Intelligent water-saving building water supply and drainage system and control method thereof

Abstract

The application discloses an intelligent water-saving building water supply and drainage system and a control method thereof. The water supply and drainage system comprises a rainwater collection module, a reclaimed water collection module, a water quality monitoring module, a water storage tank module, an intelligent decision module and an electromagnetic shunt water valve device. The control method comprises real-time collection of water quality parameters, real-time collection of water demand data, decision of water dynamic shunting, control of water dynamic shunting and self-optimization of the control system. By integrating turbidity, chemical oxygen demand and pH sensors in the rainwater pipe and the reclaimed water pipe to obtain water quality data in real time, the water quality grades of rainwater and reclaimed water and water demand at different times and in different scenarios are dynamically evaluated according to a neural network model, higher shunting precision than traditional threshold control is achieved, health and safety problems such as pipeline pollution and soil heavy metal accumulation caused by direct reuse of non-traditional water sources without effective monitoring are avoided, and water supply demand at different times and in different scenarios can be effectively met.

Description

Technical Field

[0001] This invention belongs to the field of control system technology and relates to an intelligent water-saving building water supply and drainage system and its control method. Background Technology

[0002] Large buildings, such as public buildings and high-rise residential buildings, are among the main scenarios for water consumption. Water consumption in non-potable water applications, such as toilet flushing, vehicle washing, equipment cooling, and green space irrigation, accounts for a consistently high proportion of total building water consumption. Traditional building water supply and drainage systems rely on tap water to meet all water needs, leading to a significant amount of high-quality water being consumed in unnecessary scenarios and exacerbating the water supply-demand imbalance.

[0003] Currently, the collection and utilization of non-traditional water sources, such as rainwater and greywater, face the following challenges: 1) Direct reuse of non-traditional water sources without effective monitoring may lead to hygiene and safety issues or equipment damage. For example, rainwater typically contains air pollutants and roof deposits, while greywater typically contains detergents and organic matter, which may cause pollution and scaling in water supply and drainage pipes and heavy metal accumulation in green soil; 2) Traditional water supply and drainage systems rely on manual testing or simplified treatment, making it difficult to accurately assess water quality in real time; 3) The lack of a non-traditional water source allocation mechanism makes it impossible to dynamically adjust the use of water based on water quality, potentially leading to limited use or equipment blockage due to substandard water quality; 4) Traditional building water supply and drainage systems involving non-traditional water sources lack intelligent and automated management, relying on manual valve operation to switch water sources, which may result in response delays, operational errors, and an inability to dynamically adjust based on real-time water quality and water demand, leading to water waste or idle reuse systems; 5) Rainwater pollution is usually severe at the beginning of heavy rainfall. If rainwater is not differentiated by water quality and is collected in its entirety, pollutants will enter the storage device, which is detrimental to the safety and sustainability of rainwater recycling.

[0004] Faced with the contradiction between water scarcity and high water consumption in buildings, traditional building water supply and drainage systems suffer from drawbacks in utilizing non-traditional water sources such as rainwater and greywater. These shortcomings include "extensive collection, simplistic treatment, haphazard allocation, and inefficient management," leading to serious waste of high-quality water resources and high risks associated with the reuse of non-traditional water sources. This fails to meet the demands of intelligent water conservation, green building, and sustainable development. Currently, there is an urgent need to integrate water quality monitoring, intelligent diversion, and automated control technologies to construct an efficient, safe, and intelligent water resource recycling system, addressing the building water conservation problem at its source. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention proposes an intelligent water-saving building water supply and drainage system and its control method, which is applicable to the collection and utilization of non-traditional water sources such as rainwater and greywater in buildings, and helps to improve the intelligence, safety and efficiency of water-saving building water supply and drainage systems.

