A multi-terminal cooperative central hot water control method and system based on artificial intelligence

Abstract

This invention relates to the field of hot water control technology, and more particularly to a multi-terminal collaborative central hot water control method and system based on artificial intelligence. The system includes a sensor acquisition layer, an edge control layer, an equipment execution layer, a cloud AI layer, and a terminal layer. The terminal layer includes a local human-machine interface terminal and a mobile remote terminal. The local human-machine interface terminal is used for setting on-site parameters and issuing manual control commands. The mobile remote terminal is used for remotely scheduling water usage, setting parameters, and receiving status updates. Through the structure of the sensor acquisition layer, edge control layer, equipment execution layer, cloud AI layer, and terminal layer, this invention enables the multi-terminal collaborative central hot water control method and system to significantly improve the efficiency of multi-heat source collaborative scheduling and energy utilization, the resolution of multi-terminal command conflicts and the consistency of status synchronization, and the proactive load prediction capability based on water usage behavior through AI pre-scheduling, conflict arbitration, and closed-loop control of multi-terminal collaboration.

Description

Technical Field

[0001] This invention relates to the field of hot water control technology, specifically to a multi-terminal collaborative central hot water control method and system based on artificial intelligence. Background Technology

[0002] Central hot water systems are widely used in hotels, hospitals, schools, and large residences to centrally supply domestic hot water. With the development of control technology, existing central hot water control systems have gradually evolved from early manual control to automatic control based on PLC or microcontroller. Currently, most systems use temperature threshold triggering to control the start and stop of each heat source and circulating water pump, and can achieve parameter setting and remote monitoring through local touch screens or mobile APPs, which improves the automation level of the system to a certain extent.

[0003] However, existing technologies still have the following shortcomings: On the one hand, each heat source and actuator usually operates independently based on a fixed temperature threshold, lacking a collaborative pre-scheduling mechanism based on user water usage behavior. This results in low utilization of low-grade energy sources such as solar energy, high energy consumption due to frequent start-stop of electric heaters, and users needing to wait for heating before using water. On the other hand, with the widespread use of various control entry points such as local control panels and mobile remote terminals, command conflicts are prone to occur when multiple terminals operate concurrently. Existing systems lack an effective conflict arbitration mechanism, and the status displays of each terminal are not synchronized, affecting control reliability. In addition, existing systems mostly rely on passive temperature feedback control and cannot perform load forecasting and heat source planning in advance based on historical water usage data, resulting in insufficient intelligence. Therefore, based on the above problems, a multi-terminal collaborative central hot water control method and system based on artificial intelligence is proposed. Summary of the Invention

[0004] The purpose of this invention is to provide a multi-terminal collaborative central hot water control method and system based on artificial intelligence, so as to solve the problems of high energy consumption when each heat source operates independently, easy conflict and asynchronous state of multi-terminal operation, and lack of active load prediction of water use behavior in existing systems.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A multi-terminal collaborative central hot water control system based on artificial intelligence includes a sensor acquisition layer, an edge control layer, an equipment execution layer, a cloud AI layer, and a terminal layer; The terminal layer includes a local human-machine interaction terminal and a mobile remote terminal; the local human-machine interaction terminal is used for setting on-site parameters and issuing manual control commands; the mobile remote terminal is used for remotely scheduling water use, setting parameters, and receiving status push notifications. The sensing and acquisition layer includes a dual-tank temperature sensor group, a water level sensor, a flow sensor, and an electrical parameter acquisition module, which is used to collect data on the water temperature, water level, water flow rate of each circuit, and electrical parameters of the relay circuit in the dual-tank water in real time, and upload them to the edge control layer. The edge control layer includes an edge control node, which includes a data aggregation and preprocessing module, a conflict arbitration module, an instruction execution module, a state synchronization module, and a communication module. The local human-computer interaction terminal is connected to the edge control node through a local communication bus. The mobile remote terminal accesses the wireless network through the communication module and establishes a communication connection with the cloud AI layer. The data aggregation and preprocessing module is used to aggregate the sensor data uploaded by the sensor acquisition layer and the user setting data uploaded by the terminal layer, perform feature extraction and preprocessing, and upload the preprocessed feature data to the cloud AI layer through the communication module; the conflict arbitration module is used to receive and cache the control commands issued by each terminal and the pre-scheduling strategy issued by the cloud AI layer, and perform conflict arbitration on the concurrent control commands of multiple terminals and the pre-scheduling strategy according to the preset priority matrix to generate a unique execution command; the command execution module is used to output multiple drive control signals according to the arbitrated execution command; the status synchronization module is used to synchronously feed back the real-time operating status of the system to the local human-machine interaction terminal through the local communication bus, and synchronously feed back to the mobile remote terminal through the communication module; the communication module is used to access the wireless network to realize bidirectional data communication between the edge control node, the cloud AI layer and the mobile remote terminal; The device execution layer includes a water pump, a solar circulating pump, a pipeline return water pump, an electric heater, a constant temperature water outlet pump, a constant temperature water return pump, a cooling water valve, and an antifreeze heating cable; each device in the device execution layer is communicatively connected to the instruction execution module, receives the multi-channel drive control signals, and executes corresponding actions; The cloud AI layer includes a cloud AI management terminal, which has a built-in time series prediction model and a multi-objective optimization algorithm. It is used to receive preprocessed feature data uploaded by the edge control layer, perform user water use behavior learning and load prediction, generate a multi-heat source collaborative pre-scheduling strategy, and send it to the edge control layer through the communication module.

