An intelligent premixing control method and device based on machine learning, equipment and medium

Abstract

The application provides a kind of intelligent premix control method, device, equipment and medium based on machine learning, related to the technical field of machine learning, method includes: by constructing machine learning model based on historical environmental control data, plant growth data and environmental disturbance response data training, embedded into control system, realize the intelligent preliminary adjustment and closed-loop feedback control of premix box environment;When environmental parameter is stable, stable gas is transported to plant box by air duct, and dynamic compensation is carried out on environmental change in conveying path based on model;Plant box is deployed sensor to monitor disturbance data and input into disturbance adaptation submodule, predicts disturbance trend and outputs feedforward compensation data;When the internal environment of plant box is abnormal, the system automatically exhausts and supplements stable gas, while dynamically adjusting control strategy, and builds a whole-process adaptive regulation mechanism.The application can realize the adaptive pre-regulation of breeding environment.

Description

Technical Field

[0001] This application relates to the technical field of machine learning, specifically to a machine learning-based intelligent premixed control method, device, equipment, and medium. Background Technology

[0002] In the process of breeding new varieties, the traditional premixed control method mainly sets target values ​​for environmental parameters such as temperature, humidity and carbon dioxide concentration, and directly imports the set target parameters into the breeding plant box. It relies on environmental sensors deployed in the box to collect and feedback the actual environmental data in real time. The control system continuously executes a closed-loop adjustment strategy based on the deviation between the actual environmental values ​​in the box and the set target values ​​to drive the control components to correct the premixed gas. This forms a traditional control closed loop that relies on a combination of static setting parameters and passive feedback adjustment. This method has significant deficiencies in response speed and stability when facing scenarios with frequent environmental disturbances and high sensitivity of the breeding process, making it difficult to achieve high-precision adaptive control of complex and variable breeding environments.

[0003] Because plants are highly sensitive to changes in environmental conditions, even small fluctuations in temperature, humidity, or carbon dioxide concentration within the breeding chamber can stimulate plant growth, leading to inconsistent stress responses at different growth stages and disrupting physiological metabolic rhythms. If such environmental disturbances are not controlled promptly and precisely, they can cause abnormalities in plant growth rate and direction, directly affecting the balanced development of its morphology and structure. This manifests as decreased plant quality, significant differences in plant height, uneven leaf distribution, and inconsistent fruit ripening times, severely restricting the stability of breeding efforts and the consistency of new varieties' appearance and commercial potential. Therefore, a method is needed to achieve adaptive pre-control of the breeding environment. Summary of the Invention

[0004] This application provides a machine learning-based intelligent premixed control method, device, equipment, and medium that can achieve adaptive pre-regulation of the breeding environment.

[0005] A first aspect of this application provides a machine learning-based intelligent premixed control method, the method comprising:

[0006] A machine learning model was constructed based on historical environmental control data, plant growth data, and environmental disturbance response data.

[0007] The machine learning model is embedded in the control system, and the control system controls the adjustment components to make preliminary adjustments to the environment inside the premixing box according to the target environmental parameters.

[0008] The first real-time environmental data is acquired by sensors deployed in the premixing tank and fed back to the control system to form a closed-loop regulation;

[0009] When the environmental parameters in the premix box are stable within the target range, the air duct is opened to deliver the stable environment corresponding to the environmental parameters to the plant box at a constant flow rate. At the same time, the changes in environmental parameters in the delivery path are compensated and corrected in real time based on the machine learning model.

[0010] The disturbance data monitored in real time by the sensors inside the plant box is input into the disturbance adaptation submodule in the machine learning model to predict the disturbance trend and output environmental disturbance compensation data.

[0011] When the second real-time environmental parameter inside the plant box exceeds the set threshold, the control system drives the exhaust mechanism to discharge the air from the plant box, while controlling the premix box to output a new round of stable environmental gas, and adaptively adjusting the control strategy of the premix box based on the environmental disturbance compensation data.

[0012] Based on the above technical solutions, preferably, the step of constructing a machine learning model based on historical environmental control data, plant growth data, and environmental disturbance response data specifically includes:

[0013] Historical environmental control data of plants were collected over multiple growth cycles. The historical environmental control data included environmental parameter setpoints, actual environmental parameter values, environmental parameter adjustment ranges, environmental parameter adjustment response times, environmental parameter adjustment frequencies, and environmental parameter stability indicators.