[0006] The first aspect of this application discloses an intelligent water-saving building water supply and drainage system, including a rainwater collection module, a greywater collection module, a water quality monitoring module, a water storage tank module, an intelligent decision-making module, and an electromagnetic diversion valve device;

[0007] The rainwater harvesting module includes a roof water collection device, a ground water collection device, and a rainwater pipe; the roof water collection device includes a roof guide slope and a roof water collection trough, used to collect and guide roof rainwater into the rainwater pipe; the ground water collection device includes a ground guide slope and a ground water collection trough, used to collect and guide ground rainwater into the rainwater pipe; the rainwater pipe is used for the connection between the roof water collection device, the ground water collection device, and the electromagnetic diversion valve device;

[0008] The greywater collection module includes a dedicated greywater interface, a greywater pipe, a check valve, and an impurity interception device. The dedicated greywater interface is used to connect to non-fecal sewage drainage points inside the building. The greywater pipe is used to connect the dedicated greywater interface and the electromagnetic diversion valve. The check valve ensures that the greywater flows in one direction within the greywater pipe. The impurity interception device uses a stainless steel filter screen to intercept impurities in the greywater.

[0009] The water quality monitoring module is arranged inside the rainwater pipe and the greywater pipe, and forms a wireless connection with the intelligent decision-making module; the water quality monitoring module includes a turbidity sensor, a chemical oxygen demand (COD) sensor, and a pH sensor; the turbidity sensor is used to acquire turbidity data of rainwater and greywater; the COD sensor is used to acquire COD data of rainwater and greywater; the pH sensor is used to acquire pH data of rainwater and greywater.

[0010] The water storage tank module includes a toilet flushing water tank, a greening irrigation water tank, a vehicle washing water tank, an equipment cooling water tank, a backup water tank, and a water level sensor. The toilet flushing water tank, the greening irrigation water tank, the vehicle washing water tank, and the equipment cooling water tank are used to meet the water source requirements for toilet flushing, greening irrigation, vehicle washing, and equipment cooling, respectively. The backup water tank is used to store rainwater and greywater whose water quality does not meet the water source requirements for toilet flushing, greening irrigation, vehicle washing, and equipment cooling. The water level sensor is installed in the toilet flushing water tank, the greening irrigation water tank, the vehicle washing water tank, the equipment cooling water tank, and the backup water tank to monitor the water level and record the current season and time, and is wirelessly connected to the intelligent decision-making module.

[0011] The electromagnetic diversion valve device is located between the rainwater pipe and the greywater pipe and the water storage tank module; the direction and opening of the water flow channel of the electromagnetic diversion valve device are controlled according to the direction and intensity of the current applied to the electromagnetic diversion valve device, thereby controlling the water flow direction and flow rate;

[0012] The intelligent decision-making module receives data from the water quality monitoring module and the water level sensor via wireless connection, and outputs the prediction results to the electromagnetic diversion valve device via wireless connection. The intelligent decision-making module adopts neural network model technology, which includes an input layer, a hidden layer, and an output layer. The input layer has 13 nodes, used to receive normalized water quality data and water demand data. The water quality data includes turbidity data, chemical oxygen demand (COD) data, and pH data of rainwater, as well as turbidity data, COD data, and pH data of reclaimed water. The water demand data includes water level data of toilet flushing tanks and greening irrigation tanks. The system includes water level data for the toilet flushing water tank, the greening irrigation water tank, the vehicle washing water tank, the equipment cooling water tank, the backup water tank, the current season data, and the current time data; the hidden layer is used for data processing and feature extraction; the output layer is used to generate prediction results, which are the amounts of water entering the toilet flushing water tank, the greening irrigation water tank, the vehicle washing water tank, the equipment cooling water tank, and the backup water tank, respectively; based on the prediction results, control quantities are further generated, which include the direction and intensity of the current applied to the electromagnetic diversion valve device; the neural network model uses historical data of the intelligent water-saving building water supply and drainage system as training data.

[0013] The second aspect of this application discloses a control method for an intelligent water-saving building water supply and drainage system, comprising the following steps:

[0014] S1. Real-time acquisition of water quality data: When rainwater flows through rainwater pipes and reclaimed water flows through reclaimed water pipes, the water quality monitoring module acquires the water quality data of rainwater and reclaimed water.

[0015] S2. Real-time acquisition of water demand data: Water demand data is acquired through water level sensors;

[0016] S3. Decision on dynamic water diversion: The normalized water quality data and water demand data are input into the intelligent decision module. The generated prediction results are the water volume entering the toilet flushing water tank, the greening irrigation water tank, the vehicle washing water tank, the equipment cooling water tank and the standby water tank respectively. Based on the prediction results, control quantities are further generated. The control quantities include the current direction and current intensity applied to the electromagnetic diversion valve device.