[0006] Preferably, the priority matrix of the conflict arbitration module is: local emergency stop / manual operation command > remote active operation command > cloud pre-scheduling strategy command; the conflict arbitration module maintains a timestamped command queue, and when a conflict is detected between multiple terminal concurrent control commands and the pre-scheduling strategy within a preset time window, arbitration is performed according to the priority matrix.

[0007] Preferably, the time-series prediction model is based on the LSTM or Prophet algorithm, generating a future water load prediction curve based on historical water usage time, water consumption, set temperature, and ambient temperature data; the multi-objective optimization algorithm takes the lowest operating energy consumption and the highest water comfort as dual objectives, and generates the multi-heat source collaborative pre-scheduling strategy based on the future water load prediction curve combined with the current solar collector temperature, water tank temperature, and ambient temperature. The multi-heat source collaborative pre-scheduling strategy includes the solar circulating pump pre-start period, the electric heater preheating period and power, and the pipeline return water pump circulation period.

[0008] Preferably, the dual-tank temperature sensor group includes a main water tank temperature sensor, an auxiliary water tank temperature sensor, a solar collector temperature sensor, and an ambient temperature sensor; the water level sensor includes a main water tank water level sensor and an auxiliary water tank water level sensor; the electrical parameter acquisition module is used to acquire current and voltage data from each output circuit of the instruction execution module.

[0009] Preferably, the instruction execution module includes 8 relay outputs, which respectively drive the water pump, the solar circulation pump, the pipeline return water pump, the electric heater, the constant temperature water pump, the constant temperature water return pump, the cold water valve, and the antifreeze heating cable.

[0010] Preferably, the communication module is a wireless communication module integrating a QR code network configuration unit and a WiFi / 4G communication unit; the mobile remote terminal completes device binding by scanning the QR code network configuration unit, and interacts with the cloud AI management terminal through the WiFi / 4G communication unit.

[0011] Preferably, the local human-machine interface terminal includes a touch screen and a physical button panel. The touch screen is used to display the water temperature and level of the dual water tanks and the operating status of each device. The physical button panel is used for local emergency stop and manual operation. The mobile remote terminal is a smartphone or WeChat mini-program terminal equipped with a central hot water control APP.

