Green building rainwater collection and recycling intelligent control system

Abstract

The application discloses a kind of green building rainwater collection and recycling intelligent control system, including acquisition module, the acquisition module is for through the sensor array of deployment in old community confluence system drainage pipe network key node.By deploying sensor array, hydrological and meteorological data are collected in real time, rainfall-runoff-pipe network load coupling prediction model is used, pipe network load and rainwater runoff are accurately predicted, realize the intelligent scheduling of drainage system, effectively deal with waterlogging risk;According to pipe network load and regulation reservoir liquid level data, dynamically generate rainwater and sewage separation control strategy and regulation reservoir variable frequency water pump control instruction, improve rainwater collection efficiency, reduce sewage treatment plant load;Based on historical water consumption data and future water demand prediction, combined with regulation reservoir storage capacity and rainfall recharge, optimize the circulating water supply strategy, reduce tap water supply, realize water-saving target.

Description

Technical Field

[0001] This invention relates to the field of pipeline design optimization technology, specifically to an intelligent control system for rainwater harvesting and recycling in green buildings. Background Technology

[0002] With the acceleration of urbanization, the concept of green building has gradually gained popularity and become an important way to solve urban environmental problems and promote sustainable development. In urban water resource management, rainwater, as a precious water resource, is of great significance for the effective collection and recycling of rainwater to alleviate urban water shortages, reduce pressure on urban drainage systems, prevent urban flooding, and improve the urban ecological environment.

[0003] However, traditional older residential communities often use combined sewer systems, where rainwater and sewage share the same pipe system. During rainfall, especially heavy downpours, this system is prone to overloading the pipes, leading to frequent combined sewer overflows (CSOs). CSOs not only carry large amounts of pollutants directly into receiving water bodies, severely polluting surface and groundwater, but also exacerbate urban flooding risks, impacting urban traffic safety and residents' daily lives.

[0004] Furthermore, traditional drainage systems lack intelligent scheduling and optimization control methods, making it difficult to dynamically adjust based on real-time rainfall, pipeline load status, and rainwater storage facility capacity. This results in low rainwater resource utilization and fails to fully realize the potential for rainwater collection and recycling. Although some existing technologies attempt to improve drainage performance by adding rainwater storage tanks and optimizing pipeline design, the lack of a systematic intelligent control strategy makes it difficult to effectively solve the problem of combined sewer overflows and rainwater resource waste. Summary of the Invention

[0005] To address the aforementioned technical problems, an intelligent control system for rainwater harvesting and recycling in green buildings is provided. This technical solution solves the problems mentioned in the background section.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: In a first aspect of the present invention, an intelligent control system for rainwater harvesting and recycling in green buildings is provided, comprising: The data acquisition module is used to collect hydrological and meteorological data of the combined sewer system in real time through a sensor array deployed at key nodes of the combined sewer system in old residential areas. The hydrological data includes the liquid level in the combined sewer inspection well, the flow rate and velocity in the pipeline, and the meteorological data includes the cumulative rainfall, the real-time rainfall intensity and the predicted rainfall duration. The data processing module is used to transmit the collected raw data to the edge computing gateway, perform data cleaning and spatiotemporal alignment processing, generate a standard time series dataset, and store it in the time series database. The output module is used to calculate the predicted values ​​of the network load and the predicted values ​​of the rainwater runoff within a preset time window based on a standard time series dataset and a pre-trained rainfall-runoff-network load coupling prediction model. The instruction generation module is used to generate rainwater and sewage diversion control strategies and variable frequency pump control instructions for the storage tank based on the predicted values ​​of pipeline load and rainwater runoff, combined with the real-time liquid level data of the current rainwater storage tank. The execution module is used to send control strategies and control commands to the field actuators, which include rainwater and sewage diversion electric gates, regulating tank inlet electric butterfly valves and variable frequency water pumps, to realize intelligent scheduling of drainage systems in old residential areas.

