Artificial intelligence climate chamber adaptive control system based on reinforcement learning algorithm

Abstract

This invention discloses an artificial intelligence climate chamber adaptive control system based on a reinforcement learning algorithm, relating to the field of intelligent control technology. Specifically, it comprises a data acquisition unit, a reinforcement learning control unit, a control execution unit, and a communication unit. The data acquisition unit, the reinforcement learning control unit, and the control execution unit form a closed-loop feedback loop through the communication unit. The reinforcement learning control unit generates a control strategy that minimizes energy consumption and meets response speed based on the dynamic changes in environmental parameters and target values ​​for the crop growth stage. The control execution unit then implements adaptive adjustment of the climate chamber environment. The system is designed to autonomously learn and adjust the climate chamber control strategy based on the crop growth stage, real-time environmental changes, and resource consumption.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to an artificial intelligence climate chamber adaptive control system based on a reinforcement learning algorithm. Background Art

[0002] With the continuous advancement of modern agricultural technology, the role of precision and intelligent control in agricultural production is becoming increasingly prominent. In particular, in greenhouse cultivation, climate chamber cultivation, and other facility agriculture sectors, the crop growth environment (including parameters such as temperature, humidity, light intensity, and carbon dioxide concentration) has a direct and significant impact on crop yield and quality. To maintain a suitable growing environment, traditional climate control systems typically rely on preset environmental parameter thresholds. When detected environmental data exceeds or falls below the preset range, the control system activates appropriate regulatory devices, such as heaters, humidifiers, lighting, or ventilation, to restore the set environmental conditions.

[0003] However, crops have significant differences in their requirements for environmental conditions during different growth stages (such as germination, seedling, flowering, and fruiting), and traditional control systems find it difficult to dynamically adjust control strategies according to the crop growth cycle. At the same time, external climatic conditions (such as seasonal changes, day and night temperature differences, and sudden weather changes) also have a significant impact on the environment in greenhouses or climate chambers. Traditional systems lack the ability to adapt to external changes in real time, resulting in lags and inaccuracies in environmental control.

[0004] Therefore, how to autonomously learn and adjust the climate chamber control strategy based on the crop growth stage, real-time environmental changes and resource consumption status has become a technical problem that needs to be solved urgently. Summary of the Invention

[0005] The main purpose of this invention is to provide an artificial intelligence climate chamber adaptive control system based on a reinforcement learning algorithm, which aims to be able to autonomously learn and adjust the climate chamber control strategy based on the crop growth stage, real-time environmental changes and resource consumption status.

[0006] To achieve the above objectives, the present invention proposes an artificial intelligence climate chamber adaptive control system based on a reinforcement learning algorithm, comprising:

[0007] A data acquisition unit, used to collect multiple environmental parameters in the climate chamber in real time;

[0008] a reinforcement learning control unit that autonomously learns and generates a control strategy based on the environmental parameters, crop growth stage, real-time environmental changes, and resource consumption status through a reward function that comprehensively optimizes environmental parameter proximity, energy consumption, and response speed;

[0009] a control execution unit, configured to adjust environmental parameters within the climate chamber according to the control strategy;

[0010] Communication unit, used to realize data transmission and interaction between units;

[0011] Among them, the data acquisition unit, reinforcement learning control unit, and control execution unit form a closed-loop feedback through the communication unit. The reinforcement learning control unit generates a control strategy that minimizes energy consumption and meets response speed based on the dynamic changes of environmental parameters and the target values ​​of the crop growth stage, and realizes adaptive adjustment of the climate chamber environment through the control execution unit.

[0012] In one embodiment of the present application, the data acquisition unit includes a temperature sensor, a humidity sensor, a light sensor, a carbon dioxide concentration sensor and an oxygen concentration sensor. The temperature sensor, humidity sensor, light sensor, carbon dioxide concentration sensor and oxygen concentration sensor constitute a sensor group, and a sensor group is provided in each layered structure of the culture area in the climate chamber.

[0013] In one embodiment of the present application, the control execution unit includes a heating device, a refrigeration device, a light adjustment device, a fan, an oxygen valve and a carbon dioxide valve. The heating device and the refrigeration device realize uniform diffusion of hot and cold air through a circulating fan, and the oxygen valve and the carbon dioxide valve adjust the gas concentration through the side panel pores.

[0014] In one embodiment of the present application, the reward function of the reinforcement learning control unit is:

[0015]

[0016] in, Indicates that the status Next action The immediate reward value obtained after Indicates the current environmental status, including real-time parameters such as temperature, humidity, light, and gas concentration; Indicates control actions; Indicates the environmental parameter deviation weight coefficient; is the current environment parameter; is the ideal value for the corresponding crop growth stage; Indicates the number of types of environmental parameters monitored; represents the energy consumption weight coefficient; E ( a ) is an action a energy consumption; Represents the response speed weight coefficient; δ ( s ) is the response speed reward.

[0017] In one embodiment of the present application, the communication unit includes an Internet of Things gateway and an industrial computer. The Internet of Things gateway is connected to each sensor through the Modbus protocol and communicates with the industrial computer through the TCP protocol to realize the transmission of sensor data and control instructions.

[0018] In one embodiment of the present application, the reinforcement learning control unit further includes:

[0019] The crop growth stage recognition module is used to determine the current crop growth stage based on the historical data of environmental parameters and the preset crop growth model, and dynamically adjust the ideal value in the reward function. and weight coefficient , , ;

[0020] Among them, the growth stage identification module cooperates with the data acquisition unit and the user interface unit to match the real-time leaf age, plant height and environmental accumulation through the crop growth model, determine the stage switching threshold, and trigger the staged adaptive update of the control strategy.

[0021] In one embodiment of the present application, the heating device is a resistance heating wire, the refrigeration device is a compressor and a condensing fan, the heating wire reuses the waste heat of the compressor, and the environmental parameters are uniformly adjusted through the porous backplate and the circulating fan.