[0012] Preferably, a control method includes the following steps: S1. Multi-terminal data aggregation and water usage behavior learning: The local human-machine interface terminal collects and records the target water temperature, reserved water usage time period, and mode selection parameters input by the user on-site via a touch screen or physical button panel, forming local setting data; the mobile remote terminal uploads the remote reserved water usage data and parameter adjustment commands input by the user in the mobile APP or WeChat mini-program to the edge control node through the communication module; the data aggregation and preprocessing module of the edge control node receives the above-mentioned local setting data and remote reserved water usage data, and simultaneously receives the dual-tank temperature sensor group, water level sensor, flow sensor, and electrical sensor from the sensing acquisition layer. The parameter acquisition module uploads real-time data on the water temperature and level of the dual water tanks, the water flow rate of each circuit, and the electrical parameters of the relay circuit. The data aggregation and preprocessing module cleans, aligns, and extracts features from the above multi-source heterogeneous data to generate structured preprocessed feature data containing user water usage time, water consumption, set temperature, ambient temperature, and system operating status. This data is then uploaded to the cloud AI management terminal via the communication module. The time-series prediction model built into the cloud AI management terminal receives the structured preprocessed feature data, trains and learns based on historical water usage behavior, identifies user water usage patterns, and generates a water load prediction curve for future periods. S2, AI multi-heat source pre-scheduling decision-making: The cloud-based AI management terminal calls a multi-objective optimization algorithm with the dual objective functions of minimizing operating energy consumption and maximizing water comfort. It performs joint calculations on the future water load prediction curve generated by S1 and the current solar collector temperature, main water tank temperature, auxiliary water tank temperature, and ambient temperature. The multi-objective optimization algorithm outputs a multi-heat source collaborative pre-scheduling strategy, which includes the pre-start time and running duration of the solar circulating pump, the preheating start time and heating power of the electric heater, the circulation time of the pipeline return water pump, and the coordinated start-stop sequence of the constant temperature outlet water pump and the constant temperature return water pump. The cloud-based AI management terminal distributes the above pre-scheduling strategy to the edge control node through the communication module. S3. Intelligent Arbitration of Multi-Terminal Command Conflicts: The conflict arbitration module of the edge control node establishes a timestamped command queue, which listens for and caches control commands from three sources in real time: manual control commands issued by the local human-machine interaction terminal through the local communication bus, remote active operation commands issued by the mobile remote terminal through the communication module, and pre-scheduled strategy commands issued by the cloud AI management terminal through the communication module. The conflict arbitration module detects whether the commands from the above three sources exist simultaneously within a preset time window and whether there is a control conflict for the same target device. If a conflict is detected, arbitration is performed according to a priority matrix, which is in the following order from high to low: local emergency stop / manual operation command > remote active operation command > cloud pre-scheduled strategy command. The conflict arbitration module outputs the only execution command after arbitration to the command execution module, and the commands that are not adopted are returned to the corresponding terminal with a conflict reason prompt. S4. Edge Execution and State Synchronization: The instruction execution module of the edge control node receives the unique execution instruction output by S3, parses it into the corresponding switch control signals output by the 8 relays, and drives the water pump, solar circulation pump, pipeline return water pump, electric heater, constant temperature water outlet pump, constant temperature water return pump, cold water supply valve, and antifreeze heating cable to perform corresponding actions. During execution, the sensing acquisition layer continuously feeds back real-time operating parameters to the edge control node. The state synchronization module of the edge control node divides the real-time operating status of the system into two synchronous outputs: one pushes the water temperature, water level, and operating status of each device in the dual water tanks to the touch screen of the local human-machine interaction terminal for local display through the local communication bus; the other pushes the same operating status to the mobile remote terminal through the communication module using the MQTT protocol to achieve consistency of status across multiple terminals.