[0007] Preferably, the step of transmitting the collected raw data to an edge computing gateway for data cleaning and spatiotemporal alignment specifically includes: The 3σ criterion was used to remove outliers from the real-time collected liquid level and flow rate data. Data points that exceeded the mean plus or minus three standard deviations were marked as outliers and filled using linear interpolation. Using a timestamp alignment algorithm, meteorological data and pipeline hydrological data with different sampling frequencies are uniformly resampled to the same time granularity to generate a time series index; The Kalman filter algorithm is used to denoise the aligned liquid level data and output the filtered liquid level state estimate as the core data field of the standard time series dataset.

[0008] Preferably, the construction of the rainfall-runoff-pipeline load coupled prediction model specifically includes: A rainfall runoff prediction sub-model based on a long short-term memory neural network was constructed. Its input layer includes the cumulative rainfall over the past hour, soil moisture index, and the proportion of impervious area. The hidden layer consists of two layers, each with 64 neurons. The output layer provides the predicted rainfall runoff for the next 15 minutes. ; A sub-model for hydraulic simulation of pipe networks based on the one-dimensional Saint-Venant equations is constructed, and the flow continuity equation and momentum equation of combined sewer systems are discretized and solved using the finite difference method. The predicted rainwater runoff value As the upstream boundary condition of the pipeline hydraulic simulation sub-model, the filling degree of each key inspection well within the next 15 minutes is calculated iteratively by combining pipeline topology data and Manning roughness coefficient. and the risk index of combined flow overflow .

[0009] Preferably, the combined overflow risk index The calculation expression is: ; in, This represents the current pipeline flow rate. The pipeline is designed to carry the maximum water. This is the normalized value of the rate of increase in liquid level. The weight coefficients are obtained by training with historical overflow event data, and ; when At that time, it was determined to be a state of high overflow risk.

[0010] Preferably, the control commands for the variable frequency water pump in the regulating reservoir specifically include: Obtain the current liquid level of the rainwater storage tank. With the upper limit of the safe liquid level of the storage tank ; Based on the fuzzy PID control algorithm, with liquid level deviation and the rate of change of deviation As input variables, where The target liquid level is dynamically adjusted based on the pipeline load forecast. The proportional gain of the PID controller is obtained by querying the fuzzy inference rule table. Integral coefficient and differential coefficients Real-time adjustment amount ; Calculate the frequency adjustment of the variable frequency water pump The calculation formula is as follows: ; Will The frequency is superimposed on the current water pump base frequency to generate the final variable frequency water pump control command, limiting the frequency output range to between 30Hz and 50Hz.

[0011] Preferably, the generation of the rainwater and sewage separation control strategy specifically includes: Real-time reading of total flow in combined sewer systems Wastewater flow Compared with the current treatment load rate of the wastewater treatment plant ; Establish a diversion control logic matrix and set the diversion multiple. Three times the dry season flow rate; when And the pipeline network is full At that time, a "full diversion" command is generated to control the action of the rainwater and sewage diversion electric gate, switch the combined pipe to the inlet of the rainwater storage tank, and close the interceptor valve leading to the sewage treatment plant. when or At that time, a "merging and interception" command is generated to control the opening of the electric gate to the interception position, so that all sewage in the dry season enters the sewage treatment plant, and the initial rainwater and rainwater exceeding the standard enter the storage tank. The number of pulses of the stepper motor is calculated based on the difference between the current opening feedback signal and the target opening of the electric gate, and the gate drive signal is generated.

[0012] Preferably, it also includes an initial rainwater diversion control step: An online water quality monitoring instrument is installed at the inlet of the rainwater storage tank to monitor the chemical oxygen demand (COD) concentration and turbidity (NTU) of the incoming water in real time. A model for initial rainwater pollutant runoff was constructed, and the change curve of COD concentration with rainfall duration was fitted using an exponential decay function: ; in, Rainfall duration COD concentration at time 10:00 The maximum initial concentration, The scouring attenuation coefficient, Background concentration; Set the flow rejection threshold concentration When real-time monitoring At that time, a diversion command is generated to control the opening of the initial rainwater electric valve, discharging the rainwater into the sewage pipe network; when If the duration exceeds 5 minutes, a switching command is generated to close the diversion valve and open the inlet valve of the storage tank, so that the later cleaner rainwater is collected into the storage tank.