[0022] In one embodiment of the present application, a user interface unit is further included, which provides real-time environmental parameter visualization, a manual control interface and an abnormal alarm function, and supports historical data export and comparative experiment configuration.

[0023] In one embodiment of the present application, the Internet of Things gateway controls the power on and off of each device in the execution unit through a solid-state relay.

[0024] In one embodiment of the present application, the training process of the reinforcement learning control unit includes:

[0025] Initialize model parameters based on historical data or simulation environment;

[0026] Generate adaptive control strategies through environment state acquisition, action generation, reward calculation and strategy iterative optimization;

[0027] When the reward value converges or reaches the preset number of iterations, the optimal control strategy is output.

[0028] By adopting the above technical solution, the artificial intelligence climate chamber adaptive control system based on the reinforcement learning algorithm can realize dynamic adaptive adjustment of the internal environment of the climate chamber. By comprehensively considering the reward function of environmental parameter proximity, energy consumption minimization and response speed optimization, the reinforcement learning control unit can continuously optimize the control strategy, effectively improving the accuracy of environmental regulation and energy efficiency utilization. At the same time, through the closed-loop feedback mechanism and efficient data interaction of the communication unit, the system achieves highly automated operation and stability, greatly reducing the need for manual intervention, enhancing the climate chamber's adaptability to complex environmental changes, ensuring that crops obtain the best growth environment conditions at different growth stages, and improving crop growth quality and experimental controllability. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The present invention will be described in detail below with reference to specific embodiments and accompanying drawings, wherein:

[0030] Figure 1 This is a schematic structural diagram of a first embodiment of the present invention;

[0031] Figure 2 It is a schematic diagram of the three-dimensional structure of the climate box of the present invention.

[0032] 10. Cabinet area; 11. Upper groove area; 12. Cultivation area; 20. Display screen. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the following specific embodiments are only used to explain the present invention and do not constitute a limitation of the present invention.

[0034] like Figure 1 As shown, in order to achieve the above purpose, the present invention proposes an artificial intelligence climate chamber adaptive control system based on a reinforcement learning algorithm, comprising:

[0035] A data acquisition unit, used to collect multiple environmental parameters in the climate chamber in real time;

[0036] a reinforcement learning control unit that autonomously learns and generates a control strategy based on the environmental parameters, crop growth stage, real-time environmental changes, and resource consumption status through a reward function that comprehensively optimizes environmental parameter proximity, energy consumption, and response speed;

[0037] a control execution unit, configured to adjust environmental parameters within the climate chamber according to the control strategy;

[0038] Communication unit, used to realize data transmission and interaction between units;

[0039] Among them, the data acquisition unit, reinforcement learning control unit, and control execution unit form a closed-loop feedback through the communication unit. The reinforcement learning control unit generates a control strategy that minimizes energy consumption and meets response speed based on the dynamic changes of environmental parameters and the target values ​​of the crop growth stage, and realizes adaptive adjustment of the climate chamber environment through the control execution unit.

[0040] Specifically, this embodiment provides an artificial intelligence climate chamber adaptive control system based on a reinforcement learning algorithm, which specifically includes the following steps:

[0041] First, multiple environmental monitoring sensors installed inside the climate chamber collect environmental parameters in real time. The environmental parameters include at least air temperature, humidity, light intensity, carbon dioxide concentration, and oxygen concentration. Each environmental parameter is connected to the communication unit via the RS485 bus protocol and transmitted to the reinforcement learning control unit.

[0042] Secondly, the reinforcement learning control unit receives the environmental parameters and, based on a preset crop growth stage database, real-time environmental change data, and resource consumption status data, calls an internal reinforcement learning model for processing. The reinforcement learning model initializes parameters such as network weights, learning rate, and discount factor, and sets a reward function with the goals of approaching the target value of the environmental parameters, minimizing energy consumption, and optimizing the response speed. The reward function is defined as:

[0043]

[0044] in, Indicates that the status Next action The immediate reward value obtained after Indicates the current environmental status, including real-time parameters such as temperature, humidity, light, and gas concentration; Indicates control actions; Indicates the environmental parameter deviation weight coefficient; is the current environment parameter; is the ideal value for the corresponding crop growth stage; Indicates the number of types of environmental parameters monitored; represents the energy consumption weight coefficient; E ( a ) is an action a energy consumption; Represents the response speed weight coefficient; δ ( s) is a response speed reward. The reinforcement learning control unit constructs a standardized state vector according to the current environmental state and inputs it into the reinforcement learning network. After neural network reasoning, a control strategy is generated. The control strategy includes specific operating instructions for the internal equipment of the climate chamber, such as heating power adjustment of the heating plate, cooling intensity adjustment of the condenser, brightness adjustment of the illumination panel, speed adjustment of the fan, opening and closing degree adjustment of the oxygen valve, and opening and closing degree adjustment of the carbon dioxide valve.

[0045] Subsequently, the control execution unit receives the control strategy generated by the reinforcement learning control unit, generates a specific execution signal after analysis, and controls the actions of various environmental conditioning devices through the solid-state relay module to ensure that each environmental parameter is adjusted to the target value corresponding to the crop growth stage.

[0046] After each device completes its action, the data acquisition unit collects the updated environmental parameters again. The updated environmental parameters are transmitted to the reinforcement learning control unit via the communication unit. The reinforcement learning control unit updates the state vector based on the new round of environmental parameter inputs and performs learning iterations through the strategy update process to form a closed-loop feedback mechanism.

[0047] The communication unit includes a Modbus RTU bus communication interface, a TCP / IP communication interface, and a WebSocket protocol module, which is responsible for ensuring real-time data transmission and interaction between the data acquisition unit, the reinforcement learning control unit, and the control execution unit. A buffer queue management mechanism is set up inside the communication unit to ensure data integrity and timeliness during high-concurrency data requests.