[0013] Compared with the prior art, the beneficial effects of the present invention are: In this invention, a structure consisting of a sensor acquisition layer, an edge control layer, a device execution layer, a cloud AI layer, and a terminal layer enables the sensor acquisition layer to achieve real-time sensing and data uploading of the water temperature, water level, flow rate of each circuit, and electrical parameters of the relay circuit in both water tanks through a dual-tank temperature sensor group, water level sensor, flow sensor, and electrical parameter acquisition module. The edge control layer aggregates, cleans, and extracts features from multi-source heterogeneous sensor data and multi-terminal user setting data through a data aggregation and preprocessing module. A conflict arbitration module, based on a priority matrix that prioritizes local emergency stop / manual operation commands over remote active operation commands, and remote active operation commands over cloud pre-scheduling strategy commands, arbitrates conflicts between concurrent control commands from multiple terminals and the cloud pre-scheduling strategy, generating a unique execution command. The command execution module outputs multiple drive control signals. A status synchronization module, using a local communication bus and communication module with the MQTT protocol, pushes status synchronization information to the local human-machine interface terminal and the mobile remote terminal. The communication module also enables communication between the edge control node and the cloud AI layer and the mobile terminal. The system features two-way data interaction between remote terminals; the cloud-based AI layer, through its built-in time-series prediction model and multi-objective optimization algorithm, performs load prediction and multi-heat source collaborative pre-scheduling decisions based on historical water usage behavior, generating pre-scheduling strategies that include the start-up and shutdown times and power allocation of each heat source and distributing them to the edge control layer; the terminal layer provides multi-entry interaction capabilities for on-site control and remote reservation through local human-machine interaction terminals and mobile remote terminals respectively; and the equipment execution layer receives drive signals and executes corresponding actions through water pumps, solar circulating pumps, pipeline return water pumps, electric heaters, constant temperature water pumps, constant temperature water return pumps, cold water valves, and anti-freeze heating cables. This achieves a closed-loop control method and system for multi-terminal collaborative central hot water control based on artificial intelligence. Through AI pre-scheduling, conflict arbitration, and multi-terminal collaboration, it significantly improves the collaborative scheduling and energy utilization efficiency of multiple heat sources, the resolution of multi-terminal command conflicts and the consistency of state synchronization, as well as the proactive load prediction capability based on water usage behavior. This solves the problems of high energy consumption when each heat source operates independently, easy conflict and asynchronous state of multi-terminal operation, and lack of proactive load prediction based on water usage behavior in existing systems. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the overall system architecture of the present invention; Figure 2 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0015] Please see Figure 1-2 The present invention provides a technical solution: A multi-terminal collaborative central hot water control system based on artificial intelligence includes a sensor acquisition layer, an edge control layer, an equipment execution layer, a cloud AI layer, and a terminal layer; The terminal layer includes a local human-machine interaction terminal and a mobile remote terminal; the local human-machine interaction terminal is used for setting on-site parameters and issuing manual control commands; the mobile remote terminal is used for remotely scheduling water use, setting parameters, and receiving status push notifications. The sensing and acquisition layer includes a dual-tank temperature sensor group, a water level sensor, a flow sensor, and an electrical parameter acquisition module, which is used to collect data on the water temperature, water level, water flow rate of each circuit, and electrical parameters of the relay circuit in the dual-tank water in real time, and upload them to the edge control layer. The edge control layer includes edge control nodes, which include a data aggregation and preprocessing module, a conflict arbitration module, an instruction execution module, a state synchronization module, and a communication module. The local human-machine interaction terminal connects to the edge control node through a local communication bus. The mobile remote terminal accesses the wireless network through the communication module and establishes a communication connection with the cloud AI layer. The data aggregation and preprocessing module aggregates sensor data uploaded from the sensor acquisition layer and user-defined data uploaded from the terminal layer, performs feature extraction and preprocessing, and uploads the preprocessed feature data to the cloud AI layer through the communication module. The conflict arbitration module receives and caches control commands issued by each terminal and pre-scheduling strategies issued by the cloud AI layer. Based on a preset priority matrix, it arbitrates conflicts between concurrent control commands and pre-scheduling strategies from multiple terminals to generate a unique execution command. The command execution module outputs multiple drive control signals based on the arbitrated execution command. The status synchronization module synchronously feeds back the real-time operating status of the system to the local human-machine interaction terminal through the local communication bus and to the mobile remote terminal through the communication module. The communication module is used to access the wireless network to realize bidirectional data communication between the edge control node, the cloud AI layer, and the mobile remote terminal. The equipment execution layer includes a water supply pump, a solar circulating pump, a pipeline return water pump, an electric heater, a constant temperature outlet water pump, a constant temperature return water pump, a cooling water valve, and an anti-freeze heating cable; each device in the equipment execution layer is communicatively connected to the instruction execution module, receives multiple drive control signals, and executes corresponding actions; The cloud AI layer includes a cloud AI management terminal, which has a built-in time series prediction model and multi-objective optimization algorithm. It is used to receive preprocessed feature data uploaded by the edge control layer, perform user water use behavior learning and load prediction, generate multi-heat source collaborative pre-scheduling strategies, and send them to the edge control layer through the communication module.