[0013] Preferably, it also includes a recycling optimization step based on water demand: Collect historical water usage data for green buildings, including water usage for greening irrigation, road spraying, and toilet flushing, to form a historical water usage characteristic sequence; The ARIMA(p,d,q) time series forecasting model is used to predict water demand for the next 24 hours. The input variables include historical water consumption, day type (weekday / weekend) and weather characteristics. The output is the hourly water demand forecast curve for the next 24 hours. Obtain the current water storage capacity of the regulating reservoir. With predicted rainfall replenishment ; Construct water balance constraint equations: ; in, For the first Planned water supply per hour This refers to the dead water level volume of the storage tank. Using the minimization of tap water replenishment as the objective function, a linear programming solver is used to calculate the optimal water supply strategy and generate the start-up and shutdown schedule of the circulating water supply pump.

[0014] Preferably, the generation of the rainwater and sewage diversion control strategy and the variable frequency pump control command for the storage tank further includes a multi-objective collaborative optimization process: Define the system operating cost function J, whose expression is: ; in, For water pump energy consumption data, For the predicted overflow pollution load, To improve the utilization rate of the storage tank volume, These are weighting coefficients; An improved non-dominated sorting genetic algorithm was used to optimize the control strategy. The decision variables included the target upper limit of the storage tank liquid level and the interception ratio. and water pump start / stop thresholds; Set the constraint as the pipeline fullness. And the liquid level in the regulating tank ; The Pareto optimal solution set is obtained through iterative calculation. The optimal combination of control parameters is selected according to the current rainfall intensity level, and the current control strategy is updated.

[0015] In a second aspect of the invention, an electronic device is also provided. The electronic device includes at least one processor; and a memory communicatively connected to the at least one processor; the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the system of the first aspect of the invention.

[0016] Compared with existing technologies, this invention provides an intelligent control system for rainwater harvesting and recycling in green buildings, which has the following beneficial effects: This invention utilizes a sensor array to collect real-time hydrological and meteorological data from the pipe network. A coupled prediction model of rainfall, runoff, and pipe network load accurately predicts pipe network load and rainwater runoff, enabling intelligent scheduling of the drainage system and effectively addressing the risk of urban flooding. Based on pipe network load and storage tank level data, it dynamically generates rainwater and sewage separation control strategies and control commands for the storage tank's variable frequency pumps, improving rainwater collection efficiency and reducing the load on sewage treatment plants. An online water quality monitoring instrument is installed, and an initial rainwater pollutant flushing model is constructed to achieve automatic initial rainwater diversion and subsequent clean rainwater collection, ensuring the quality of reclaimed water. Based on historical water usage data and future water demand predictions, combined with storage tank capacity and rainfall replenishment, the circulating water supply strategy is optimized to reduce tap water replenishment and achieve water conservation goals. Furthermore, by defining a system operating cost function, a non-dominated sorting genetic algorithm is used to perform multi-objective optimization of the control strategy, improving the overall operational efficiency of the system. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the system framework structure in this invention; Figure 2 This is a schematic diagram of the method flow for S101-S103 in this invention; Figure 3 This is a schematic diagram of the method flow for S201-S203 in this invention; Figure 4 This is a schematic diagram of the method flow for S301-S305 in this invention. Detailed Implementation