[0048] In this embodiment, the data acquisition unit and the reinforcement learning control unit communicate using the TCP / IP protocol, and the reinforcement learning control unit and the control execution unit communicate using the Modbus RTU protocol, thereby ensuring low-latency transmission of control signals. At the same time, the reinforcement learning control unit pushes environmental parameters and control status to the host computer display system in real time through the WebSocket protocol, realizing visual monitoring and manual intervention; when generating the control strategy, the reinforcement learning control unit analyzes the target environmental range corresponding to the crop growth stage in real time, adjusts the action amplitude according to the current deviation degree, and gives priority to low-energy consumption action paths during the action adjustment process. If sudden environmental interference occurs, the reinforcement learning control unit immediately re-evaluates the status and dynamically generates an emergency control strategy to ensure that the environment quickly returns to the target range.

[0049] By adopting the above technical solution, the artificial intelligence climate chamber adaptive control system based on the reinforcement learning algorithm can realize dynamic adaptive adjustment of the internal environment of the climate chamber. By comprehensively considering the reward function of environmental parameter proximity, energy consumption minimization and response speed optimization, the reinforcement learning control unit can continuously optimize the control strategy, effectively improving the accuracy of environmental regulation and energy efficiency utilization. At the same time, through the closed-loop feedback mechanism and efficient data interaction of the communication unit, the system achieves highly automated operation and stability, greatly reducing the need for manual intervention, enhancing the climate chamber's adaptability to complex environmental changes, ensuring that crops obtain the best growth environment conditions at different growth stages, and improving crop growth quality and experimental controllability.

[0050] In one embodiment of the present application, the data acquisition unit includes a temperature sensor, a humidity sensor, a light sensor, a carbon dioxide concentration sensor and an oxygen concentration sensor. The temperature sensor, humidity sensor, light sensor, carbon dioxide concentration sensor and oxygen concentration sensor constitute a sensor group, and a sensor group is provided in each layered structure of the culture area in the climate chamber.

[0051] Specifically, the data acquisition unit includes a temperature sensor, a humidity sensor, a light sensor, a carbon dioxide concentration sensor, and an oxygen concentration sensor. These sensors form a sensor group through a unified data interface. Each sensor group is connected to the communication unit via a 485 bus, and the communication unit transmits the collected data to the reinforcement learning control unit.

[0052] There are multiple layered culture structures in the climate chamber. A sensor group is fixedly installed inside each culture layer area. The specific installation method of each sensor group is as follows:

[0053] The temperature sensor and humidity sensor are installed in the center of each cultivation area to collect the micro-environment temperature and humidity around the crops in the layer in real time.

[0054] The light sensor is installed in the middle area below the light panel above the cultivation area to monitor the actual light intensity received by the crops.

[0055] The carbon dioxide concentration sensor and the oxygen concentration sensor are installed near the air holes on the side panels of the culture area to detect the concentration of air components in the culture area.

[0056] Each sensor group is connected to the communication interface module in the climate chamber cabinet area through a shielded cable. The communication interface module uniformly transmits the signal to the gateway and forwards it to the reinforcement learning control unit.

[0057] The data collected by each sensor group is timestamped, and the communication unit is equipped with a priority management mechanism to ensure that abnormal environmental data is transmitted first. CRC check is used in the communication process to ensure data integrity. By setting up sensor groups in the hierarchical structure of each cultivation area of ​​the climate chamber, accurate monitoring of the microenvironment of crops at different levels and in different areas can be achieved. The multi-layer environmental data collected by the data acquisition unit can be analyzed and processed by the reinforcement learning control unit to form a customized environmental adjustment strategy, thereby improving the locality and adaptability of the climate chamber environmental control.

[0058] By adopting the above technical solution, the data acquisition unit can realize real-time collection of the microenvironmental parameters of each layer inside the climate chamber by setting a sensor group inside each culture layer of the climate chamber, ensuring that the data received by the reinforcement learning control unit covers the environmental changes in different growth areas in the climate chamber. The reinforcement learning control unit generates a more reasonable control strategy based on more detailed data, thereby improving the accuracy of environmental control and the consistency of the crop growth environment.

[0059] In one embodiment of the present application, the control execution unit includes a heating device, a refrigeration device, a light adjustment device, a fan, an oxygen valve and a carbon dioxide valve. The heating device and the refrigeration device realize uniform diffusion of hot and cold air through a circulating fan, and the oxygen valve and the carbon dioxide valve adjust the gas concentration through the side panel pores.

[0060] Specifically, the control execution unit provided in this embodiment includes a heating device, a cooling device, a light control device, a fan, an oxygen valve, and a carbon dioxide valve. The heating device uses a high-power heating resistor installed in the heating layer above the cooling device. The heating resistor is controlled by a solid-state relay module to switch on and off, and to adjust the heating power.

[0061] The refrigeration device adopts a compressor refrigeration system and cooperates with a condensing fan to dissipate heat. The compressor is installed at the bottom of the climate chamber and connected to the refrigeration area inside the climate chamber through a condensing pipe. The refrigeration compressor is controlled on and off by a solid-state relay to achieve precise temperature control.

[0062] The circulating fan is installed above the heating device and the refrigeration device. The circulating fan adopts variable frequency control and can adjust the speed according to needs. The working principle of the circulating fan is to inhale the hot air generated by the heating device or the cold air generated by the refrigeration device and diffuse it evenly along the inside of the climate box through the porous back plate, so as to achieve rapid and uniform distribution of temperature inside the box and avoid local temperature differences affecting crop growth.

[0063] The light adjustment device consists of two detachable light panels installed on the top of the cultivation area. Each light panel is equipped with an LED light source with adjustable brightness and spectrum type. The light adjustment device is connected to the solid-state relay through a PWM dimming controller. It can dynamically adjust the light intensity and spectrum ratio according to the lighting strategy output by the reinforcement learning control unit to ensure the lighting requirements of crops at different growth stages.