[0016] The priority matrix of the conflict arbitration module is: local emergency stop / manual operation command > remote active operation command > cloud pre-scheduling strategy command; the conflict arbitration module maintains a timestamped command queue. When a conflict is detected between multiple terminal concurrent control commands and the pre-scheduling strategy within a preset time window, arbitration is performed according to the priority matrix. This setting ensures that local on-site control always has the highest response priority to guarantee operational safety, while also ensuring that the user's remote active intent takes precedence over the system's automatic pre-scheduling strategy, achieving orderly resolution of multi-terminal concurrent conflicts and reliable output of a single execution command; the time series prediction model is based on the LSTM or Prophet algorithm, according to historical water usage time, Water consumption, set temperature, and ambient temperature data generate a water load prediction curve for future periods. A multi-objective optimization algorithm, with the dual objectives of minimizing operating energy consumption and maximizing water comfort, generates a multi-heat source collaborative pre-scheduling strategy based on the future water load prediction curve and current solar collector temperature, water tank temperature, and ambient temperature. This strategy includes a solar circulating pump pre-start period, an electric heater pre-heating period and power, and a pipeline return water pump circulation period. This configuration enables the system to predict water load in advance based on historical water usage behavior and collaboratively schedule multiple heat sources, achieving on-demand preheating, significantly reducing operating energy consumption and minimizing user waiting time. A dual-tank temperature sensor group is also included. This includes a main water tank temperature sensor, an auxiliary water tank temperature sensor, a solar collector temperature sensor, and an ambient temperature sensor; water level sensors include a main water tank water level sensor and an auxiliary water tank water level sensor; an electrical parameter acquisition module is used to collect current and voltage data from each output circuit of the command execution module, enabling real-time and accurate monitoring of the entire chain of water temperature, water level, water flow rate in each circuit, and electrical parameters of the relay circuits in both water tanks; the command execution module includes 8 relay outputs, corresponding to drive the water supply pump, solar circulation pump, pipeline return pump, electric heater, constant temperature outlet pump, constant temperature return pump, cooling water valve, and antifreeze heating cable; the communication module integrates a QR code distribution network unit and... The wireless communication module of the WiFi / 4G communication unit; the mobile remote terminal completes device binding by scanning the QR code of the network configuration unit, and interacts with the cloud AI management terminal through the WiFi / 4G communication unit. This setting enables the mobile remote terminal to quickly complete device binding and network access with one-click scanning, and ensures the stability and real-time performance of remote data interaction through WiFi / 4G dual-mode communication, significantly reducing the user configuration threshold and improving the system communication reliability; the local human-machine interaction terminal includes a touch screen and a physical button panel. The touch screen is used to display the water temperature, water level and operating status of each component in the dual water tanks, and the physical button panel is used for local emergency stop and manual operation;The mobile remote terminal is a smartphone or WeChat mini-program terminal equipped with a central hot water control APP. This setup allows users to perform on-site visual monitoring and emergency control via the local touchscreen and physical buttons, while also remotely scheduling and checking status via the mobile APP or WeChat mini-program, achieving seamless interaction between local and remote channels.

