of the water plant, and the water pump, dosing system and filtration system in the water plant are used as game participants A. i , where i = 1, 2, 3;

[0037] S32. Define each game participant A i The strategy set S i , the policy set contains different operation modes and operating parameters:

[0038] For pumps, the strategy set includes different flow settings;

[0039] For dosing systems, the strategy set includes different dosing amount settings;

[0040] For the filtering system, the policy set includes different filter speed settings;

[0041] S33. Define the payoff function U for each game participant i (S i ,S -i ), the revenue function aims at optimizing water quality compliance rate, energy consumption and cost:

[0042] U i (S i ,S -i )=α i P(t)-β i E(t)-γ i C(t);

[0043] Among them, P(t) represents the water quality compliance rate, E(t) represents the energy consumption of the water pump, dosing system or filtration system, C(t) represents the cost of the corresponding process link, α i , β i , γ i The weight coefficients for each participant correspond to the goals of water quality compliance rate, energy consumption and cost optimization.

[0044] Optionally, the process of constructing the local optimal scheduling solution in the current stage includes, in each operation stage, game participants updating strategies according to a complex replication dynamic equation:

[0045]

[0046] Among them, x i For strategy S i The proportion of water pump, dosing system and filtration system in the game, U i (S i ,S -i ,t) is the game participant A i The profit function changes with time t, is the rate of change of water quality compliance rate over time, indicating the changing trend of water quality in the current water plant. is the rate of change of energy consumption over time, indicating the changing trend of energy consumption in each process link. is the rate of change of equipment status over time, indicating the change trend of the operating status of the water pump, dosing system and filtration system, λ i , μ i , ν i They are weight coefficients respectively:

[0047] λ i Indicates process step A i Sensitivity to changes in water quality compliance rates, reflecting the impact of water pumps, dosing systems and filtration systems on water quality;

[0048] μ i Indicates process step A i Sensitivity to changes in energy consumption determines regulation to minimize energy consumption;

[0049] ν i Indicates process step A i The sensitivity to changes in equipment status determines the degree of impact on equipment maintenance and adjustments;

[0050] is the average payoff of all game participants:

[0051]

[0052] Among them, n is the number of process links participating in the game.

[0053] Optionally, the S5 includes:

[0054] S51. Using the adaptive feedback control mechanism, according to the local optimal scheduling scheme output by the evolutionary game model, the operating parameters of each process link of the water plant are adjusted in real time, including the pump flow rate Q(t), the dosage M(t) and the filtration speed V(t);

[0055] S52. During the adjustment process, the real-time monitoring data set X'(t) of the water transfer plant is collected in real time, and the real-time monitoring data set X'(t) of the water transfer plant is transmitted to the control system through the feedback mechanism to update the participant benefit function U' in the multi-stage evolutionary game model i (S i ,S -i ,t), so that the multi-stage evolutionary game model can be dynamically adjusted according to the latest operating status and external environment parameters;

[0056] S53. Based on the feedback data and the strategy combination output by the evolutionary game model, the process parameters are optimized and adjusted using the adaptive feedback control equation:

[0057]

[0058] Among them, λ1, λ2, λ3 are adjustment coefficients, which control the response of the water pump, dosing system and filtration system to the adjustment.

[0059] The speed of ″′, U Q (t),U M (t),U V (t) are the profit functions of the water pump, dosing system and filtration system respectively, is the average payoff function of all game participants, is the rate of change of water quality parameters, is the second-order derivative of the external environment parameter, is the rate of change of energy consumption, is the square of the rate of change of equipment status parameters, α1, α2, α3 are the weights of the impact of water quality parameter changes on each link, and β1, β2, β3 are the weights of the impact of energy consumption changes on each link.

[0060] S54. Through the closed-loop feedback regulation mechanism, the system dynamically optimizes the operating parameters of the process links according to the real-time monitoring data of the water plant. During the adjustment process, each process link continuously updates its profit function and control parameters to achieve the optimal dynamic regulation effect, forming an adaptive closed-loop regulation system.

[0061] Optionally, the S6 includes:

[0062] S61. When the water plant operating conditions or external environmental parameters W(t) change, real-time monitoring of the change rate of the external environment Combined with the real-time monitoring data set of the water plant, the local optimal scheduling plan is calculated through a multi-stage evolutionary game model;

[0063] S62. Dynamically adjust the operation phase of the water plant according to the calculated local optimal scheduling plan, switch between the low demand phase, the medium demand phase and the high demand phase, and update the operating parameters of the process link through the combination of participant strategies in the multi-stage evolutionary game model;

[0064] S63. Use the adaptive feedback control mechanism to dynamically adjust the strategy combination of the process link according to the changes in the external environment and operating conditions. When the strategy combination of the process link changes, the control system automatically optimizes the operating parameters of each link of the water plant according to the adaptive feedback control mechanism to achieve the global optimal scheduling at each stage;

[0065] S64. As the operation phase switches, the control system adjusts the profit function and strategy combination of each process link according to real-time data so that the water plant always maintains the global optimal scheduling when the external environment and operating conditions change.