[0018] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0019] Example 1 Please refer to Figure 1 As shown, in a first aspect of the present invention, a smart control system for rainwater harvesting and recycling in green buildings is provided, comprising: The data acquisition module is used to collect real-time hydrological and meteorological data of the combined sewer system by using sensor arrays deployed at key nodes of the combined sewer system in old residential areas. The hydrological data includes the liquid level in the combined sewer inspection well, the flow rate and velocity in the pipeline, and the meteorological data includes the cumulative rainfall, real-time rainfall intensity and predicted rainfall duration. The data processing module transmits the collected raw data to the edge computing gateway for data cleaning and spatiotemporal alignment, generates a standard time-series dataset, and stores it in the time-series database. The output module is used to calculate the predicted values ​​of the network load and the predicted values ​​of the rainwater runoff within a preset time window based on a standard time series dataset and a pre-trained rainfall-runoff-network load coupling prediction model. The instruction generation module is used to generate rainwater and sewage diversion control strategies and variable frequency pump control instructions for the storage tank based on the predicted values ​​of pipeline load and rainwater runoff, combined with the real-time liquid level data of the current rainwater storage tank. The execution module is used to send control strategies and control commands to the field actuators, which include rainwater and sewage diversion electric gates, regulating tank inlet electric butterfly valves, and variable frequency water pumps, to realize intelligent scheduling of drainage systems in old residential areas.

[0020] As will be understood by those skilled in the art, this invention utilizes a sensor array to collect real-time hydrological and meteorological data from the pipe network, and employs a rainfall-runoff-pipe network load coupling prediction model to accurately predict pipe network load and rainwater runoff, thereby achieving intelligent scheduling of the drainage system and effectively addressing the risk of urban flooding. Based on pipe network load and storage tank level data, it dynamically generates rainwater and sewage separation control strategies and storage tank variable frequency pump control commands, improving rainwater collection efficiency and reducing the load on sewage treatment plants. It also sets up an online water quality monitoring instrument and constructs an initial rainwater pollutant flushing model to achieve automatic initial rainwater diversion and subsequent clean rainwater collection, ensuring the quality of reclaimed water. Based on historical water usage data and future water demand predictions, combined with storage tank capacity and rainfall replenishment, it optimizes the circulating water supply strategy, reducing tap water replenishment and achieving water conservation goals. Furthermore, by defining a system operating cost function and using a non-dominated sorting genetic algorithm to perform multi-objective optimization of the control strategy, it improves the overall operational efficiency of the system.

[0021] Please refer to Figure 2 As shown, the collected raw data is transmitted to the edge computing gateway for data cleaning and spatiotemporal alignment processing, specifically including: S101. The 3σ criterion is used to remove outliers from the real-time collected liquid level and flow rate data. Data points that exceed the mean plus or minus three times the standard deviation are marked as outliers and filled using linear interpolation. S102. Using a timestamp alignment algorithm, meteorological data and pipeline hydrological data with different sampling frequencies are uniformly resampled to the same time granularity to generate a time series index. S103. Based on the Kalman filter algorithm, the aligned liquid level data is denoised, and the filtered liquid level state estimate is output as the core data field of the standard time series dataset.

[0022] Please refer to Figure 3 As shown, the construction of the rainfall-runoff-pipeline load coupled prediction model specifically includes: S201. Construct a rainfall runoff prediction sub-model based on a long short-term memory neural network. Its input layer includes the cumulative rainfall, soil moisture index, and impervious area ratio of the past hour. The hidden layer contains two layers, each with 64 neurons. The output layer is the predicted rainfall runoff value for the next 15 minutes. ; S202. Construct a sub-model for hydraulic simulation of pipe networks based on the one-dimensional Saint-Venant equations, and use the finite difference method to discretize and solve the continuity equation and momentum equation of the combined sewer system. S203, Predicted stormwater runoff value As the upstream boundary condition of the pipeline hydraulic simulation sub-model, the filling degree of each key inspection well within the next 15 minutes is calculated iteratively by combining pipeline topology data and Manning roughness coefficient. and the overflow risk index of the combined system .

[0023] Combined Flow Overflow Risk Index The calculation expression is: ; in, The current pipeline flow rate, The pipeline is designed to carry the maximum water. This is the normalized value of the rate of increase in liquid level. The weight coefficients are obtained by training with historical overflow event data, and ; when At that time, it was determined to be a state of high overflow risk.