[0017] Workflow: The steps of the AI-based multi-terminal collaborative central hot water control system are as follows: S1. Multi-terminal data aggregation and water usage behavior learning: The local human-machine interface terminal collects and records the target water temperature, reserved water usage time period, and mode selection parameters input by the user on-site through the touch screen or physical button panel, forming local setting data; The mobile remote terminal uploads the remote reserved water usage data and parameter adjustment instructions input by the user in the mobile APP or WeChat mini-program to the edge control node through the communication module; The data aggregation and preprocessing module of the edge control node receives the above-mentioned local setting data and remote reserved water usage data, and simultaneously receives the dual-tank temperature sensor group, water level sensor, flow sensor, and electrical parameter acquisition data from the sensor acquisition layer. The module uploads real-time data on dual-tank water temperature, water level, water flow rate of each circuit, and electrical parameters of relay circuits. The data aggregation and preprocessing module cleans, aligns, and extracts features from the above multi-source heterogeneous data, generating structured preprocessed feature data containing user water usage time, water consumption, set temperature, ambient temperature, and system operating status. This data is then uploaded to the cloud-based AI management terminal via the communication module. The cloud-based AI management terminal's built-in time-series prediction model receives the structured preprocessed feature data, trains and learns based on historical water usage behavior, identifies user water usage patterns, and generates a water load prediction curve for future periods. S2, AI multi-heat source pre-scheduling decision-making: The cloud-based AI management terminal calls a multi-objective optimization algorithm to achieve the lowest operating energy consumption and the best user experience. The optimal water comfort level is determined by a dual objective function. The predicted water load for future periods generated in S1 is jointly calculated with the current solar collector temperature, main water tank temperature, auxiliary water tank temperature, and ambient temperature. A multi-objective optimization algorithm outputs a multi-heat source collaborative pre-scheduling strategy. This strategy includes the pre-start time and runtime of the solar circulating pump, the pre-heating start time and heating power of the electric heater, the circulation time of the pipeline return pump, and the coordinated start-stop sequence of the constant temperature outlet pump and the constant temperature return pump. The cloud-based AI management terminal distributes this pre-scheduling strategy to the edge control node via a communication module. S3 involves intelligent arbitration of multi-terminal command conflicts. The conflict arbitration module of the edge control node establishes a timestamped command queue, real-time monitoring and caching of three sources. The control commands are as follows: manual control commands issued by the local human-machine interaction terminal through the local communication bus, remote active operation commands issued by the mobile remote terminal through the communication module, and pre-scheduling strategy commands issued by the cloud AI management terminal through the communication module. The conflict arbitration module detects whether the commands from the above three sources exist simultaneously within a preset time window and whether there is a control conflict for the same target device. If a conflict is detected, arbitration is performed according to the priority matrix, which is in the following order from high to low: local emergency stop / manual operation command > remote active operation command > cloud pre-scheduling strategy command. The conflict arbitration module outputs the only execution command after arbitration to the command execution module. The command that is not adopted is returned to the corresponding terminal with a conflict reason prompt.S4. Edge Execution and State Synchronization: The instruction execution module of the edge control node receives the unique execution instruction output by S3, parses it into the corresponding switch control signals output by the 8 relays, and drives the water pump, solar circulation pump, pipeline return water pump, electric heater, constant temperature water outlet pump, constant temperature water return pump, cold water supply valve, and antifreeze heating cable to perform corresponding actions. During execution, the sensing acquisition layer continuously feeds back real-time operating parameters to the edge control node. The state synchronization module of the edge control node outputs the real-time operating status of the system in two synchronous outputs: one output pushes the water temperature, water level, and operating status of each device in the dual water tanks to the local human-machine interface terminal in real time via the local communication bus. One terminal displays the information locally on its touchscreen; another pushes the same operating status to a mobile remote terminal via the MQTT protocol through a communication module, achieving consistency across multiple terminals. This realizes an AI-based multi-terminal collaborative central hot water control method and system. Through AI pre-scheduling, conflict arbitration, and closed-loop control with multi-terminal collaboration, it significantly improves the collaborative scheduling and energy utilization efficiency of multiple heat sources, the resolution of multi-terminal command conflicts and the consistency of state synchronization, as well as the proactive load prediction capability based on water usage behavior. This solves the problems of high energy consumption when each heat source operates independently, easy conflicts and asynchronous states in multi-terminal operations, and the lack of proactive load prediction based on water usage behavior in existing systems.

[0018] This article uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only for the purpose of helping to understand the method and core ideas of the present invention. The above descriptions are only preferred embodiments of the present invention. It should be noted that due to the limitations of textual expression, while there are objectively infinite specific structures, those skilled in the art can make several improvements, modifications, or changes without departing from the principles of the present invention, and can also combine the above technical features in an appropriate manner. These improvements, modifications, changes, or combinations, or the direct application of the inventive concept and technical solution to other situations without modification, should all be considered within the scope of protection of the present invention.