[0017] S4. Control of dynamic water diversion: Based on the control quantity generated by the intelligent decision module, the direction and intensity of the current applied to the electromagnetic diversion valve device are controlled, thereby controlling the direction and opening of the water flow channel of the electromagnetic diversion valve device, and further controlling the direction and flow rate of the water flow, so that the rainwater collected by the rainwater collection module or the greywater collected by the greywater collection module enters the toilet flushing water tank, the greening irrigation water tank, the vehicle washing water tank, the equipment cooling water tank, or the backup water tank.

[0018] S5. Self-optimization of the control system: During use, the intelligent water-saving building water supply and drainage system continuously collects water demand data through water level sensors. From this data, it obtains the correlation between the water demand for toilet flushing, greening irrigation, vehicle washing, and equipment cooling and the changes in water demand with the seasons and time. The system then uses the gradient descent method to update the weight parameters of the neural network model, ensuring that the neural network model achieves self-optimization based on the water demand data. This ensures that the intelligent water-saving building water supply and drainage system automatically adapts to scenarios of water quality fluctuations and changes in user water usage habits.

[0019] Preferably, in step S5, a comprehensive loss function is constructed that includes the water demand for toilet flushing, green space irrigation, vehicle washing, and equipment cooling, thereby optimizing the parameters θ of the neural network model; the comprehensive loss function L(θ) satisfies the following expression:

[0020]

[0021] Where, β i (i = 1, 2, 3, 4) are the weighting coefficients for water demand for toilet flushing, green space irrigation, vehicle washing, and equipment cooling, respectively. β i All are not less than 0, and satisfy β1+β2+β3+β4=1; Q i (i = 1, 2, 3, 4) represent the water demand over time for toilet flushing, greening irrigation, vehicle washing, and equipment cooling, respectively.

[0022] The gradient descent method is used to update the weight parameters of the neural network model. The parameters θ of the neural network model satisfy the following expression:

[0023]

[0024] Where η is the learning rate.

[0025] Compared with existing technologies, the beneficial effects of this invention are as follows: For the collection and utilization of rainwater and greywater in buildings, an intelligent water-saving building water supply and drainage system and its control method are disclosed. The system includes a rainwater collection module, a greywater collection module, a water quality monitoring module, a water storage tank module, an intelligent decision-making module, and an electromagnetic diversion valve device. The control method includes real-time acquisition of water quality parameters, real-time acquisition of water demand data, decision-making on dynamic water diversion, control of dynamic water diversion, and self-optimization of the control system. By integrating turbidity, chemical oxygen demand (COD), and pH sensors into rainwater and greywater pipes to acquire water quality data in real time, and dynamically evaluating the water quality levels of rainwater and greywater, as well as water demand at different times and in different scenarios, based on a neural network model, higher efficiency than traditional threshold control is achieved. The system achieves high diversion accuracy, avoiding hygiene and safety issues such as pipeline pollution and heavy metal accumulation in soil caused by the direct reuse of non-traditional water sources without effective monitoring. It can also effectively meet water supply needs at different times and in different scenarios. By constructing a self-optimizing mechanism based on water supply demand, a closed loop of "data acquisition-decision control-feedback optimization" is formed, ensuring that the control system automatically adapts to scenarios of water quality fluctuations and changes in user water usage habits. Through the deep integration of "sensor networks-intelligent algorithms-automated equipment", a safe, efficient, and adaptive building water resource recycling system is constructed, filling the technical gaps in traditional systems in terms of non-real-time monitoring, non-intelligent allocation, and non-closed-loop management. It provides a green, economical, standardized, and replicable solution for water-saving renovation of large public buildings and high-rise residential buildings. Attached Figure Description

[0026] Figure 1 This is a connection diagram of an intelligent water-saving building water supply and drainage system according to the present invention;

[0027] Figure 2 This is a schematic diagram of the neural network model shown in an embodiment of the present invention;

[0028] Figure 3 This is a flowchart of a control method for an intelligent water-saving building water supply and drainage system according to the present invention;