[0064] The oxygen valve and carbon dioxide valve are installed at the corresponding positions of the air holes on the side panel of the climate chamber respectively. Both the oxygen valve and the carbon dioxide valve adopt a proportional electric control valve structure, and the gas flow is precisely controlled by adjusting the opening and closing angles. The oxygen valve and the carbon dioxide valve receive instructions from the control execution unit and adjust the air intake volume in real time according to the oxygen concentration and carbon dioxide concentration inside the climate chamber to keep the gas concentration inside the climate chamber in the optimal range for crop growth. The oxygen valve and the carbon dioxide valve are connected to the gas cylinder through independent pipelines, and the gas cylinder is equipped with a pressure reducing valve to ensure a stable gas supply; the various parts of the control execution unit are connected in real time with the reinforcement learning control unit by the communication unit. After receiving the control strategy generated by the reinforcement learning control unit, the communication unit parses the instructions and drives the heating device, refrigeration device, lighting adjustment device, fan, oxygen valve and carbon dioxide valve to act on demand in sequence. After the action is executed, the control execution unit returns the execution status through the communication unit for the reinforcement learning control unit to update the feedback information, ensuring the formation of a complete closed-loop control process.

[0065] By adopting the above technical solution, the control execution unit realizes the uniform diffusion of hot and cold air inside the climate chamber through the heating device and the refrigeration device in conjunction with the circulating fan, effectively avoiding the problem of uneven crop growth caused by local environmental fluctuations. At the same time, the oxygen valve and the carbon dioxide valve are combined with the structure of the air holes on the side panel of the climate chamber to achieve precise control of the gas concentration inside the climate chamber, further ensuring the dynamic stability of the gas composition in the crop growth environment.

[0066] In one embodiment of the present application, the reward function of the reinforcement learning control unit is:

[0067]

[0068] in, Indicates that the status Next action The immediate reward value obtained after Indicates the current environmental status, including real-time parameters such as temperature, humidity, light, and gas concentration; Indicates control actions; Indicates the environmental parameter deviation weight coefficient; is the current environment parameter; is the ideal value for the corresponding crop growth stage; Indicates the number of types of environmental parameters monitored; represents the energy consumption weight coefficient; E ( a ) is an action a energy consumption; Represents the response speed weight coefficient; δ ( s ) is the response speed reward.

[0069] Specifically, the reward function of the reinforcement learning control unit in this embodiment is set as:

[0070]

[0071] in, Indicates the current environment status Execute control action The immediate reward value obtained after the current environmental state s includes the set of parameters such as temperature, humidity, light intensity, carbon dioxide concentration and oxygen concentration inside the climate box collected in real time by the data acquisition unit, and the control action Represents the reinforcement learning control unit according to the current state The generated environmental adjustment command combination includes heating power adjustment of the heating device, cooling intensity adjustment of the cooling device, brightness and spectrum adjustment of the lighting adjustment device, fan speed change, oxygen valve opening change and carbon dioxide valve opening change; in the reward function The environmental parameter deviation weight coefficient is used to measure the importance of the deviation between the environmental parameter and the ideal target value. It is the current real-time monitored environmental values. is the ideal environmental parameter value set for the corresponding crop growth stage, The number of environmental parameters involved in the monitoring includes five categories: temperature, humidity, light intensity, carbon dioxide concentration and oxygen concentration. It is a penalty term for environmental parameter deviation. By finding the absolute deviation between each environmental parameter and the corresponding ideal value and taking a weighted sum, a negative reward corresponding to the degree of deviation is given to encourage the environment to quickly approach the target value.

[0072] is the energy consumption weight coefficient, used to evaluate the execution action The energy consumption caused by To control the action The corresponding energy consumption data includes the weighted sum of the energy consumption of the heating device, refrigeration device, light control device, fan, oxygen valve and carbon dioxide valve. It is an energy penalty term that encourages the reinforcement learning control unit to generate a lower energy consumption adjustment strategy by penalizing high energy consumption actions; is the response speed weight coefficient, which is used to measure the rate at which the environmental state converges to the target value. Reward for response speed, It is defined as the improvement of the degree to which the environmental parameters approach the target value per unit time. The third item The reward item is the response speed reward item. By rewarding the control action that quickly improves the environmental state, the system's response ability to sudden environmental changes is improved. The reinforcement learning control unit obtains the immediate reward value after each control action during the training process. To adjust the weight parameters of the internal strategy network and continuously optimize the control strategy, so that the internal environment of the climate chamber can meet the growth needs of crops while minimizing energy consumption and optimizing response speed.

[0073] By adopting the above technical solution, the reinforcement learning control unit can accurately optimize the climate chamber control strategy under different environmental changes by constructing a reward function that comprehensively considers environmental proximity, energy consumption and response speed. This allows the system to effectively reduce energy consumption and improve the rapid response capability of environmental control while maintaining an ideal growth environment.

[0074] In one embodiment of the present application, the communication unit includes an Internet of Things gateway and an industrial computer. The Internet of Things gateway is connected to each sensor through the Modbus protocol and communicates with the industrial computer through the TCP protocol to realize the transmission of sensor data and control instructions.

[0075] Specifically, in this embodiment, the communication unit specifically includes an Internet of Things gateway and an industrial computer. The Internet of Things gateway and each sensor are connected by wire using the Modbus RTU protocol. The Internet of Things gateway has multiple RS485 interfaces integrated inside, which correspond to the temperature sensors, humidity sensors, light sensors, carbon dioxide concentration sensors and oxygen concentration sensors distributed on each layer inside the climate chamber. The Internet of Things gateway collects real-time environmental parameter data of each sensor at a set time interval through a polling mechanism; the Internet of Things gateway performs preliminary formatting processing on the collected sensor data, including adding timestamps, sensor identification codes and data integrity verification, and then encapsulates the processed environmental parameter data into TCP data packets through the built-in TCP protocol stack and sends them to the industrial computer.