Claims

1. A multi-terminal collaborative central hot water control system based on artificial intelligence, characterized in that: It includes the sensing and acquisition layer, the edge control layer, the device execution layer, the cloud AI layer, and the terminal layer; The terminal layer includes a local human-machine interaction terminal and a mobile remote terminal; the local human-machine interaction terminal is used for setting on-site parameters and issuing manual control commands; the mobile remote terminal is used for remotely scheduling water use, setting parameters, and receiving status push notifications. The sensing and acquisition layer includes a dual-tank temperature sensor group, a water level sensor, a flow sensor, and an electrical parameter acquisition module, which is used to collect data on the water temperature, water level, water flow rate of each circuit, and electrical parameters of the relay circuit in the dual-tank water in real time, and upload them to the edge control layer. The edge control layer includes an edge control node, which includes a data aggregation and preprocessing module, a conflict arbitration module, an instruction execution module, a state synchronization module, and a communication module. The local human-computer interaction terminal is connected to the edge control node through a local communication bus. The mobile remote terminal accesses the wireless network through the communication module and establishes a communication connection with the cloud AI layer. The data aggregation and preprocessing module is used to aggregate the sensor data uploaded by the sensor acquisition layer and the user setting data uploaded by the terminal layer, perform feature extraction and preprocessing, and upload the preprocessed feature data to the cloud AI layer through the communication module; the conflict arbitration module is used to receive and cache the control commands issued by each terminal and the pre-scheduling strategy issued by the cloud AI layer, and perform conflict arbitration on the concurrent control commands of multiple terminals and the pre-scheduling strategy according to the preset priority matrix to generate a unique execution command; The instruction execution module is used to output multiple drive control signals according to the arbitrated execution instruction; The status synchronization module is used to synchronously feed back the real-time operating status of the system to the local human-machine interaction terminal through the local communication bus, and synchronously feed back to the mobile remote terminal through the communication module; the communication module is used to access the wireless network to realize bidirectional data communication between the edge control node, the cloud AI layer and the mobile remote terminal; The device execution layer includes a water pump, a solar circulating pump, a pipeline return water pump, an electric heater, a constant temperature water outlet pump, a constant temperature water return pump, a cooling water valve, and an antifreeze heating cable; each device in the device execution layer is communicatively connected to the instruction execution module, receives the multi-channel drive control signals, and executes corresponding actions; The cloud AI layer includes a cloud AI management terminal, which has a built-in time series prediction model and a multi-objective optimization algorithm. It is used to receive preprocessed feature data uploaded by the edge control layer, perform user water use behavior learning and load prediction, generate a multi-heat source collaborative pre-scheduling strategy, and send it to the edge control layer through the communication module.

2. The multi-terminal collaborative central hot water control system based on artificial intelligence according to claim 1, characterized in that: The priority matrix of the conflict arbitration module is: local emergency stop / manual operation command > remote active operation command > cloud pre-scheduling strategy command; the conflict arbitration module maintains a timestamped command queue, and when a conflict is detected between multiple terminal concurrent control commands and the pre-scheduling strategy within a preset time window, arbitration is performed according to the priority matrix.

3. The multi-terminal collaborative central hot water control system based on artificial intelligence according to claim 1, characterized in that: The time-series prediction model is based on the LSTM or Prophet algorithm, which generates a water load prediction curve for future periods based on historical water usage time, water consumption, set temperature, and ambient temperature data. The multi-objective optimization algorithm takes the lowest operating energy consumption and the highest water comfort as dual objectives. Based on the water load prediction curve for future periods and the current solar collector temperature, water tank temperature, and ambient temperature, it generates the multi-heat source collaborative pre-scheduling strategy. The multi-heat source collaborative pre-scheduling strategy includes the solar circulating pump pre-start period, the electric heater preheating period and power, and the pipeline return water pump circulation period.

4. The multi-terminal collaborative central hot water control system based on artificial intelligence according to claim 1, characterized in that: The dual-tank temperature sensor group includes a main water tank temperature sensor, an auxiliary water tank temperature sensor, a solar collector temperature sensor, and an ambient temperature sensor; the water level sensor includes a main water tank water level sensor and an auxiliary water tank water level sensor; the electrical parameter acquisition module is used to collect current and voltage data from each output circuit of the instruction execution module.

5. The multi-terminal collaborative central hot water control system based on artificial intelligence according to claim 1, characterized in that: The instruction execution module includes 8 relay outputs, which respectively drive the water pump, solar circulation pump, pipeline return water pump, electric heater, constant temperature water pump, constant temperature return water pump, cold water valve, and antifreeze heating cable.

6. The multi-terminal collaborative central hot water control system based on artificial intelligence according to claim 1, characterized in that: The communication module is a wireless communication module that integrates a QR code network configuration unit and a WiFi / 4G communication unit; the mobile remote terminal completes device binding by scanning the QR code network configuration unit and interacts with the cloud AI management terminal through the WiFi / 4G communication unit.