[0014] Collect plant growth data within the time period corresponding to the historical environmental control data. The plant growth data includes plant species, growth stage, plant height, leaf area index, chlorophyll content, dry matter and growth rate.

[0015] Collect environmental disturbance response data recorded during plant growth, including disturbance type, disturbance amplitude, disturbance duration, disturbance recovery time, and disturbance regulation feedback characteristics;

[0016] The historical environmental control data, plant growth data and environmental disturbance response data are time-aligned to construct a unified training sample dataset.

[0017] A deep neural network is used to train the training sample dataset to form a machine learning model for outputting target environmental parameters and environmental disturbance compensation data. This model comprises a multi-layer feedforward neural network with an input layer, multiple hidden layers, and an output layer. The input layer receives plant species, growth stages, historical sequences of environmental parameters, and phenotypic features. The output layer generates the target environmental parameters and disturbance compensation strategies. The training process of the machine learning model employs a supervised learning mechanism. The objective function is set to minimize the mean square error between the predicted environmental parameters and the actual adjustment results. A regularization term is introduced to constrain the complexity of the controller parameters and prevent overfitting. Cross-validation is used to evaluate the model's generalization ability during training, and hyperparameter grid search is used to optimize the model structure and control parameters.

[0018] Based on the above technical solutions, preferably, the step of embedding the machine learning model into the control system, and the control system controlling the adjustment components to perform preliminary adjustment of the environment inside the premixing chamber according to the target environmental parameters, specifically includes:

[0019] The machine learning model is deployed in the core control unit of the control system, which includes a data receiving module, a model parsing module, a control signal generation module, and an execution feedback module.

[0020] The data receiving module is used to receive the target environmental parameters output by the machine learning model, and the model parsing module generates corresponding adjustment instructions based on the target environmental parameters. The adjustment instructions include temperature adjustment instructions, humidity adjustment instructions and carbon dioxide concentration adjustment instructions.

[0021] The control signal generation module outputs control signals to the temperature regulation component, humidity regulation component, and carbon dioxide concentration regulation component according to the regulation command, so as to drive the regulation components to perform regulation operations;

[0022] The execution feedback module collects current environmental parameters based on environmental parameter sensors deployed inside the premixing box. The current environmental parameters include the current temperature value, the current humidity value, and the current carbon dioxide concentration value.

[0023] The error between the current environmental parameters and the target environmental parameters is calculated, and the control signal is corrected in real time based on the error to form a closed-loop regulation control logic, so that the environmental parameters in the premixing box are stabilized to the target environmental parameters.

[0024] Based on the above technical solutions, preferably, the step of acquiring first real-time environmental data through sensors deployed in the premixing box and feeding it back to the control system to form a closed-loop regulation specifically includes:

[0025] An environmental parameter sensor module is deployed in multiple spatial locations inside the premixing chamber. The environmental parameter sensor module includes a temperature sensor, a humidity sensor, and a carbon dioxide concentration sensor.

[0026] The environmental parameter sensor module collects first real-time environmental parameters at a preset sampling period. The first real-time environmental parameters include a first real-time temperature value, a first real-time humidity value, and a first real-time carbon dioxide concentration value.

[0027] The first real-time environmental parameters are packaged into an environmental parameter data frame and sent to the control system through the data transmission module;

[0028] The control system calculates the error between the environmental parameter data and the target environmental parameters to obtain the temperature error, humidity error, and carbon dioxide concentration error. Based on the proportional-integral-derivative control algorithm, it generates control signals to drive the temperature regulation component, humidity regulation component, and carbon dioxide concentration regulation component to perform regulation operations, forming a closed-loop regulation path with the environmental parameter sensor module as the sensing end, the control system as the decision end, and the regulation components as the execution end.

[0029] Based on the above technical solution, preferably, when the environmental parameters inside the premixing box are stable within the target range, the air duct is opened to deliver the stable environment corresponding to the environmental parameters to the plant box at a constant flow rate. Simultaneously, based on the machine learning model, real-time compensation and correction are performed on changes in environmental parameters along the delivery path, specifically including:

[0030] When the control system determines that the first real-time temperature value, the first real-time humidity value, and the first real-time carbon dioxide concentration value are all within the error range of the target temperature parameter, the target humidity parameter, and the target carbon dioxide concentration parameter, it outputs a duct opening command to drive the duct gas delivery mechanism to start. The duct gas delivery mechanism includes a blower, an airflow control valve, and a pressure stabilization module.