[0024] The control commands for the variable frequency pumps in the regulating reservoir specifically include: Obtain the current liquid level of the rainwater storage tank. With the upper limit of the safe liquid level of the storage tank ; Based on the fuzzy PID control algorithm, with liquid level deviation and the rate of change of deviation As input variables, where The target liquid level is dynamically adjusted based on the pipeline load forecast. The proportional gain of the PID controller is obtained by querying the fuzzy inference rule table. Integral coefficient and differential coefficients Real-time adjustment amount ; Calculate the frequency adjustment of the variable frequency water pump The calculation formula is as follows: ; Will The frequency is superimposed on the current water pump base frequency to generate the final variable frequency water pump control command, limiting the frequency output range to between 30Hz and 50Hz.

[0025] Please refer to Figure 4 As shown, the generated rainwater and sewage separation control strategy specifically includes: S301, Real-time reading of total flow in combined sewer network Wastewater flow Compared with the current treatment load rate of the wastewater treatment plant ; S302. Establish the diversion control logic matrix and set the diversion multiple. This is three times the flow rate during the dry season; S303, when And the pipeline network is full At that time, a "full diversion" command is generated to control the action of the rainwater and sewage diversion electric gate, switch the combined pipe to the inlet of the rainwater storage tank, and close the interceptor valve leading to the sewage treatment plant. S304, when or At that time, a "merging and interception" command is generated to control the opening of the electric gate to the interception position, so that all sewage in the dry season enters the sewage treatment plant, and the initial rainwater and rainwater exceeding the standard enter the storage tank. S305. Based on the difference between the current opening feedback signal and the target opening of the electric gate, calculate the number of pulses of the stepper motor and generate the gate drive signal.

[0026] It also includes initial rainwater diversion control steps: An online water quality monitoring instrument is installed at the inlet of the rainwater storage tank to monitor the chemical oxygen demand (COD) concentration and turbidity (NTU) of the incoming water in real time. A model for initial rainwater pollutant runoff was constructed, and the change curve of COD concentration with rainfall duration was fitted using an exponential decay function: ; in, Rainfall duration COD concentration at time 10:00 The maximum initial concentration, The scouring attenuation coefficient is the coefficient of friction. Background concentration; Set the flow rejection threshold concentration When real-time monitoring At that time, a diversion command is generated to control the opening of the initial rainwater electric valve, discharging the rainwater into the sewage pipe network; when If the duration exceeds 5 minutes, a switching command is generated to close the diversion valve and open the inlet valve of the storage tank, so that the later cleaner rainwater is collected into the storage tank.

[0027] It also includes recycling optimization steps based on water demand: Collect historical water usage data for green buildings, including water usage for greening irrigation, road spraying, and toilet flushing, to form a historical water usage characteristic sequence; The ARIMA(p,d,q) time series forecasting model is used to predict water demand for the next 24 hours. The input variables include historical water consumption, day type (weekday / weekend) and weather characteristics. The output is the hourly water demand forecast curve for the next 24 hours. Obtain the current water storage capacity of the regulating reservoir. With predicted rainfall replenishment ; Construct water balance constraint equations: ; in, For the first Planned water supply per hour This refers to the dead water level volume of the storage tank. Using the minimization of tap water replenishment as the objective function, a linear programming solver is used to calculate the optimal water supply strategy and generate the start-up and shutdown schedule of the circulating water supply pump.

[0028] The generation of rainwater and sewage separation control strategies and variable frequency pump control instructions for the storage tank also includes a multi-objective collaborative optimization process: Define the system operating cost function J, whose expression is: ; in, For water pump energy consumption data, For the predicted overflow pollution load, To improve the utilization rate of the storage tank volume, These are weighting coefficients; An improved non-dominated sorting genetic algorithm was used to optimize the control strategy. The decision variables included the target upper limit of the storage tank liquid level and the interception ratio. and water pump start / stop thresholds; Set the constraint as the pipeline fullness. And the liquid level in the storage tank ; The Pareto optimal solution set is obtained through iterative calculation. The optimal combination of control parameters is selected according to the current rainfall intensity level, and the current control strategy is updated.

[0029] In a second aspect of the invention, an electronic device is also provided. The electronic device includes at least one processor; and a memory communicatively connected to the at least one processor; the memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to perform the system of the first aspect of the invention.