[0066] Optionally, the S7 includes:

[0067] S71. Combined with water plant historical monitoring data X hist (t) and real-time monitoring data X(t) to build a prediction model for analyzing system change trends;

[0068] S72. In the prediction model, the change trend equation of water quality parameter P, energy consumption data E, water demand D and equipment operation status parameter S is calculated by time series analysis method:

[0069]

[0070] Among them, X pred(t+Δt) is the predicted value at the future time t+Δt, X(t) represents any water quality parameter P(t), energy consumption data E(t), water demand D(t) or equipment operation status parameter S(t), α X and β X are the weight coefficients of the corresponding parameters respectively;

[0071] S74. Optimize the water plant dispatch plan based on the water quality fluctuations, energy consumption, water demand and equipment load change trends output by the prediction model, and adjust the operating parameters of each process link of the water plant before system changes occur.

[0072] The beneficial effects of the present invention are:

[0073] (1) The present invention combines real-time monitoring data and a multi-stage evolutionary game model through an adaptive feedback control mechanism to dynamically adjust the operating parameters of each process link during the operation of the water plant. The system performs closed-loop regulation based on the real-time collected data. When the external environment or internal operating conditions change, it can respond quickly and automatically adjust the scheduling plan to ensure that the water plant always operates in the optimal state. Compared with traditional fixed parameter scheduling, the present invention shows extremely high flexibility and adaptability in dealing with water quality fluctuations, energy consumption changes, and equipment load changes, greatly improving the response speed and efficiency of the system.

[0074] (2) The present invention regards each process link in the water plant as a game participant by constructing a multi-stage evolutionary game model, and defines the benefit function of each participant, comprehensively considering multiple goals such as water quality compliance rate, energy consumption optimization and cost control. By replicating the strategy evolution of the dynamic equation, it is possible to achieve local optimality in each stage, and dynamically adjust the strategy combination of different stages through an adaptive feedback control mechanism, thereby achieving global optimal scheduling.

[0075] (3) The present invention combines historical monitoring data with real-time monitoring data using time series analysis and prediction models, which can accurately predict future system change trends of the water plant. The prediction results can be used to optimize the scheduling plan in advance before system changes occur, thereby avoiding the impact of emergencies on water plant operations. Compared with the lagging adjustment in the prior art, the present invention effectively improves the predictability and foresight of the scheduling system through the prediction model, ensuring that the water plant is always in an efficient and safe operating state. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0077] Figure 1 This is a flow chart of a method for intelligent optimization scheduling of the entire process of a water plant based on adaptive feedback control proposed by the present invention;

[0078] Figure 2 The present invention provides a flow chart of a combination of game strategies for process links based on a multi-stage evolutionary game model in a method for intelligent optimization scheduling of the entire process of a water plant based on adaptive feedback control. DETAILED DESCRIPTION

[0079] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0080] refer to Figure 1-2 , a water plant full-process intelligent optimization scheduling method based on adaptive feedback control, comprising the following steps:

[0081] S1. Dynamically divide the operation process of the water plant into low demand stage, medium demand stage and high demand stage based on water quality parameters, energy consumption data, water demand and real-time parameters of equipment operation status;

[0082] S2. Obtain real-time monitoring data sets of the water plant in the low demand stage, medium demand stage and high demand stage, including water quality parameters, energy consumption data, equipment status and external environment parameters, and establish a real-time data collection system related to each process link;

[0083] S3. A multi-stage evolutionary game model is constructed based on the real-time monitoring data set of the water plant. Each process link in the water plant is taken as a game participant, and the strategy set and profit function of each game participant are defined. The profit function includes the water quality compliance rate, energy consumption and cost optimization goals;

[0084] S4. In each stage, the evolutionary game model is used to simulate the strategy evolution between each process link, and the strategy combination of each game participant is calculated and optimized by replicating the dynamic equation to generate the local optimal scheduling plan for the current stage;

[0085] S5. Using the adaptive feedback control mechanism, according to the local optimal scheduling plan output by the evolutionary game model, the pump flow, dosage and filtration speed of each process link of the water plant are adjusted in real time, and new real-time monitoring data are continuously collected during the adjustment process and fed back to the control system to form a closed-loop feedback regulation;

[0086] S6. When the operating conditions or external environment of the water plant change, the operation stage is dynamically adjusted according to the local optimal scheduling scheme of the evolutionary game model and the adaptive feedback control mechanism, and the strategy combination of each process link is switched to achieve the global optimal scheduling of the system at each stage;

[0087] S7. Combine historical monitoring data with real-time monitoring data to predict the possible system change trend of the water plant in the future and optimize the dispatching plan in advance;

[0088] S8. The optimized scheduling effect is displayed through the integrated intelligent scheduling platform, the operating status of the entire water plant is monitored and a scheduling report is generated for the reference of managers.