[0031] An environmental parameter monitoring module for the transport path is installed along the inside of the air duct. The environmental parameter monitoring module for the transport path includes a temperature sensor, a humidity sensor, and a carbon dioxide concentration sensor for the transport path, which are used to acquire environmental change data within the transport path.

[0032] The environmental change data is input into the environmental transmission prediction sub-model in the machine learning model. The environmental transmission prediction sub-model generates an environmental compensation instruction. The control system dynamically adjusts the output amplitude of the temperature regulation component, humidity regulation component, and carbon dioxide concentration regulation component according to the environmental compensation instruction, so as to realize real-time compensation and correction of the environmental parameters at the terminal of the transport path.

[0033] Based on the above technical solutions, preferably, the step of inputting the disturbance data monitored in real time by the sensors inside the plant box into the disturbance adaptation submodule of the machine learning model for disturbance trend prediction and outputting environmental disturbance compensation data specifically includes:

[0034] An environmental disturbance monitoring sensor module is deployed at multiple locations inside the plant box. The environmental disturbance monitoring sensor module includes a plant box temperature sensor, a plant box humidity sensor, and a plant box carbon dioxide concentration sensor, which are used to collect real-time temperature disturbance values, real-time humidity disturbance values, and real-time carbon dioxide concentration disturbance values ​​inside the plant box to form a disturbance data sequence.

[0035] The perturbation data sequence is packaged into a perturbation data input frame and transmitted to the perturbation adaptation submodule of the machine learning model. The perturbation adaptation submodule performs perturbation trend prediction based on the neural network structure of time series modeling and outputs the perturbation trend prediction result.

[0036] Environmental disturbance compensation data is generated based on the disturbance trend prediction results. The environmental disturbance compensation data includes the pre-adjustment amplitude, adjustment rate and duration for controlling the temperature regulation component, humidity regulation component and carbon dioxide concentration regulation component, so as to achieve feedforward compensation correction of the disturbance.

[0037] Based on the above technical solutions, preferably, when the second real-time environmental parameter inside the plant box exceeds a set threshold, the control system drives the exhaust mechanism to discharge the air from the plant box, simultaneously controls the premixing box to output a new round of stable environmental gas, and adaptively adjusts the control strategy of the premixing box based on the environmental disturbance compensation data, specifically including:

[0038] The environmental parameter judgment module receives second real-time environmental parameters collected by the plant box temperature sensor, the plant box humidity sensor, and the plant box carbon dioxide concentration sensor. The second real-time environmental parameters include a second real-time temperature value, a second real-time humidity value, and a second real-time carbon dioxide concentration value. The module then compares the real-time temperature value, the real-time humidity value, and the real-time carbon dioxide concentration value with the error thresholds of the target temperature parameter, the target humidity parameter, and the target carbon dioxide concentration parameter.

[0039] When any of the second real-time environmental parameters exceeds the corresponding error threshold range, the control system outputs an exhaust command to drive the exhaust fan to discharge non-standard environmental gases from the plant box through the exhaust channel.

[0040] At the same time, the control system outputs a gas replenishment command, driving the premix box to generate a new round of stable environmental gas based on the current target temperature parameters, target humidity parameters, and target carbon dioxide concentration parameters, and delivers it to the plant box through the air duct;

[0041] The control system further inputs environmental disturbance compensation data into the strategy adjustment module to adaptively optimize parameters such as the adjustment cycle, adjustment amplitude, sensor sampling density, and control signal gain of the current premix box, forming a control strategy with enhanced dynamic response.

[0042] A second aspect of this application provides a machine learning-based intelligent premixing control device, the device being used to execute a machine learning-based intelligent premixing control method as described in any of the above-described methods, the device comprising an acquisition module, a processing module, and an output module, wherein:

[0043] The acquisition module is used to construct a machine learning model based on historical environmental control data, plant growth data, and environmental disturbance response data.

[0044] The processing module is used to embed the machine learning model into the control system, and the control system controls the adjustment component to make preliminary adjustments to the environment inside the premix box according to the target environmental parameters.