[0029] Reference numerals: 1-Rainwater harvesting module, 11-Roof water collection device, 111-Roof drainage slope, 112-Roof water collection trough, 12-Ground water collection device, 121-Ground drainage slope, 122-Ground water collection trough, 13-Rainwater pipe, 2-Grey water collection module, 21-Grey water dedicated interface, 22-Grey water pipe, 23-Check valve device, 24-Impurity interception device, 3-Water quality monitoring module, 31-Turbidity sensor, 32-Chemical oxygen demand sensor, 33-pH sensor, 4-Intelligent decision-making module, 5-Electromagnetic diversion valve device, 6-Water storage tank module, 61-Toilet flushing water storage tank, 62-Green irrigation water storage tank, 63-Vehicle washing water storage tank, 64-Equipment cooling water storage tank, 65-Spare water storage tank, 66-Water level sensor. Detailed Implementation

[0030] The embodiments of the present invention will be described in more detail below with reference to the accompanying drawings and reference numerals, so that those skilled in the art can implement them after reading this specification. It should be understood that the specific embodiments described herein are only for explaining the present invention and are not intended to limit the present invention.

[0031] The first aspect of this application discloses as follows: Figure 1-2 The intelligent water-saving building water supply and drainage system shown includes a rainwater collection module 1, a greywater collection module 2, a water quality monitoring module 3, a water storage tank module 6, an intelligent decision-making module 4, and an electromagnetic diversion valve device 5.

[0032] The rainwater collection module 1 includes a roof water collection device 11, a ground water collection device 12, and a rainwater pipe 13. The roof water collection device 11 includes a roof guide slope 111 and a roof water collection trough 112, used to collect and guide roof rainwater into the rainwater pipe 13. The ground water collection device 12 includes a ground guide slope 121 and a ground water collection trough 122, used to collect and guide ground rainwater into the rainwater pipe 13. The rainwater pipe 13 is used for the connection between the roof water collection device 11 and the ground water collection device 12 and the electromagnetic diversion valve device 5.

[0033] The greywater collection module 2 includes a greywater-specific interface 21, a greywater pipe 22, a check valve device 23, and an impurity interception device 24. The greywater-specific interface 21 is used to connect to non-fecal sewage drainage points inside the building, including drainage points for laundry water and bathing water. The greywater pipe 22 is used to connect the greywater-specific interface 21 and the electromagnetic diversion valve device 5. The check valve device 23 is used to ensure that the greywater flows unidirectionally in the greywater pipe 22. The impurity interception device 24 uses a stainless steel filter screen to intercept impurities in the greywater.

[0034] The water quality monitoring module is arranged inside the rainwater pipe and the greywater pipe, and forms a wireless connection with the intelligent decision-making module; the water quality monitoring module 3 includes a turbidity sensor 31, a chemical oxygen demand (COD) sensor 32, and a pH sensor 33; the turbidity sensor 31 is used to acquire turbidity data of rainwater and greywater; the COD sensor 32 is used to acquire COD data of rainwater and greywater; the pH sensor 33 is used to acquire pH data of rainwater and greywater;

[0035] The water storage tank module 6 includes a toilet flushing water tank 61, a greening irrigation water tank 62, a vehicle washing water tank 63, an equipment cooling water tank 64, a backup water tank 65, and a water level sensor 66. The toilet flushing water tank 61, the greening irrigation water tank 62, the vehicle washing water tank 63, and the equipment cooling water tank 64 are used to meet the water source needs for toilet flushing, greening irrigation, vehicle washing, and equipment cooling, respectively. The backup water tank 65 is used to store rainwater and medium-temperature water whose water quality does not meet the water source needs for toilet flushing, greening irrigation, vehicle washing, and equipment cooling. Water; the water level sensor 66 is installed in the toilet flushing water tank 61, the greening irrigation water tank 62, the vehicle washing water tank 63, the equipment cooling water tank 64, and the backup water tank 65, for monitoring the water level and recording the current season and time, and is wirelessly connected to the intelligent decision-making module 4; in specific implementation, the toilet flushing water tank 61, the greening irrigation water tank 62, the vehicle washing water tank 63, the equipment cooling water tank 64, and the backup water tank 65 are equipped with ultraviolet disinfection devices, which are activated periodically for sterilization;

[0036] The electromagnetic diversion valve device 5 is located between the rainwater pipe 13, the greywater pipe 22, and the water storage tank module 6; the direction and opening of the water flow channel of the electromagnetic diversion valve device 5 are controlled according to the direction and intensity of the current applied to the electromagnetic diversion valve device 5, thereby controlling the direction and flow rate of the water flow.