[0089] In this implementation, S1 includes:

[0090] S11. Obtain water quality parameters P, energy consumption data E, water demand D and equipment operation status parameters S of the water plant operation, and use the data as input to form a real-time data set X(t) of the water plant;

[0091] S12. Based on the real-time data set X(t), the demand stage division standard of the water plant is set as low demand stage L, medium demand stage M and high demand stage H. The demand stage division standard is defined by the threshold range of each parameter:

[0092] The interval of water quality parameter P is [P low ,P high ], the energy consumption data E is in the range of [E low ,E high ], the interval of water demand D is [D low ,D high ], the interval of the equipment state parameter S is [S low ,S high ];

[0093] S13. Determine the demand stage to which each parameter belongs at a certain time t by calculating the real-time data set X(t), and use the classification function F(X(t)) to classify the operation status of the water plant:

[0094] L, if P(t)∈[P low ,P mid ]and E(t)∈[E low ,E mid ]and D(t)∈[D low ,D mid ]and S(t)∈[S low ,S mid ]F(X(t))={M,if P(t)∈[P mid ,P high ]and E(t)∈[E mid ,E high ]andD(t)∈[D mid ,D high ]andS(t)∈[S mid ,S high ]; H, if P(t)>P high or E(t)>E high orD(t)>D highorS(t)>S high

[0095] S14. According to the real-time data set X(t) and the classification function F(X(t)), the operation process of the water plant is dynamically divided into a low demand stage, a medium demand stage and a high demand stage.

[0096] In this implementation, S2 includes:

[0097] S21. Obtain real-time monitoring data sets of each process link of the water plant during the low demand stage, medium demand stage and high demand stage of the water plant, including water quality parameters P(t), energy consumption data E(t), equipment status S(t) and external environment parameters W(t):

[0098] P(t) represents the relevant parameters of the water quality of the water plant at a certain time t, including turbidity, pH value and chemical oxygen demand (COD);

[0099] E(t) represents the energy consumption data of the water plant at time t, including total energy consumption and sub-item energy consumption of each process link;

[0100] S(t) represents the operating status parameters of the main equipment in the water plant, including the pump pressure, filter status, and the working status of the dosing system;

[0101] W(t) represents the external environmental parameters, including temperature, humidity and water quality parameters of the external water source;

[0102] S22. Based on the real-time monitoring of water quality parameters, energy consumption data, equipment status and external environmental parameters of each process link, a real-time data acquisition system DCS(t) is constructed. The real-time data acquisition system independently monitors and collects data for each process link in different demand stages, and sets the acquisition frequency f of the real-time data acquisition system. Different real-time data acquisition frequencies f are set in the low demand stage, medium demand stage and high demand stage respectively. L ,f M ,f H :

[0103] f L The frequency of data collection for the low demand phase;

[0104] f M The frequency of data collection for the medium demand stage;

[0105] f H The frequency of data collection for the high-demand phase;

[0106] S24. Dynamically adjust the parameter settings of the real-time data acquisition system DCS(t) according to the monitoring requirements of each stage to build a real-time monitoring data set for the water plant:

[0107] X(t)={P(t),E(t),S(t),W(t)}.

[0108] In this implementation, S3 includes:

[0109] S31. A multi-stage evolutionary game model is constructed based on the real-time monitoring data set of the water plant, and the water pump, dosing system and filtration system in the water plant are used as game participants A. i , where i = 1, 2, 3;

[0110] S32. Define each game participant A i The strategy set S i , the policy set contains different operation modes and operating parameters:

[0111] For pumps, the strategy set includes different flow settings;

[0112] For dosing systems, the strategy set includes different dosing amount settings;

[0113] For the filtering system, the policy set includes different filter speed settings;

[0114] S33. Define the payoff function U for each game participant i (S i ,S -i ), the revenue function aims at optimizing water quality compliance rate, energy consumption and cost:

[0115] U i (S i ,S -i )=α i P(t)-β i E(t)-γ i C(t);

[0116] Among them, P(t) represents the water quality compliance rate, E(t) represents the energy consumption of the water pump, dosing system or filtration system, C(t) represents the cost of the corresponding process link, α i , β i , γ i The weight coefficients for each participant correspond to the goals of water quality compliance rate, energy consumption and cost optimization.