7. The multi-terminal collaborative central hot water control system based on artificial intelligence according to claim 1, characterized in that: The local human-machine interface terminal includes a touch screen and a physical button panel. The touch screen is used to display the water temperature and level of the dual water tanks and the operating status of each device. The physical button panel is used for local emergency stop and manual operation. The mobile remote terminal is a smartphone or WeChat mini-program terminal with a central hot water control APP installed.

8. A control method for a multi-terminal collaborative central hot water control system based on artificial intelligence as described in claims 1-7, characterized in that, Includes the following steps: S1. Multi-terminal data aggregation and water use behavior learning: The local human-computer interaction terminal collects and records the target water temperature, scheduled water use time period and mode selection parameters entered by the user on site through the touch screen or physical button panel to form local setting data. The mobile remote terminal uploads remote water reservation data and parameter adjustment commands input by the user in a mobile APP or WeChat mini-program to the edge control node via the communication module. The data aggregation and preprocessing module of the edge control node receives the aforementioned local setting data and remote water reservation data, and simultaneously receives real-time data on the water temperature, water level, flow rate of each circuit, and electrical parameters of the relay circuit uploaded by the dual-tank temperature sensor group, water level sensor, flow sensor, and electrical parameter acquisition module in the sensing acquisition layer. The data aggregation and preprocessing module cleans, aligns, and extracts features from the aforementioned multi-source heterogeneous data to generate structured preprocessed feature data containing user water usage time, water consumption, set temperature, ambient temperature, and system operating status, and uploads it to the cloud AI management terminal via the communication module. The time series prediction model built into the cloud AI management terminal receives the aforementioned structured preprocessed feature data, trains and learns based on historical water usage behavior, identifies user water usage patterns, and generates a water load prediction curve for future periods. S2, AI multi-heat source pre-scheduling decision-making: The cloud-based AI management terminal calls a multi-objective optimization algorithm with the dual objective functions of minimizing operating energy consumption and maximizing water comfort. It performs joint calculations on the future water load prediction curve generated by S1 and the current solar collector temperature, main water tank temperature, auxiliary water tank temperature, and ambient temperature. The multi-objective optimization algorithm outputs a multi-heat source collaborative pre-scheduling strategy, which includes the pre-start time and running duration of the solar circulating pump, the preheating start time and heating power of the electric heater, the circulation time of the pipeline return water pump, and the coordinated start-stop sequence of the constant temperature outlet water pump and the constant temperature return water pump. The cloud-based AI management terminal distributes the above pre-scheduling strategy to the edge control node through the communication module. S3, Intelligent Arbitration of Multi-Terminal Command Conflicts: The conflict arbitration module of the edge control node establishes a timestamped command queue, which listens to and caches control commands from three sources in real time: manual control commands issued by the local human-machine interaction terminal through the local communication bus, remote active operation commands issued by the mobile remote terminal through the communication module, and pre-scheduling strategy commands issued by the cloud AI management terminal through the communication module. The conflict arbitration module detects whether the instructions from the above three sources exist simultaneously within a preset time window and whether there is a control conflict for the same target device; If a conflict is detected, arbitration is performed based on a priority matrix, which is in the following order from high to low: local emergency stop / manual operation command > remote active operation command > cloud pre-scheduling strategy command; The conflict arbitration module outputs the sole execution instruction after arbitration to the instruction execution module. Instructions that are not accepted are returned to the corresponding terminal along with a conflict reason prompt. S4. Edge Execution and State Synchronization: The instruction execution module of the edge control node receives the unique execution instruction output by S3, parses it into the corresponding switch control signals output by the 8 relays, and drives the water pump, solar circulation pump, pipeline return water pump, electric heater, constant temperature water outlet pump, constant temperature water return pump, cold water supply valve, and antifreeze heating cable to perform corresponding actions. During execution, the sensing acquisition layer continuously feeds back real-time operating parameters to the edge control node. The state synchronization module of the edge control node divides the real-time operating status of the system into two synchronous outputs: one pushes the water temperature, water level, and operating status of each device in the dual water tanks to the touch screen of the local human-machine interaction terminal for local display through the local communication bus; the other pushes the same operating status to the mobile remote terminal through the communication module using the MQTT protocol to achieve consistency of status across multiple terminals.

Images