[0037] The intelligent decision-making module 4 receives data from the water quality monitoring module 3 and the water level sensor 66 via wireless connection, and outputs the prediction result to the electromagnetic diversion valve device 5 via wireless connection; the intelligent decision-making module 4 adopts neural network model technology, and the neural network model includes an input layer (I1, I2, ... I... 12 ), hidden layer (h1) [1] h2 [1] ,…h n [3]The system consists of an input layer (O1, O2, O3, O4) and an output layer (O4). The input layer has 13 nodes, used to receive normalized water quality data and water demand data. The water quality data includes turbidity data of rainwater, chemical oxygen demand (COD) data of rainwater, pH data of rainwater, turbidity data of reclaimed water, COD data of reclaimed water, and pH data of reclaimed water. The water demand data includes water level data of toilet flushing tanks, water level data of greening irrigation tanks, water level data of vehicle washing tanks, water level data of equipment cooling tanks, water level data of standby tanks, and water level data of the current season. The system includes section data and current time data; the hidden layer is used for data processing and feature extraction; the output layer is used to generate prediction results, which are the water volumes entering the toilet flushing water tank 61, the greening irrigation water tank 62, the vehicle washing water tank 63, the equipment cooling water tank 64, and the spare water tank 65, respectively; based on the prediction results, control quantities are further generated, which include the direction and intensity of the current applied to the electromagnetic diversion valve device 5; the neural network model uses the historical data of the intelligent water-saving building water supply and drainage system as training data.

[0038] The second aspect of this application discloses, as follows: Figure 3 The control method for an intelligent water-saving building water supply and drainage system, as shown, includes the following steps:

[0039] S1. Real-time acquisition of water quality parameters: When rainwater flows through rainwater pipe 13 and greywater flows through greywater pipe 22, the water quality data of rainwater and greywater are acquired through water quality monitoring module 3; the sampling period Δt is 60s, thereby balancing data real-time performance and system energy consumption.

[0040] S2. Real-time acquisition of water demand data: Water demand data is acquired through water level sensor 64;

[0041] S3. Decision on dynamic water diversion: The normalized water quality data and water demand data are input into the intelligent decision module 4. The generated prediction results are the water volume entering the toilet flushing water tank 61, the greening irrigation water tank 62, the vehicle washing water tank 63, the equipment cooling water tank 64, and the spare water tank 65 respectively. Based on the prediction results, control quantities are further generated. The control quantities include the direction and intensity of the current applied to the electromagnetic diversion valve device 5.

[0042] S4. Control of dynamic water diversion: Based on the control quantity generated by the intelligent decision module 4, the direction and intensity of the current applied to the electromagnetic diversion water valve device 5 are controlled, thereby controlling the direction and opening of the water flow channel of the electromagnetic diversion water valve device 5, and further controlling the direction and flow rate of the water flow, so that the rainwater collected by the rainwater collection module 1 or the greywater collected by the greywater collection module 2 enters the toilet flushing water storage tank 61, the greening irrigation water storage tank 62, the vehicle washing water storage tank 63, the equipment cooling water storage tank 64, or the backup water storage tank 65.

[0043] S5. Self-optimization of the control system: During use, the intelligent water-saving building water supply and drainage system continuously collects water demand data through a water level sensor 66. This data reveals the correlation between water demand for toilet flushing, green space irrigation, vehicle washing, and equipment cooling and seasonal and temporal variations. The system then uses gradient descent to update the weight parameters of the neural network model, ensuring self-optimization based on the water demand data. This allows the intelligent water-saving building water supply and drainage system to automatically adapt to water quality fluctuations and changes in user water usage habits. Specifically, a comprehensive loss function is constructed that includes water demand for toilet flushing, green space irrigation, vehicle washing, and equipment cooling, thereby optimizing the parameters θ of the neural network model. The comprehensive loss function L(θ) satisfies the following expression:

[0044]

[0045] Where, β i (i = 1, 2, 3, 4) are the weighting coefficients for water demand for toilet flushing, green space irrigation, vehicle washing, and equipment cooling, respectively. β i All are not less than 0, and satisfy β1+β2+β3+β4=1; Q i (i = 1, 2, 3, 4) represent the water demand over time for toilet flushing, greening irrigation, vehicle washing, and equipment cooling, respectively.