[0045] The processing module is used to acquire first real-time environmental data through sensors deployed in the premixing box and feed it back to the control system to form a closed-loop regulation;

[0046] The processing module is used to open the air duct when the environmental parameters in the premix box are stable within the target range, and to deliver the stable environment corresponding to the environmental parameters to the plant box at a constant flow rate. At the same time, it performs real-time compensation and correction on the changes in environmental parameters in the delivery path based on the machine learning model.

[0047] The processing module is used to input the disturbance data monitored in real time by the sensors inside the plant box into the disturbance adaptation submodule in the machine learning model to predict the disturbance trend and output environmental disturbance compensation data.

[0048] The output module is used to drive the exhaust mechanism to discharge the air from the plant box when the second real-time environmental parameter inside the plant box exceeds a set threshold, and at the same time control the premix box to output a new round of stable environmental gas, and adaptively adjust the control strategy of the premix box based on the environmental disturbance compensation data.

[0049] A third aspect of this application provides an electronic device including a processor, a memory, a user interface, and a network interface, wherein the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any of the foregoing.

[0050] A fourth aspect of this application provides a computer-readable storage medium storing instructions that, when executed, perform the method described in any of the preceding descriptions.

[0051] In summary, one or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0052] 1. This application constructs a machine learning model trained based on historical environmental control data, plant growth data, and environmental disturbance response data, and embeds it into the control system. This enables dynamic adjustment of the premixing chamber environment, real-time compensation for changes in the transport path environment, and predictive response to disturbance trends in the plant chamber. When the plant chamber environment is abnormal, the control system can adaptively adjust the adjustment strategy in conjunction with disturbance compensation data, forming a closed-loop mechanism of perception-prediction-compensation-control. This enables the system to have feedforward adjustment capability and adaptive response capability, effectively solving the problems of delayed response to environmental disturbances and insufficient control precision in the traditional mode, and ultimately realizing adaptive pre-control of the breeding environment.

[0053] 2. By collecting historical environmental control data, plant growth data, and environmental disturbance response data over multiple periods, and training them with deep neural networks, a machine learning model with multi-factor joint reasoning capabilities was constructed. This enabled accurate prediction of target environmental parameters and disturbance compensation strategies, effectively improving the personalized adaptability and forward-looking response capability of environmental regulation, and providing a highly reliable data-driven foundation for subsequent environmental control.

[0054] 3. By embedding machine learning models into the core control unit of the control system, a closed-loop control path is established from receiving target environmental parameters, parsing instructions, generating control signals to executing feedback corrections. This ensures that the environmental regulation within the premixing chamber has full-process automated regulation capabilities, significantly improving regulation response efficiency, control accuracy, and environmental stability.

[0055] 4. By deploying multiple environmental parameter sensor modules inside the premixing chamber and implementing feedback regulation based on proportional-integral-derivative control algorithms, a local closed-loop control chain of perception-decision-execution is constructed. This chain can identify minute disturbances at high frequency and precisely adjust the control output, thereby significantly improving the environmental stability of the premixing chamber and establishing a highly consistent gas foundation for subsequent gas delivery.

[0056] 5. After determining that the environmental parameters inside the premix box are stable, the air duct is opened, and an environmental parameter monitoring module is deployed in the delivery path. An environmental transmission prediction sub-model is introduced for real-time compensation and correction, so as to achieve consistent control of the gas from the source to the plant box. This effectively avoids problems such as temperature decay, humidity loss and gas concentration deviation during the delivery process, and ensures that the actual receiving environment of the plant is highly matched with the set value.

[0057] 6. By deploying environmental disturbance monitoring sensor modules inside the plant box and using the disturbance adaptation submodule in the neural network model to predict disturbance trends, the system can achieve dynamic perception and trend modeling before disturbances occur, and generate environmental disturbance compensation data based on the prediction results, thereby realizing feedforward regulation and enhancing the timeliness and proactivity of environmental disturbance response.

[0058] 7. When abnormal fluctuations occur in the environmental parameters inside the plant box, non-standard environmental gases are discharged through the exhaust mechanism. At the same time, the premix box is controlled to replenish a new round of stable environmental gases. Based on the disturbance compensation data, the control strategy parameters are dynamically optimized to form a disturbance closed-loop response and adaptive strategy reconstruction mechanism, which significantly improves the system's steady-state maintenance capability and anti-disturbance recovery capability, and effectively ensures that the plant growth environment is continuously in an optimized state. Attached Figure Description

[0059] Figure 1 This is a flowchart illustrating a machine learning-based intelligent premixed control method disclosed in an embodiment of this application.