[0076] The industrial computer serves as the upper data processing center, receives environmental parameter data transmitted from the IoT gateway, performs data analysis, standardization and storage, and transmits the standardized environmental status data to the reinforcement learning control unit for subsequent control strategy generation. During the control instruction issuance process, the reinforcement learning control unit sends the generated control strategy in the control instruction format to the industrial computer via the TCP protocol. After parsing the control instruction, the industrial computer sends it to the IoT gateway again via the TCP protocol. The IoT gateway re-encodes the received control instruction according to the Modbus RTU protocol and sends it to the corresponding control devices, such as the heating device, cooling device, lighting control device, fan, oxygen valve and carbon dioxide valve execution module, to ensure that the environmental parameters are adjusted in real time according to the control strategy generated by the reinforcement learning control unit.

[0077] To improve the stability and fault tolerance of communication, the IoT gateway is equipped with a cache area and a disconnection reconnection mechanism. When network communication is abnormal, data can be cached and reissued to ensure the continuity and reliability of data collection and command transmission. The industrial computer is equipped with a dual network port structure to support redundant communication path design, further improving the stability of system operation and data security.

[0078] By adopting the above technical solution, the communication unit realizes the efficient collection and standardized transmission of sensor data and the accurate issuance of control instructions through the collaboration of the IoT gateway and the industrial computer, combining the Modbus protocol and the TCP protocol. It ensures the real-time data interaction and synchronous instruction execution between the modules inside the climate chamber, and greatly improves the communication reliability of the system.

[0079] In one embodiment of the present application, the reinforcement learning control unit further includes:

[0080] The crop growth stage recognition module is used to determine the current crop growth stage based on the historical data of environmental parameters and the preset crop growth model, and dynamically adjust the ideal value in the reward function. and weight coefficient , , ;

[0081] Among them, the growth stage identification module cooperates with the data acquisition unit and the user interface unit to match the real-time leaf age, plant height and environmental accumulation through the crop growth model, determine the stage switching threshold, and trigger the staged adaptive update of the control strategy.

[0082] Specifically, in this embodiment, the reinforcement learning control unit further includes a crop growth stage identification module. The crop growth stage identification module works in conjunction with the data acquisition unit and the user interface unit. First, the environmental parameter data of each cultivation area in the climate chamber is obtained in real time through the data acquisition unit, and compared and analyzed with the crop growth model data preset in the user interface unit. The crop growth model includes parameter information such as typical environmental demand characteristics, cumulative temperature, cumulative light intensity, humidity change curve, and carbon dioxide concentration requirements corresponding to different growth stages of crops. The crop growth stage identification module calculates real-time environmental cumulative indicators, such as cumulative leaf age, cumulative plant height, cumulative light radiation, cumulative temperature time integral, etc., and matches them with the standard cumulative values ​​in the growth model.

[0083] During the matching process, the crop growth stage recognition module is set with a stage switching threshold. When the real-time cumulative amount reaches or exceeds the set stage switching threshold, the crop growth stage recognition module determines that the crop growth stage has switched, triggering the reward function parameter update process, specifically including dynamically adjusting the ideal value in the reward function. , set the target value of each environmental parameter to the standard environmental value corresponding to the new growth stage, and adjust the weight coefficient appropriately according to the different requirements of the new stage for environmental stability, energy consumption control and response speed , , , for example, increasing the weight of environmental parameter proximity during the seedling stage , increase the weight of energy consumption optimization during the growth period , increase the response speed weight during the flowering and fruiting period , ensuring that the control strategy generated by the reinforcement learning control unit is more in line with the actual needs of crops at different growth stages; the crop growth stage recognition module can also receive manual correction signals input by the user through the user interface unit. When the user inputs the measured leaf age or plant height data through the interface, the crop growth stage recognition module can correct the growth stage judgment according to the manual data to improve the accuracy and flexibility of recognition.

[0084] After the crop growth stage recognition module recognizes a new growth stage, it sends the updated 、 , , The parameters are passed to the reinforcement learning control unit, which regenerates the control strategy based on the new reward function parameters to achieve phased adaptive update of the control strategy, thereby ensuring that the internal environment of the climate chamber always matches the actual growth needs of the crops.

[0085] Using the above technical solution, the reinforcement learning control unit introduces a crop growth stage recognition module and combines historical data of environmental parameters with a preset crop growth model to achieve dynamic judgment of the crop growth stage and adaptive adjustment of the reward function parameters, enabling the system to optimize environmental control objectives and control strategies in real time according to the actual growth process of the crops.

[0086] In one embodiment of the present application, the heating device is a resistance heating wire, the refrigeration device is a compressor and a condensing fan, the heating wire reuses the waste heat of the compressor, and the environmental parameters are uniformly adjusted through the porous backplate and the circulating fan.

[0087] Specifically, in this embodiment, the heating device in the control execution unit is a resistance heating wire. The resistance heating wire is made of high-efficiency alloy material and has the performance of rapid heating and high-temperature aging resistance. The resistance heating wire is installed in a partition area above the refrigeration compressor inside the climate chamber. The refrigeration device includes a refrigeration compressor and a condensing fan. The refrigeration compressor adopts a variable frequency compressor structure, and the temperature inside the climate chamber is reduced by refrigerant circulation. The condensing fan is used to assist the compressor in dissipating heat and improve the refrigeration efficiency and stability of the compressor.

[0088] During the operation of the system, when the refrigeration compressor completes cooling, the heat generated by the long-term operation of the compressor body is transferred to the upper resistance heating wire area through physical contact. The resistance heating wire uses the waste heat of the compressor for preheating, thereby reducing the startup energy consumption and heating response time when the heating mode needs to be started, realizing the secondary utilization of energy and improving the overall energy utilization efficiency of the climate chamber.

[0089] A porous back panel is provided inside the climate chamber, which is distributed in the cooling area, heating area and the junction of the air duct. The circulating fan is installed in the center position above the refrigeration compressor and the resistance heating wire. The circulating fan adopts variable frequency control and can adjust the air volume according to the real-time environmental parameters. When the circulating fan is running, the cold air output by the refrigeration compressor or the hot air generated by the resistance heating wire is sucked into the air duct, and dispersed to each cultivation area of ​​the climate chamber through the uniform aperture on the porous back panel, so that the cold and hot air can be quickly and evenly diffused in the space, realizing the uniform adjustment of the temperature, humidity and gas composition inside the climate chamber, avoiding the impact of local uneven cold and heat or humidity fluctuations on crop growth.