[0117] In this implementation, the process of constructing the local optimal scheduling solution at the current stage includes, in each operation stage, game participants updating their strategies according to a complex replication dynamic equation:

[0118]

[0119] Among them, x i For strategy S iThe proportion of water pump, dosing system and filtration system in the game, U i (S i ,S -i ,t) is the game participant A i The profit function changes with time t, is the rate of change of water quality compliance rate over time, indicating the changing trend of water quality in the current water plant. is the rate of change of energy consumption over time, indicating the changing trend of energy consumption in each process link. is the rate of change of equipment status over time, indicating the change trend of the operating status of the water pump, dosing system and filtration system, λ i , μ i , ν i They are weight coefficients respectively:

[0120] λ i Indicates process step A i Sensitivity to changes in water quality compliance rates, reflecting the impact of water pumps, dosing systems and filtration systems on water quality;

[0121] μ i Indicates process step A i Sensitivity to changes in energy consumption determines regulation to minimize energy consumption;

[0122] ν i Indicates process step A i The sensitivity to changes in equipment status determines the degree of impact on equipment maintenance and adjustments;

[0123] is the average payoff of all game participants:

[0124]

[0125] Among them, n is the number of process links participating in the game.

[0126] In this implementation, S5 includes:

[0127] S51. Using the adaptive feedback control mechanism, according to the local optimal scheduling scheme output by the evolutionary game model, the operating parameters of each process link of the water plant are adjusted in real time, including the pump flow rate Q(t), the dosage M(t) and the filtration speed V(t);

[0128] S52. During the adjustment process, the real-time monitoring data set X'(t) of the water transfer plant is collected in real time, and the real-time monitoring data set X'(t) of the water transfer plant is transmitted to the control system through the feedback mechanism to update the participant benefit function U' in the multi-stage evolutionary game model i (S i ,S -i,t), so that the multi-stage evolutionary game model can be dynamically adjusted according to the latest operating status and external environment parameters;

[0129] S53. Based on the feedback data and the strategy combination output by the evolutionary game model, the process parameters are optimized and adjusted using the adaptive feedback control equation:

[0130]

[0131]

[0132] Among them, λ1, λ2, λ3 are adjustment coefficients, which control the response of the water pump, dosing system and filtration system to the adjustment.

[0133] The speed of ″′, U Q (t),U M (t),U V (t) are the profit functions of the water pump, dosing system and filtration system respectively, is the average payoff function of all game participants, is the rate of change of water quality parameters, is the second-order derivative of the external environment parameter, is the rate of change of energy consumption, is the square of the rate of change of equipment status parameters, α1, α2, α3 are the weights of the impact of water quality parameter changes on each link, and β1, β2, β3 are the weights of the impact of energy consumption changes on each link.

[0134] S54. Through the closed-loop feedback regulation mechanism, the system dynamically optimizes the operating parameters of the process links according to the real-time monitoring data of the water plant. During the adjustment process, each process link continuously updates its profit function and control parameters to achieve the optimal dynamic regulation effect, forming an adaptive closed-loop regulation system.

[0135] In this implementation, S6 includes:

[0136] S61. When the water plant operating conditions or external environmental parameters W(t) change, real-time monitoring of the change rate of the external environment Combined with the real-time monitoring data set of the water plant, the local optimal scheduling plan is calculated through a multi-stage evolutionary game model;

[0137] S62. Dynamically adjust the operation phase of the water plant according to the calculated local optimal scheduling plan, switch between the low demand phase, the medium demand phase and the high demand phase, and update the operating parameters of the process link through the combination of participant strategies in the multi-stage evolutionary game model;

[0138] S63. Use the adaptive feedback control mechanism to dynamically adjust the strategy combination of the process link according to the changes in the external environment and operating conditions. When the strategy combination of the process link changes, the control system automatically optimizes the operating parameters of each link of the water plant according to the adaptive feedback control mechanism to achieve the global optimal scheduling at each stage;

[0139] S64. As the operation phase switches, the control system adjusts the profit function and strategy combination of each process link according to real-time data so that the water plant always maintains the global optimal scheduling when the external environment and operating conditions change.