[0030] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. An intelligent control system for rainwater harvesting and recycling in green buildings, characterized in that, include: The data acquisition module is used to collect hydrological and meteorological data of the combined sewer system in real time through a sensor array deployed at key nodes of the combined sewer system in old residential areas. The hydrological data includes the liquid level in the combined sewer inspection well, the flow rate and velocity in the pipeline, and the meteorological data includes the cumulative rainfall, the real-time rainfall intensity and the predicted rainfall duration. The data processing module is used to transmit the collected raw data to the edge computing gateway, perform data cleaning and spatiotemporal alignment processing, generate a standard time series dataset, and store it in the time series database. The output module is used to calculate the predicted values ​​of the network load and the predicted values ​​of the rainwater runoff within a preset time window based on a standard time series dataset and a pre-trained rainfall-runoff-network load coupling prediction model. The instruction generation module is used to generate rainwater and sewage diversion control strategies and variable frequency pump control instructions for the storage tank based on the predicted values ​​of pipeline load and rainwater runoff, combined with the real-time liquid level data of the current rainwater storage tank. The execution module is used to send control strategies and control commands to the field actuators, which include rainwater and sewage diversion electric gates, regulating tank inlet electric butterfly valves and variable frequency water pumps, to realize intelligent scheduling of drainage systems in old residential areas.

2. The intelligent control system for rainwater harvesting and recycling in green buildings according to claim 1, characterized in that, The process of transmitting the collected raw data to the edge computing gateway for data cleaning and spatiotemporal alignment specifically includes: The 3σ criterion was used to remove outliers from the real-time collected liquid level and flow rate data. Data points that exceeded the mean plus or minus three standard deviations were marked as outliers and filled using linear interpolation. Using a timestamp alignment algorithm, meteorological data and pipeline hydrological data with different sampling frequencies are uniformly resampled to the same time granularity to generate a time series index; The Kalman filter algorithm is used to denoise the aligned liquid level data and output the filtered liquid level state estimate as the core data field of the standard time series dataset.

3. The intelligent control system for rainwater harvesting and recycling in green buildings according to claim 2, characterized in that, The construction of the rainfall-runoff-pipeline load coupled prediction model specifically includes: A rainfall runoff prediction sub-model based on a long short-term memory neural network was constructed. Its input layer includes the cumulative rainfall over the past hour, soil moisture index, and the proportion of impervious area. The hidden layer consists of two layers, each with 64 neurons. The output layer provides the predicted rainfall runoff for the next 15 minutes. ; A sub-model for hydraulic simulation of pipe networks based on the one-dimensional Saint-Venant equations is constructed, and the flow continuity equation and momentum equation of combined sewer systems are discretized and solved using the finite difference method. The predicted rainwater runoff value As the upstream boundary condition of the pipeline hydraulic simulation sub-model, the filling degree of each key inspection well within the next 15 minutes is calculated iteratively by combining pipeline topology data and Manning roughness coefficient. and the risk index of combined flow overflow .

4. The intelligent control system for rainwater harvesting and recycling in green buildings according to claim 3, characterized in that, The combined overflow risk index The calculation expression is: ； in, This represents the current pipeline flow rate. The pipeline is designed to carry the maximum water. This is the normalized value of the rate of increase in liquid level. The weight coefficients are obtained by training with historical overflow event data, and ; when At that time, it was determined to be a state of high overflow risk.

5. The intelligent control system for rainwater harvesting and recycling in green buildings according to claim 4, characterized in that, The control commands for the variable frequency water pumps in the storage tank specifically include: Obtain the current liquid level of the rainwater storage tank. With the upper limit of the safe liquid level of the storage tank ; Based on the fuzzy PID control algorithm, with liquid level deviation and the rate of change of deviation As input variables, where The target liquid level is dynamically adjusted based on the pipeline load forecast. The proportional gain of the PID controller is obtained by querying the fuzzy inference rule table. Integral coefficient and differential coefficients Real-time adjustment amount ; Calculate the frequency adjustment of the variable frequency water pump The calculation formula is as follows: ； Will The frequency is superimposed on the current water pump base frequency to generate the final variable frequency water pump control command, limiting the frequency output range to between 30Hz and 50Hz.