[0046] The gradient descent method is used to update the weight parameters of the neural network model. The parameters θ of the neural network model satisfy the following expression:

[0047]

[0048] Where η is the learning rate, with a value range of [0.01, 0.1], used to balance the convergence speed and accuracy of the gradient descent method.

[0049] Therefore, by integrating turbidity, chemical oxygen demand (COD), and pH sensors into rainwater and greywater pipes to acquire water quality data in real time, and dynamically assessing the water quality levels of rainwater and greywater, as well as water demand at different times and in different scenarios, based on a neural network model, higher diversion accuracy than traditional threshold control can be achieved. This avoids hygiene and safety issues such as pipe pollution and soil heavy metal accumulation caused by the direct reuse of non-traditional water sources without effective monitoring, and can effectively meet water supply needs at different times and in different scenarios. By constructing a self-optimizing mechanism based on water supply demand, a closed loop of "data acquisition - decision control - feedback optimization" is formed, ensuring that the control system automatically adapts to scenarios of water quality fluctuations and changes in user water usage habits. Through the deep integration of "sensor networks - intelligent algorithms - automated equipment," a safe, efficient, and adaptive building water resource recycling system is constructed, filling the technical gaps in traditional systems in terms of non-real-time monitoring, non-intelligent allocation, and non-closed-loop management. This provides a green, economical, standardized, and replicable solution for water-saving renovations of large public buildings and high-rise residential buildings.

[0050] The above describes one or more embodiments of the present invention in a relatively specific and detailed manner, but it should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this patent should be determined by the appended claims.