[0140] In this implementation, S7 includes:

[0141] S71. Combined with water plant historical monitoring data X hist (t) and real-time monitoring data X(t) to build a prediction model for analyzing system change trends;

[0142] S72. In the prediction model, the change trend equation of water quality parameter P, energy consumption data E, water demand D and equipment operation status parameter S is calculated by time series analysis method:

[0143]

[0144] Among them, X pred (t+Δt) is the predicted value at the future time t+Δt, X(t) represents any water quality parameter P(t), energy consumption data E(t), water demand D(t) or equipment operation status parameter S(t), α X and β X are the weight coefficients of the corresponding parameters respectively;

[0145] S74. Optimize the water plant dispatch plan based on the water quality fluctuations, energy consumption, water demand and equipment load change trends output by the prediction model, and adjust the operating parameters of each process link of the water plant before system changes occur.

[0146] Embodiment 1:

[0147] In order to verify the feasibility of the present invention in the intelligent optimization scheduling of the entire process of a water plant, the implementers conducted experiments using a certain city’s actual water plant operation data set and a simulated changing environment data set. The water plant’s data set covers operating data from January 2021 to June 2023, including water quality parameters, energy consumption data, equipment operating status and external environmental conditions. Based on these data, the implementers also generated a changing environment data set simulating peak periods and equipment failure conditions to evaluate the scheduling optimization capabilities of the present invention in complex scenarios.

[0148] The implementers divided the data set into a training set and a test set in a ratio of 8:2. The training set contains historical operation data from the past 2.5 years, which is used to train the strategy combination and adaptive feedback mechanism in the multi-stage evolutionary game model. The test set is used to verify the system's real-time response and scheduling optimization performance in a complex environment.

[0149] In the data collection and preprocessing stage, the implementers used multiple process parameters such as pump flow, dosage, and filtration speed as input features, and water quality compliance rate, energy consumption, and equipment load as the main output targets. At the same time, the implementers modeled the impact of the external environment, which included temperature, humidity, and water quality changes. The data was acquired in real time through a sensor network. The implementers used an adaptive moment estimation optimizer to train the multi-stage evolutionary game model and set the learning rate to 0.001. The training was carried out for a total of 500 epochs, and the batch size was set to 32.

[0150] In order to adapt to the optimization process of the present invention, the implementers took the various process links of the water plant as game participants and constructed a strategy combination model based on multi-stage game. Through the feedback mechanism, the system dynamically adjusted the strategy combination according to water quality fluctuations, energy consumption data and equipment status changes, and generated the optimal operating parameters for each process link.

[0151] In practical applications, the implementers focused on comparing the performance of the intelligent optimization scheduling method of the present invention with the traditional fixed parameter scheduling method, especially the responsiveness during peak hours and under complex environmental conditions. The specific experimental results and data are shown in Table 1 below:

[0152] Table 1 Comparison of experimental data between the intelligent optimization scheduling method of the present invention and the traditional scheduling method

[0153] Compare Projects Traditional methods Method of the present invention Pump energy consumption (kWh) 16000(per day) 13500 (per day) Dosage(g / L) 0.85 (unbalanced) 0.75(precise control) Water quality compliance rate 88% (down from peak) 97% (stable throughout) Filter system load 90% (often overloaded) 75% (stable operation) Dispatch response time (minutes) 45-60(manual adjustment) 10-15 (system automatically adjusts) System operation cost (yuan) 52000(per day) 43000(per day)

[0154] During the training process, the multi-stage evolutionary game model of the present invention performs strategy optimization based on actual data sets and simulated data sets. Traditional methods rely on manually set fixed parameters and are difficult to cope with dynamic changes in water plant operations. The present invention uses evolutionary games and adaptive feedback control mechanisms to automatically adjust water pump flow, dosage and filtration speed when water quality fluctuations or equipment status changes are detected, greatly improving the real-time and accuracy of scheduling.

[0155] In order to further prove the effectiveness of the present invention, the implementers introduced two types of extreme situations in the test set: the first type is the peak water consumption period, and the second type is the equipment failure simulation. By comparing the scheduling effects in these two types of situations, the implementers found that the traditional method increased the energy consumption of the water pump by 25% during the peak water consumption period and equipment failure, and the dosage was difficult to accurately control, resulting in water quality fluctuations and frequent overload operation of the filtration system. The optimization method of the present invention can automatically adjust the strategy under these extreme conditions, reducing the energy consumption of the water pump by 15%, controlling the dosage accurately to within 0.01g / L, maintaining the water quality compliance rate at more than 97%, and reducing the load of the filtration system by 15%.