[0090] The resistance heating wire, refrigeration compressor, condensing fan and circulating fan are all connected to the solid-state relay control module of the control execution unit, and receive specific control instructions issued by the reinforcement learning control unit through the communication unit to realize dynamic intelligent control of the heating, cooling and air circulation processes, ensuring that the internal environmental parameters of the climate chamber are stably maintained within the ideal range according to the requirements of the crop growth stage.

[0091] By adopting the above technical solution, the heating device adopts a resistance heating wire and combines it with the waste heat reuse design of the refrigeration device compressor, which can effectively reduce the energy consumption of heating startup and shorten the temperature rise response time, thereby improving energy utilization. The refrigeration device achieves efficient and stable cooling through the cooperation of the compressor and the condensing fan. The synergistic effect of the porous backplate and the circulating fan enables the internal environmental parameters of the climate chamber to diffuse quickly and evenly in the space, avoiding the adverse effects of local environmental fluctuations on crop growth.

[0092] In one embodiment of the present application, a user interface unit is further included, which provides real-time environmental parameter visualization, a manual control interface and an abnormal alarm function, and supports historical data export and comparative experiment configuration.

[0093] Specifically, in this embodiment, the system also includes a user interface unit, which is installed on the outside of the climate chamber cabinet area and realizes functional interaction by connecting to the industrial computer. The user interface unit is equipped with a high-definition display and a touch control module to provide real-time environmental parameter visualization function. The real-time environmental parameter visualization function receives the environmental data processed by the reinforcement learning control unit through the communication unit, and dynamically displays the temperature, humidity, light intensity, carbon dioxide concentration and oxygen concentration of each culture area inside the climate chamber in the form of graphs, curves and numerical values ​​on the display screen. The user interface unit is provided with a manual control interface. The manual control interface provides separate control switches and parameter adjustment sliders for the heating device, refrigeration device, light adjustment device, fan, oxygen valve and carbon dioxide valve through a graphical operation panel. The user can temporarily adjust the working status or parameters of each device according to actual needs. The manual operation instructions are directly sent to the control execution unit through the communication unit to realize real-time control. At the same time, the user interface unit is provided with an authority management system to ensure that manual operation authority is limited by authorized users.

[0094] The user interface unit also includes an abnormal alarm function. The abnormal alarm function sets upper and lower warning thresholds for real-time environmental parameters. When any environmental parameter is monitored to exceed the safety range or there is a communication failure or equipment abnormality, the user is immediately notified through pop-up windows, sound and light alarms, and log records, prompting the user to intervene in time to ensure crop cultivation safety.

[0095] The user interface unit supports the historical data export function, which can export environmental parameter records by day, week, month or custom time period. The data format supports common formats such as CSV and Excel, which is convenient for users to perform local archiving and subsequent analysis.

[0096] The user interface unit also provides a comparative experiment configuration function, which allows users to set different environmental target parameter groups in different layers of the culture area, set the environmental condition differences between the experimental group and the control group through the interface, and monitor and record the changes in the two groups of data in real time to facilitate scientific experiments and data analysis; all interactive operations of the user interface units are synchronized with the industrial computer and the reinforcement learning control unit through the internal system bus to ensure that the user interface operation is consistent with the actual operating status of the climate chamber, avoiding data lag or operation failure.

[0097] By adopting the above technical solution, the user interface unit can effectively improve the operability and safety of the system by providing real-time environmental parameter visualization, manual control interface and abnormal alarm function. Users can intuitively understand the changes in the internal environment of the climate chamber and intervene and adjust as needed. The abnormal alarm mechanism ensures timely risk warning of the crop cultivation process. At the same time, the historical data export and comparative experiment configuration functions provide convenient data support and experimental basis for scientific research and environmental parameter optimization.

[0098] In one embodiment of the present application, the Internet of Things gateway controls the power on and off of each device in the execution unit through a solid-state relay.

[0099] Specifically, in this embodiment, the power on / off control is realized between the Internet of Things gateway and each device of the control execution unit through a solid-state relay. The Internet of Things gateway is provided with multiple digital output interfaces inside, each digital output interface is connected to the input end of the corresponding solid-state relay, and the output end of the solid-state relay is respectively connected to the power circuit of the heating device, refrigeration device, lighting adjustment device, fan, oxygen valve and carbon dioxide valve. After receiving the control instruction issued by the reinforcement learning control unit or the industrial computer, the Internet of Things gateway parses the instruction content and switches the corresponding digital output interface between high and low levels according to the instruction requirements. When the Internet of Things gateway outputs a high-level signal, the corresponding solid-state relay closes the circuit and starts the corresponding control execution unit device. When the Internet of Things gateway outputs a low-level signal, the corresponding solid-state relay circuit is disconnected and the corresponding control execution unit device is turned off, thereby realizing precise management of the power on and off of each control device.

[0100] To ensure the reliability of control signal transmission and the timeliness of action execution, the digital output control of the IoT gateway adopts a photoelectric isolation design to effectively prevent false operations caused by external electrical interference. At the same time, a state feedback module is set up inside the IoT gateway to detect the output status of the solid-state relay in real time, and feedback is sent to the industrial computer and reinforcement learning control unit through the communication unit to ensure correct closed-loop confirmation of the execution action.

[0101] By adopting the above technical solution, the IoT gateway controls the power on and off of each device in the control execution unit through solid-state relays. This not only greatly simplifies the control circuit structure of the climate chamber and improves the system response speed and stability, but also improves the safety and life of the entire system through the high reliability and protection function of the solid-state relays.