[0060] Figure 2 This is a schematic diagram of a machine learning-based intelligent premixed control device disclosed in an embodiment of this application;

[0061] Figure 3 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of this application.

[0062] Explanation of reference numerals in the attached drawings: 201, acquisition module; 202, processing module; 203, output module; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. Detailed Implementation

[0063] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0064] In the description of the embodiments of this application, the words "for example" or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design that is described as "for example" or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design options. Rather, the use of the words "for example" or "for instance" is intended to present the relevant concepts in a specific manner.

[0065] In the description of the embodiments of this application, the term "multiple" means two or more. For example, multiple systems means two or more systems, and multiple screen terminals means two or more screen terminals. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0066] In the process of new variety breeding, traditional premixed control methods rely on a closed-loop mechanism of static setting and passive feedback regulation, which makes it difficult to achieve rapid response and precise control in complex and perturbed environments. Especially in the context of plants being highly sensitive to temperature, humidity and carbon dioxide concentration, any slight fluctuation may cause plant physiological rhythm disorder and asynchronous stress response, resulting in significant differences in plant morphology, development rhythm and fruit ripening, which in turn affects the phenotypic consistency and commercial potential of breeding work. Therefore, there is an urgent need for an adaptive environmental regulation method with dynamic prediction and feedforward control capabilities to achieve precise and stable control in highly sensitive breeding environments.

[0067] This embodiment discloses an intelligent premixed control method based on machine learning, referring to... Figure 1 This includes the following steps S110-S160:

[0068] S110 is a machine learning model built based on historical environmental control data, plant growth data, and environmental disturbance response data.

[0069] This application discloses a machine learning-based intelligent premixed control method applied to a server. The server includes, but is not limited to, electronic devices such as mobile phones, tablets, wearable devices, and PCs (Personal Computers), and can also be a backend server running a machine learning-based intelligent premixed control method. The server can be implemented using a standalone server or a server cluster composed of multiple servers.

[0070] In one possible implementation, a machine learning model is constructed based on historical environmental control data, plant growth data, and environmental disturbance response data. Specifically, this includes: collecting historical environmental control data for plants over multiple growth cycles, including environmental parameter setpoints, actual environmental parameter values, environmental parameter adjustment amplitudes, environmental parameter adjustment response times, environmental parameter adjustment frequencies, and environmental parameter stability indices; collecting plant growth data within the corresponding time periods of the historical environmental control data, including plant species, growth stage, plant height, leaf area index, chlorophyll content, dry matter mass, and growth rate; collecting environmental disturbance response data recorded during plant growth, including disturbance type, disturbance amplitude, disturbance duration, disturbance recovery time, and disturbance regulation feedback characteristics; and combining the historical environmental control data and plant growth data with the environmental disturbance response data. Plant growth data and environmental disturbance response data were time-aligned to construct a unified training sample dataset. A deep neural network was used to train the training sample dataset to form a machine learning model for outputting target environmental parameters and environmental disturbance compensation data. The model consisted of a multi-layer feedforward neural network containing an input layer, multiple hidden layers, and an output layer. The input layer received plant species, growth stage, historical sequences of environmental parameters, and phenotypic features. The output layer generated target environmental parameters and disturbance compensation strategies. The training process of the machine learning model adopted a supervised learning mechanism. The objective function was set to minimize the mean square error between the predicted environmental parameters and the actual adjustment results. At the same time, a regularization term was introduced to constrain the complexity of the controller parameters and prevent the model from overfitting. Cross-validation was used to evaluate the model's generalization ability during training, and the model structure and control parameters were optimized through hyperparameter grid search.

[0071] Specifically, firstly, historical environmental control data of the plants over multiple growth cycles is collected. This historical environmental control data is obtained by recording the entire process of temperature, humidity, and carbon dioxide concentration parameters during regulation within the control system over a long period. Specifically, this includes the setpoint, actual value, adjustment range, response time, adjustment frequency, and stability index of each environmental parameter at each moment. This historical environmental control data needs to be exported from the environmental regulation log system through a standardized data interface and uniformly coded with timestamps for subsequent time-series alignment with plant growth data.