[0156] In addition, in the comparative tests under different seasons and different environmental conditions, the system of the present invention showed high robustness. For example, under low temperature conditions in winter, the water quality compliance rate of the traditional dispatching system dropped to 85% due to insufficient response to environmental changes. However, the water quality compliance rate of the present invention remained above 96% by adaptively adjusting the water pump flow and dosage. Under high temperature and high humidity conditions in summer, the system of the present invention can still maintain efficient operation, and the energy consumption is reduced by about 18% compared with the traditional system. The following Table 2 further shows the performance comparison between the system of the present invention and the traditional system under different environmental conditions:

[0157] Table 2 Performance comparison between the system of the present invention and the traditional system

[0158]

[0159]

[0160] In addition, the present invention also predicts water quality fluctuations and equipment load changes in the next few hours by combining historical data with real-time data, and optimizes the scheduling plan in advance. For example, if it is predicted that the external temperature will rise and the water demand will increase in the next two hours, the system will automatically adjust the flow rate and dosage of the water pump to ensure that the water quality is stable and the equipment load does not exceed the acceptable range before the peak period.

[0161] To sum up, this embodiment uses data sets of real water plants and data sets of simulated environments to conduct detailed experimental comparisons between the method of the present invention and traditional methods, verifying the feasibility and superiority of the present invention in complex environments. Through the full-process intelligent optimization scheduling method of the water plant of the present invention, the water plant can greatly improve the scheduling efficiency, reduce energy consumption, and ensure the water quality compliance rate, and perform well in dealing with peak periods and complex equipment failures.

[0162] The present invention combines real-time monitoring data and a multi-stage evolutionary game model through an adaptive feedback control mechanism to dynamically adjust the operating parameters of each process link during the operation of the water plant. The system performs closed-loop regulation based on the real-time collected data, and can respond quickly and automatically adjust the scheduling plan when the external environment or internal operating conditions change, ensuring that the water plant always operates in the optimal state. Compared with traditional fixed parameter scheduling, the present invention exhibits extremely high flexibility and adaptability in dealing with water quality fluctuations, energy consumption changes, and equipment load changes, greatly improving the response speed and efficiency of the system.

[0163] The present invention regards each process link in the water plant as a game participant by constructing a multi-stage evolutionary game model, and defines the benefit function of each participant, comprehensively considering multiple goals such as water quality compliance rate, energy consumption optimization and cost control. By replicating the strategy evolution of the dynamic equation, it is possible to achieve local optimality in each stage, and dynamically adjust the strategy combination of different stages through an adaptive feedback control mechanism, thereby achieving global optimal scheduling.

[0164] The present invention combines historical monitoring data with real-time monitoring data using time series analysis and prediction models, and can accurately predict future system change trends of the water plant. Through the prediction results, the scheduling plan can be optimized in advance before the system changes occur, avoiding the impact of emergencies on the operation of the water plant. Compared with the lagging adjustment in the prior art, the present invention effectively improves the predictability and foresight of the scheduling system through the prediction model, ensuring that the water plant is always in an efficient and safe operating state.

[0165] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for intelligent optimization scheduling of the entire process of a water plant based on adaptive feedback control, characterized in that: The steps include: S1. Dynamically divide the operation process of the water plant into low demand stage, medium demand stage and high demand stage based on water quality parameters, energy consumption data, water demand and real-time parameters of equipment operation status; S2. Obtain real-time monitoring data sets of the water plant in the low demand stage, medium demand stage and high demand stage, including water quality parameters, energy consumption data, equipment status and external environment parameters, and establish a real-time data collection system related to each process link; S3. A multi-stage evolutionary game model is constructed based on the real-time monitoring data set of the water plant. Each process link in the water plant is taken as a game participant, and the strategy set and profit function of each game participant are defined. The profit function includes the water quality compliance rate, energy consumption and cost optimization goals; S4. In each stage, the evolutionary game model is used to simulate the strategy evolution between each process link, and the strategy combination of each game participant is calculated and optimized by replicating the dynamic equation to generate the local optimal scheduling plan for the current stage; S5. Using the adaptive feedback control mechanism, according to the local optimal scheduling plan output by the evolutionary game model, the pump flow, dosage and filtration speed of each process link of the water plant are adjusted in real time, and new real-time monitoring data are continuously collected during the adjustment process and fed back to the control system to form a closed-loop feedback regulation; S6. When the operating conditions or external environment of the water plant change, the operation stage is dynamically adjusted according to the local optimal scheduling scheme of the evolutionary game model and the adaptive feedback control mechanism, and the strategy combination of each process link is switched to achieve the global optimal scheduling of the system at each stage; S7. Combine historical monitoring data with real-time monitoring data to predict the possible system change trend of the water plant in the future and optimize the dispatching plan in advance; S8. The optimized scheduling effect is displayed through the integrated intelligent scheduling platform, the operating status of the entire water plant is monitored and a scheduling report is generated for the reference of managers.