[0102] In one embodiment of the present application, the training process of the reinforcement learning control unit includes:

[0103] Initialize model parameters based on historical data or simulation environment;

[0104] Generate adaptive control strategies through environment state acquisition, action generation, reward calculation and strategy iterative optimization;

[0105] When the reward value converges or reaches the preset number of iterations, the optimal control strategy is output.

[0106] Specifically, the training process of the reinforcement learning control unit in this embodiment includes the following steps: first, the model parameters are initialized based on historical data or a simulated environment. The historical data consists of environmental parameters, control actions, and corresponding environmental change results recorded during the previous operation of the climate chamber. The simulated environment is constructed by a preset climate chamber physical model and a crop growth model. The reinforcement learning control unit performs pre-training based on the historical data or the simulated environment, initializes the weight parameters, learning rate, discount factor, and other core training hyperparameters of the neural network, and sets the reward function:

[0107]

[0108] As the training optimization target; during the training process, the reinforcement learning control unit first obtains the current environment state through the environment state acquisition module , environmental status Including normalized real-time parameters such as temperature, humidity, light intensity, carbon dioxide concentration and oxygen concentration, and then generating a control action based on the current strategy network , control action Specific control instructions for heating devices, cooling devices, light adjustment devices, fans, oxygen valves and carbon dioxide valves; control actions After execution in the simulation environment or historical environment response model, the system calculates the new environment state based on the execution results and calculates the immediate reward value based on the reward function The reward value comprehensively considers the proximity of environmental parameters, action energy consumption and response speed. The reinforcement learning control unit stores the current state, action, reward value and new state as training samples in the experience replay pool. The policy network and value network are gradient updated by sampling the training samples in the experience pool. The reinforcement learning optimization algorithm is iteratively trained and the network parameters are continuously adjusted to improve the policy performance.

[0109] The training process continues. When the cumulative reward value reaches the preset convergence standard or the number of training rounds reaches the preset number of iterations, the reinforcement learning control unit determines that the training is complete and outputs the optimal control strategy that finally converges. The optimal control strategy has the ability to adaptively generate control instructions based on the actual environmental status of the climate chamber, and can dynamically adjust the internal temperature, humidity, light and gas concentration of the climate chamber in real time to meet various needs during the crop growth stage.

[0110] Using the above technical solution, the reinforcement learning control unit initializes the model parameters based on historical data or simulation environment, and forms an adaptive control strategy through environmental state collection, action generation, reward calculation and policy iterative optimization. It outputs the optimal control strategy after the reward converges or reaches the preset number of iterations. This can not only effectively improve the training efficiency and strategy reliability of the climate chamber intelligent control system, but also ensure that the final generated control strategy has good environmental adaptability.

[0111] The climate box structure of the present invention is as follows Figure 2 As shown, the climate box includes a cabinet area 10, a groove area, an internal assembly area, and a culture area 12. Each area is interconnected by wires to ensure stable operation of the climate box.

[0112] To further explain, various sensor power supply units, industrial computers, display screens 20 and Internet of Things gateways are installed and fixed inside the cabinet area 10, and it is responsible for routing the wires related to various control devices.

[0113] The positive and negative power cables for each sensor are consolidated and fed into a single power plug. The 485 communication A and B cables used for data transmission are also consolidated and organized in a slotted guide rail wiring box. All A cables are connected to the A port of the IoT gateway, and all B cables are connected to the B port of the IoT gateway. This creates a neat and tidy wiring arrangement, making subsequent maintenance easier.

[0114] The industrial computer's power cord connects to the recessed area through a wiring box, providing independent power supply, while the network cable connects directly to the IoT gateway. This separate wiring method facilitates troubleshooting later.

[0115] The display screen 20 is powered in the same way as the industrial computer, using an independent power supply and a normally open design to prevent damage to the screen due to frequent starts and stops.

[0116] The gateway is clearly separated from the industrial computer, display screen 20, and sensors, and uses independent power supplies to ensure stable operation of each control device. Multiple control line outputs are connected to the recessed area through a wiring box and then connected to the live and neutral wires of each electrical appliance.

[0117] A water storage barrel fixed by an iron frame is provided at the lower part of the interior of the cabinet area 10 to provide a stable water source for the humidifier to ensure the normal operation of the humidification function.

[0118] To further explain, the recessed area is primarily used to connect the hot, neutral, and ground wires of each appliance to the control output DO line, as well as the solid-state relay protection circuits required by each appliance, enabling gateway control of the appliances. Label the hot, neutral, and ground wires for each appliance and connect them to the output of the solid-state relay. Then, connect the control output DO line from cabinet area 10 to the relay input. Once the circuit modification is complete, the DO port's open and closed status can be controlled via the IoT gateway, enabling remote and intelligent control of the appliances within the cabinet.

[0119] The internal assembly area is mainly responsible for the installation of various electrical appliances to ensure the function and smooth operation of the climate chamber. The specific installation sequence is as follows:

[0120] (1) Refrigeration compressor and condensing fan: The climate chamber is equipped with two refrigeration compressors to cool the chamber. Since the compressors have high power, they need to be used with condensing fans to dissipate heat during operation to ensure long-term stable operation.

[0121] (2) Heating resistors: Two sets of heating resistors are used for cabinet heating and are installed above the refrigeration compressor. After refrigeration is completed, the system can quickly switch to heating mode to achieve constant temperature control. This layout also allows the compressor's electrical waste heat to preheat the resistors, thereby saving energy and reducing consumption.

[0122] (3) Circulation fan: The circulation fan is installed above the refrigeration compressor and the heating resistor, and near the bottom of the incubation area 12 of the box. Its function is to push the air from the bottom to the top, and at the same time, part of the air flow is blown out through the small holes on the back panel to achieve rapid and uniform diffusion of cold air and hot air, and maintain good air circulation in the box.

[0123] (4) Humidifier: The humidifier is installed behind the porous back panel in the cabinet and runs through the entire back panel. Its water source is connected to the water tank in the right cabinet through a water pipe. The coordinated operation of the fan can enhance the water mist diffusion effect and improve the humidification efficiency.