6. The intelligent control system for rainwater harvesting and recycling in green buildings according to claim 5, characterized in that, The aforementioned strategy for generating rainwater and sewage separation control specifically includes: Real-time reading of total flow in combined sewer systems Wastewater flow Compared with the current treatment load rate of the wastewater treatment plant ; Establish a diversion control logic matrix and set the diversion multiple. Three times the dry season flow rate; when And the pipeline network is full At that time, a "full diversion" command is generated to control the action of the rainwater and sewage diversion electric gate, switch the combined pipe to the inlet of the rainwater storage tank, and close the interceptor valve leading to the sewage treatment plant; when or At that time, a "merging and interception" command is generated to control the opening of the electric gate to the interception position, so that all sewage in the dry season enters the sewage treatment plant, and the initial rainwater and rainwater exceeding the standard enter the storage tank. The number of pulses of the stepper motor is calculated based on the difference between the current opening feedback signal and the target opening of the electric gate, and the gate drive signal is generated.

7. The intelligent control system for rainwater harvesting and recycling in green buildings according to claim 6, characterized in that, It also includes initial rainwater diversion control steps: An online water quality monitoring instrument is installed at the inlet of the rainwater storage tank to monitor the chemical oxygen demand (COD) concentration and turbidity (NTU) of the incoming water in real time. A model for initial rainwater pollutant runoff was constructed, and the change curve of COD concentration with rainfall duration was fitted using an exponential decay function: ； in, Rainfall duration COD concentration at time 10:00 The maximum initial concentration, The scouring attenuation coefficient is the coefficient of friction. Background concentration; Set the flow rejection threshold concentration When real-time monitoring At that time, a diversion command is generated to control the opening of the initial rainwater electric valve, discharging the rainwater into the sewage pipe network; when If the duration exceeds 5 minutes, a switching command is generated to close the diversion valve and open the inlet valve of the storage tank, so that the later cleaner rainwater is collected into the storage tank.

8. The intelligent control system for rainwater harvesting and recycling in green buildings according to claim 7, characterized in that, It also includes recycling optimization steps based on water demand: Collect historical water usage data for green buildings, including water usage for greening irrigation, road spraying, and toilet flushing, to form a historical water usage characteristic sequence; The ARIMA(p,d,q) time series forecasting model is used to predict water demand for the next 24 hours. The input variables include historical water consumption, day type (weekday / weekend) and weather characteristics. The output is the hourly water demand forecast curve for the next 24 hours. Obtain the current water storage capacity of the regulating reservoir. With predicted rainfall replenishment ; Construct water balance constraint equations: ； in, For the first Planned water supply per hour This refers to the dead water level volume of the storage tank. Using the minimization of tap water replenishment as the objective function, a linear programming solver is used to calculate the optimal water supply strategy and generate the start-up and shutdown schedule of the circulating water supply pump.

9. The intelligent control system for rainwater harvesting and recycling in green buildings according to claim 8, characterized in that, The generated rainwater and sewage separation control strategy and the variable frequency pump control command for the storage tank also include a multi-objective collaborative optimization process: Define the system operating cost function J, whose expression is: ； in, For water pump energy consumption data, For the predicted overflow pollution load, To improve the utilization rate of the storage tank volume, These are weighting coefficients; An improved non-dominated sorting genetic algorithm was used to optimize the control strategy. The decision variables included the target upper limit of the storage tank liquid level and the interception ratio. and water pump start / stop thresholds; Set the constraint as the pipeline fullness. And the liquid level in the regulating tank ; The Pareto optimal solution set is obtained through iterative calculation. The optimal combination of control parameters is selected according to the current rainfall intensity level, and the current control strategy is updated.

10. An electronic device, comprising at least one processor; and a memory communicatively connected to said at least one processor; characterized in that, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the system according to any one of claims 1-9.

Images