2. According to claim 1, a method for intelligent optimization and scheduling of the entire process of a water plant based on adaptive feedback control is characterized in that: The S1 includes: S11. Obtain water quality parameters P, energy consumption data E, water demand D and equipment operation status parameters S of the water plant operation, and use the data as input to form a real-time data set X(t) of the water plant; S12. Based on the real-time data set X(t), the demand stage division standard of the water plant is set as low demand stage L, medium demand stage M and high demand stage H. The demand stage division standard is defined by the threshold range of each parameter: The interval of water quality parameter P is [P low ,P high ], the energy consumption data E is in the range of [E low ,E high ], the interval of water demand D is [D low ,D high ], the interval of the equipment state parameter S is [S low ,S high ]; S13. Determine the demand stage to which each parameter belongs at a certain time t by calculating the real-time data set X(t), and use the classification function F(X(t)) to classify the operation status of the water plant: L,if P(t)∈[P low ,P mid ]and E(t)∈[E low ,E mid ]and D(t)∈[D low ,D mid ]and S(t)∈[S low ,S mid ] F(X(t))={M,if P(t)∈[P mid ,P high ]and E(t)∈[E mid ,E high ]and D(t)∈[D mid ,D high ]andS(t)∈[S mid ,S high ]; H,if P(t)>P high or E(t)>E high or D(t)>D high or S(t)>S high S14. According to the real-time data set X(t) and the classification function F(X(t)), the operation process of the water plant is dynamically divided into a low demand stage, a medium demand stage and a high demand stage.

3. According to claim 1, a method for intelligent optimization and scheduling of the entire process of a water plant based on adaptive feedback control is characterized in that: The S2 includes: S21. Obtain real-time monitoring data sets of each process link of the water plant during the low demand stage, medium demand stage and high demand stage of the water plant, including water quality parameters P(t), energy consumption data E(t), equipment status S(t) and external environment parameters W(t): P(t) represents the relevant parameters of the water quality of the water plant at a certain time t, including turbidity, pH value and chemical oxygen demand (COD); E(t) represents the energy consumption data of the water plant at time t, including total energy consumption and sub-item energy consumption of each process link; S(t) represents the operating status parameters of the main equipment in the water plant, including the pump pressure, filter status, and the working status of the dosing system; W(t) represents the external environmental parameters, including temperature, humidity and water quality parameters of the external water source; S22. Based on the real-time monitoring of water quality parameters, energy consumption data, equipment status and external environmental parameters of each process link, a real-time data acquisition system DCS(t) is constructed. The real-time data acquisition system independently monitors and collects data for each process link in different demand stages, and sets the acquisition frequency f of the real-time data acquisition system. Different real-time data acquisition frequencies f are set in the low demand stage, medium demand stage and high demand stage respectively. L ,f M ,f H : f L The frequency of data collection for the low demand phase; f M The frequency of data collection for the medium demand stage; f H The frequency of data collection for the high-demand phase; S24. Dynamically adjust the parameter settings of the real-time data acquisition system DCS(t) according to the monitoring requirements of each stage to build a real-time monitoring data set for the water plant: X(t)={P(t),E(t),S(t),W(t)}.

4. According to claim 1, a method for intelligent optimization and scheduling of the entire process of a water plant based on adaptive feedback control is characterized in that: The S3 includes: S31. A multi-stage evolutionary game model is constructed based on the real-time monitoring data set of the water plant, and the water pump, dosing system and filtration system in the water plant are used as game participants A. i , where i = 1, 2, 3; S32. Define each game participant A i The strategy set S i , the policy set contains different operation modes and operating parameters: For pumps, the strategy set includes different flow settings; For dosing systems, the strategy set includes different dosing amount settings; For the filtering system, the policy set includes different filter speed settings; S33. Define the payoff function U for each game participant i (S i ,S -i ), the revenue function aims at optimizing water quality compliance rate, energy consumption and cost: U i (S i ,S -i )=a i P(t)-β i E(t)-γ i C(t); Among them, P(t) represents the water quality compliance rate, E(t) represents the energy consumption of the water pump, dosing system or filtration system, C(t) represents the cost of the corresponding process link, α i , β i , γ i The weight coefficients for each participant correspond to the goals of water quality compliance rate, energy consumption and cost optimization.