[0124] Cultivation area 12 is the planting area for experimental crops. This area can be divided into experimental spaces by adding iron plates to achieve comparative experiments. The specific composition is as follows:

[0125] (1) Light board: Two detachable light boards are installed on the top of the culture area 12, which can be switched to different light sources according to experimental requirements. A gap is left between the two light boards, and a black cloth partition can be added to form left and right control groups. Each iron plate is used to place experimental crops, and a light board can be installed under the iron plate to provide light for the next layer.

[0126] (2) Side panel hooks: The hooks on the side panels are not only used to fix the iron plates and expand the crop planting area, but can also be used to hang sensor extension probes for environmental monitoring.

[0127] (3) Side panel air holes: The side panels are designed with air holes to facilitate ventilation and adjustment of carbon dioxide and oxygen to maintain the required gas concentration.

[0128] (4) Porous back panel: The porous back panel is used for fan air supply, so that the air generated by humidification, cooling and heating is evenly distributed throughout the entire box. At the same time, the back panel holes also promote gas circulation and can discharge the original gas, thereby regulating the box environment.

[0129] (5) Glass door: The door glass of the climate chamber is covered with a dimming film, which can adjust the reflective or light-transmitting properties of the glass as needed, making it easy to observe the interior and preventing external light from interfering with the experimental environment when no observation is required.

[0130] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. All equivalent structural transformations made by using the contents of the present invention description and drawings under the inventive concept of the present invention, or direct / indirect application in other related technical fields are included in the patent protection scope of the present invention.

Claims

1. An artificial intelligence climate chamber adaptive control system based on reinforcement learning algorithm, characterized in that: include: A data acquisition unit for collecting multiple environmental parameters and crop growth stages in the climate chamber in real time; a reinforcement learning control unit that autonomously learns and generates a control strategy based on the environmental parameters, crop growth stage, real-time environmental changes, and resource consumption status through a reward function that comprehensively optimizes environmental parameter proximity, energy consumption, and response speed; a control execution unit, configured to adjust environmental parameters within the climate chamber according to the control strategy; The communication unit is used to realize data transmission and interaction between various units. The communication unit includes an IoT gateway and an industrial computer. The IoT gateway connects to various sensors via the Modbus protocol and communicates with the industrial computer via the TCP protocol to realize the transmission of sensor data and control instructions. The data acquisition unit, the reinforcement learning control unit, and the control execution unit form a closed-loop feedback through the communication unit. The reward function of the reinforcement learning control unit is: in, Indicates that the status Next action The immediate reward value obtained after Indicates the current environmental status, including real-time parameters such as temperature, humidity, light, and gas concentration; Indicates control actions; Indicates the environmental parameter deviation weight coefficient; is the current environment parameter; is the ideal value for the corresponding crop growth stage; Indicates the number of types of environmental parameters monitored; represents the energy consumption weight coefficient; E(a) is the energy consumption of action a; represents the response speed weight coefficient; δ(s) is the response speed reward; the reinforcement learning control unit also includes: a crop growth stage identification module, which is used to determine the current crop growth stage based on the historical data of environmental parameters and the preset crop growth model, and dynamically adjust the ideal value in the reward function and weight coefficient , , ; Increase the weight of environmental parameter proximity during the seedling stage , increase the weight of energy consumption optimization during the growth period , increase the response speed weight during the flowering and fruiting period ; Among them, the growth stage identification module cooperates with the data acquisition unit and the user interface unit to match the real-time leaf age, plant height and environmental accumulation through the crop growth model, determine the stage switching threshold, and trigger the stage-by-stage adaptive update of the control strategy; the reinforcement learning control unit generates a control strategy that minimizes energy consumption and meets the response speed according to the dynamic changes of environmental parameters and the target values ​​of the crop growth stage, and realizes adaptive adjustment of the climate chamber environment through the control execution unit.

2. The artificial intelligence climate chamber adaptive control system based on reinforcement learning algorithm according to claim 1, characterized in that: The data acquisition unit includes a temperature sensor, a humidity sensor, a light sensor, a carbon dioxide concentration sensor and an oxygen concentration sensor. The temperature sensor, humidity sensor, light sensor, carbon dioxide concentration sensor and oxygen concentration sensor constitute a sensor group. Each layered structure of the culture area in the climate box is provided with a sensor group.

3. The artificial intelligence climate chamber adaptive control system based on reinforcement learning algorithm according to claim 1, characterized in that: The control execution unit includes a heating device, a cooling device, a light adjustment device, a fan, an oxygen valve and a carbon dioxide valve. The heating device and the cooling device achieve uniform diffusion of hot and cold air through a circulating fan, and the oxygen valve and the carbon dioxide valve adjust the gas concentration through the air holes on the side panels.

4. The artificial intelligence climate chamber adaptive control system based on reinforcement learning algorithm according to claim 3, characterized in that: The heating device is a resistance heating wire, and the refrigeration device is a compressor and a condensing fan. The heating wire reuses the waste heat of the compressor, and the environmental parameters are uniformly adjusted through the porous back plate and the circulating fan.

5. The artificial intelligence climate chamber adaptive control system based on reinforcement learning algorithm according to claim 1, characterized in that: It also includes a user interface unit, which provides real-time environmental parameter visualization, a manual control interface and abnormal alarm functions, and supports historical data export and comparative experiment configuration.

6. The artificial intelligence climate chamber adaptive control system based on reinforcement learning algorithm according to claim 1, characterized in that: The Internet of Things gateway controls the power on and off of each device in the execution unit through a solid-state relay.

7. The artificial intelligence climate chamber adaptive control system based on reinforcement learning algorithm according to claim 1, characterized in that: The training process of the reinforcement learning control unit includes: Initialize model parameters based on historical data or simulation environment; Generate adaptive control strategies through environment state acquisition, action generation, reward calculation and strategy iterative optimization; When the reward value converges or reaches the preset number of iterations, the optimal control strategy is output.

Images