Claims

1. A control method of an intelligent water-saving building water supply and drainage system, characterized by, The application discloses a water-saving building water supply and drainage system, which comprises a rainwater collecting module, a reclaimed water collecting module, a water quality monitoring module, a water storage tank module, an intelligent decision module and an electromagnetic water diversion valve device. The rainwater collecting module comprises a roof water collecting device, a ground water collecting device and a rainwater pipe; the roof water collecting device comprises a roof water diversion slope and a roof water collecting tank, which are used for converging and guiding the roof rainwater into the rainwater pipe; the ground water collecting device comprises a ground water diversion slope and a ground water collecting tank, which are used for converging and guiding the ground rainwater into the rainwater pipe; and the rainwater pipe is used for connecting the roof water collecting device and the ground water collecting device with the electromagnetic water diversion valve device. The reclaimed water collecting module comprises a reclaimed water special interface, a reclaimed water pipe, a check valve device and a foreign matter intercepting device; the reclaimed water special interface is used for being connected with a non-excrement sewage drainage point in a building; the reclaimed water pipe is used for connecting the reclaimed water special interface with the electromagnetic water diversion valve device; the check valve device is used for ensuring the one-way flow of the reclaimed water in the reclaimed water pipe; and the foreign matter intercepting device is made of a stainless steel filter screen and is used for intercepting foreign matters in the reclaimed water. The water quality monitoring module is arranged in the rainwater pipe and the reclaimed water pipe and is wirelessly connected with the intelligent decision module; the water quality monitoring module comprises a turbidity sensor, a chemical oxygen demand sensor and a pH sensor; the turbidity sensor is used for acquiring turbidity data of the rainwater and the reclaimed water; the chemical oxygen demand sensor is used for acquiring chemical oxygen demand data of the rainwater and the reclaimed water; and the pH sensor is used for acquiring pH data of the rainwater and the reclaimed water. The water storage tank module comprises a toilet flushing water storage tank, a greening irrigation water storage tank, a vehicle washing water storage tank, a device cooling water storage tank, a standby water storage tank and water level sensors; the toilet flushing water storage tank, the greening irrigation water storage tank, the vehicle washing water storage tank and the device cooling water storage tank are respectively used for meeting the water source requirements of toilet flushing, greening irrigation, vehicle washing and device cooling; the standby water storage tank is used for storing the rainwater and the reclaimed water which do not meet the water source requirements of toilet flushing, greening irrigation, vehicle washing and device cooling; and the water level sensors are installed in the toilet flushing water storage tank, the greening irrigation water storage tank, the vehicle washing water storage tank, the device cooling water storage tank and the standby water storage tank, are used for monitoring water levels and recording the current season and time, and are wirelessly connected with the intelligent decision module. The electromagnetic water diversion valve device is located between the rainwater pipe and the reclaimed water pipe and the water storage tank module; the water flow passage direction and the water flow passage opening degree of the electromagnetic water diversion valve device are controlled according to the current direction and the current intensity applied to the electromagnetic water diversion valve device. The intelligent decision module receives the data of the water quality monitoring module and the water level sensors through wireless connection and outputs the prediction results to the electromagnetic water diversion valve device through wireless connection. The control method comprises the following steps: S1, real-time acquisition of water quality parameters: when the rainwater flows through the rainwater pipe and when the reclaimed water flows through the reclaimed water pipe, the water quality data of the rainwater and the reclaimed water are acquired through the water quality monitoring module. S2, real-time collection of water demand data: water demand data is obtained through a water level sensor; S3, decision of water dynamic distribution: the normalized water quality data and the water demand data are input into an intelligent decision module, the prediction result generated is the water quantity entering the toilet flushing water storage tank, the landscaping irrigation water storage tank, the vehicle washing water storage tank, the equipment cooling water storage tank and the standby water storage tank respectively, a control quantity is further generated according to the prediction result, the control quantity includes the current direction and the current intensity applied to the electromagnetic water distribution valve device; S4, control of water dynamic distribution: according to the control quantity generated by the intelligent decision module, the current direction and the current intensity applied to the electromagnetic water distribution valve device are controlled, so as to control the water flow passage direction and the water flow passage opening degree of the electromagnetic water distribution valve device, and further control the water flow direction and the flow rate, so that the rainwater collected by the rainwater collection module or the reclaimed water collected by the reclaimed water collection module enters the toilet flushing water storage tank, the landscaping irrigation water storage tank, the vehicle washing water storage tank, the equipment cooling water storage tank or the standby water storage tank; S5, self-optimization of the control system: the water supply and drainage system continuously collects water demand data through the water level sensor during use, obtains the correlation of the water demand of toilet flushing, landscaping irrigation, vehicle washing and equipment cooling changing with the season and time, and updates the weight parameters of the neural network model using the gradient descent method, to ensure that the neural network model realizes self-optimization according to the water demand data, and further ensures that the water supply and drainage system automatically adapts to the scene of water quality fluctuations and changes in user water usage habits; wherein a comprehensive loss function containing the water demand of toilet flushing, landscaping irrigation, vehicle washing and equipment cooling is constructed, so as to optimize the parameters of the neural network model; the comprehensive loss function θ satisfies the following expression: L θ ​​ （1） wherein, β i i =1,2,3,4) are weight coefficients of water demand of toilet flushing, green irrigation, vehicle washing and equipment cooling, respectively, β i are not less than 0, and satisfy β 1 β 2 β 3 β 4 =1; Q i i =1,2,3,4) are time-varying water demands of toilet flushing, green irrigation, vehicle washing and equipment cooling, respectively.​​​​​ The weight parameters of the neural network model are updated by using a gradient descent method, and the parameters of the neural network model θ satisfy the following expression: （2） wherein η is the learning rate.

2. The control method of the intelligent water-saving building water supply and drainage system according to claim 1, characterized in that, The intelligent decision module adopts a neural network model technology, the neural network model includes an input layer, a hidden layer and an output layer; the input layer has 13 nodes in common, for receiving the normalized water quality data and the water demand data, the water quality data includes the turbidity data of rainwater, the chemical oxygen demand data of rainwater, the pH data of rainwater, the turbidity data of reclaimed water, the chemical oxygen demand data of reclaimed water and the pH data of reclaimed water; the water demand data includes the water level data of the toilet flushing water storage tank, the water level data of the landscaping irrigation water storage tank, the water level data of the vehicle washing water storage tank, the water level data of the equipment cooling water storage tank, the water level data of the standby water storage tank, the current season data and the current time data; the hidden layer is used for data processing and feature extraction; the output layer is used for generating a prediction result, the prediction result is the water quantity entering the toilet flushing water storage tank, the landscaping irrigation water storage tank, the vehicle washing water storage tank, the equipment cooling water storage tank and the standby water storage tank respectively; a control quantity is further generated according to the prediction result, the control quantity includes the current direction and the current intensity applied to the electromagnetic water distribution valve device; the neural network model takes the past historical data of the water supply and drainage system as training data.

Images