5. According to claim 1, a method for intelligent optimization and scheduling of the entire process of a water plant based on adaptive feedback control is characterized in that: The local optimal scheduling scheme construction process of the current stage includes that in each operation stage, the game participants update their strategies according to the complex replication dynamic equation: Among them, x i For strategy S i The proportion of water pump, dosing system and filtration system in the game, U i (S i ,S -i ,t) is the game participant A i The profit function changes with time t, is the rate of change of water quality compliance rate over time, indicating the changing trend of water quality in the current water plant. is the rate of change of energy consumption over time, indicating the changing trend of energy consumption in each process link. is the rate of change of equipment status over time, indicating the change trend of the operating status of the water pump, dosing system and filtration system, λ i , μ i , ν i They are weight coefficients respectively: λ i Indicates process step A i Sensitivity to changes in water quality compliance rates, reflecting the impact of water pumps, dosing systems and filtration systems on water quality; μ i Indicates process step A i Sensitivity to changes in energy consumption determines regulation to minimize energy consumption; ν i Indicates process step A i The sensitivity to changes in equipment status determines the degree of impact on equipment maintenance and adjustments; is the average payoff of all game participants: Among them, n is the number of process links participating in the game.

6. According to claim 1, a method for intelligent optimization and scheduling of the entire process of a water plant based on adaptive feedback control is characterized in that: The S5 includes: S51. Using the adaptive feedback control mechanism, according to the local optimal scheduling scheme output by the evolutionary game model, the operating parameters of each process link of the water plant are adjusted in real time, including the pump flow rate Q(t), the dosage M(t) and the filtration speed V(t); S52. During the adjustment process, the real-time monitoring data set X'(t) of the water transfer plant is collected in real time, and the real-time monitoring data set X'(t) of the water transfer plant is transmitted to the control system through the feedback mechanism to update the participant benefit function U' in the multi-stage evolutionary game model i (S i ,S -i ,t), so that the multi-stage evolutionary game model can be dynamically adjusted according to the latest operating status and external environment parameters; S53. Based on the feedback data and the strategy combination output by the evolutionary game model, the process parameters are optimized and adjusted using the adaptive feedback control equation: Among them, λ1, λ2, λ3 are adjustment coefficients, which control the speed of the water pump, dosing system and filtration system to respond to the adjustment, U′ Q (t),U′ M (t),U′ V (t) are the profit functions of the water pump, dosing system and filtration system respectively, is the average payoff function of all game participants, is the rate of change of water quality parameters, is the second-order derivative of the external environment parameter, is the rate of change of energy consumption, is the square of the rate of change of equipment status parameters, α1, α2, α3 are the weights of the impact of water quality parameter changes on each link, and β1, β2, β3 are the weights of the impact of energy consumption changes on each link. S54. Through the closed-loop feedback regulation mechanism, the system dynamically optimizes the operating parameters of the process links according to the real-time monitoring data of the water plant. During the adjustment process, each process link continuously updates its profit function and control parameters to achieve the optimal dynamic regulation effect, forming an adaptive closed-loop regulation system.

7. The method for intelligent optimization and scheduling of the whole process of a water plant based on adaptive feedback control according to claim 1 is characterized in that: The S6 includes: S61. When the water plant operating conditions or external environmental parameters W(t) change, real-time monitoring of the change rate of the external environment Combined with the real-time monitoring data set of the water plant, the local optimal scheduling plan is calculated through a multi-stage evolutionary game model; S62. Dynamically adjust the operation phase of the water plant according to the calculated local optimal scheduling plan, switch between the low demand phase, the medium demand phase and the high demand phase, and update the operating parameters of the process link through the combination of participant strategies in the multi-stage evolutionary game model; S63. Use the adaptive feedback control mechanism to dynamically adjust the strategy combination of the process link according to the changes in the external environment and operating conditions. When the strategy combination of the process link changes, the control system automatically optimizes the operating parameters of each link of the water plant according to the adaptive feedback control mechanism to achieve the global optimal scheduling at each stage; S64. As the operation phase switches, the control system adjusts the profit function and strategy combination of each process link according to real-time data so that the water plant always maintains the global optimal scheduling when the external environment and operating conditions change.

8. The method for intelligent optimization and scheduling of the whole process of a water plant based on adaptive feedback control according to claim 1 is characterized in that: The S7 includes: S71. Combined with water plant historical monitoring data X hist (t) and real-time monitoring data X(t) to build a prediction model for analyzing system change trends; S72. In the prediction model, the change trend equation of water quality parameter P, energy consumption data E, water demand D and equipment operation status parameter S is calculated by time series analysis method: Among them, X pred (t+Δt) is the predicted value at the future time t+Δt, X(t) represents any water quality parameter P(t), energy consumption data E(t), water demand D(t) or equipment operation status parameter S(t), α X and β X are the weight coefficients of the corresponding parameters respectively; S74. Optimize the water plant dispatch plan based on the water quality fluctuations, energy consumption, water demand and equipment load change trends output by the prediction model, and adjust the operating parameters of each process link of the water plant before system changes occur.

Images