[0072] Then, plant growth data were collected for the time periods corresponding to the aforementioned historical environmental control data. This data was acquired using high-frequency growth monitoring equipment and included plant species, growth stage, plant height, leaf area index, chlorophyll content, dry matter mass, and growth rate. All plant growth data were recorded in segments using timestamps and annotated with structured data through a phenotypic analysis platform. To ensure data consistency, plant phenotypic changes at different growth stages needed to be strictly time-matched with the corresponding historical environmental control data to avoid model training errors caused by recording time-series offsets.

[0073] Next, environmental disturbance response data recorded during plant growth were collected. This data primarily originated from an automatic environmental disturbance event identification system, including disturbance type, disturbance amplitude, disturbance duration, disturbance recovery time, and disturbance regulation feedback characteristics. The disturbance type was automatically classified and identified by the system based on the slope and amplitude of environmental parameter changes. The disturbance amplitude and duration were calculated as the maximum deviation and event duration, respectively. The disturbance recovery time was the time required for the control system to readjust the environmental parameters to the stable range. The disturbance regulation feedback characteristics recorded the control strategy response path and the trajectory of control signal changes.

[0074] Subsequently, historical environmental control data, plant growth data, and environmental disturbance response data were uniformly aligned along the time dimension to construct a training sample dataset with consistent structure and synchronized timestamps. Each record in the training sample dataset corresponds to a time segment, including the target environmental state, plant growth state, and corresponding disturbance response. These are organized into multidimensional input samples through a unified data structure framework. All data features are normalized and missing value processing is performed to ensure numerical stability during the training process.

[0075] Finally, a deep neural network was used to train the training sample dataset, constructing a multi-layer feedforward neural network as the core architecture of the machine learning model. The neural network includes an input layer, multiple hidden layers, and an output layer. The input layer receives features including plant species, growth stage, historical sequences of environmental parameters, and plant phenotypic indices. The output layer generates the target environmental parameters and perturbation compensation strategies. The model training process employs a supervised learning mechanism, constructing an objective function to minimize the mean squared error between the predicted environmental parameters and the historical actual adjustment results, as the optimization objective. This is specifically expressed as follows:

[0076]

[0077] in, This represents the total number of training samples. Indicates the first The target environment parameters of each sample are predicted by the neural network. This indicates the historical actual adjustment result corresponding to this sample. This is the predicted loss value of the model. This loss function measures the deviation between the model output and the actual historical control results. By averaging the squares of all sample errors, it effectively amplifies large errors and smooths small errors, enabling the model to more accurately learn the numerical mapping relationship of historical control behavior.

[0078] Meanwhile, an L2 regularization term is introduced into the model loss function to suppress the complexity of the controller parameters and prevent overfitting, as shown below:

[0079]

[0080] in, In a neural network, the first... Each connection weight, This represents the total weight. The regularization coefficient is . This represents a penalty term for controller complexity. By adding this regularization term to the total loss function, excessively large weight values ​​can be effectively suppressed, preventing the model from overfitting the training data, thereby improving the model's generalization ability on unknown data and ensuring the model's adaptability under dynamic environmental disturbances.

[0081] To improve the model's generalization ability, a cross-validation mechanism is introduced during training, and a hyperparameter grid search strategy is used to systematically optimize neural network structure parameters such as the number of hidden layers, the number of nodes, the type of activation function, and the learning rate. This results in a convergent, stable machine learning model with high generalization ability. The final model output prediction formula is expressed as follows:

[0082]

[0083] in, This represents the input feature vector, which includes plant species, growth stage, historical sequences of environmental parameters, and plant phenotypic indicators. and They represent the first Layer weight matrix and bias vector Indicates the first The activation function of the layer, Indicates the number of network layers. This formula represents the target environmental parameters and environmental disturbance compensation strategy output by the model. It fully describes the nonlinear mapping process from input to output of the deep neural network. Through multi-layer nonlinear combinations, the model can capture the complex interaction between the environment and the physiological state of plants, thus providing support for environmental control decisions under complex disturbance scenarios.

[0084] S120 embeds a machine learning model into the control system, which controls the adjustment components to make preliminary adjustments to the environment inside the premixing chamber based on the target environmental parameters.

[0085] In one possible implementation, a machine learning model is embedded in the control system. The control system controls the adjustment components to perform preliminary adjustments to the environment inside the premixing chamber based on target environmental parameters. Specifically, this includes: deploying the machine learning model in the core control unit of the control system, which includes a data receiving module, a model parsing module, a control signal generation module, and an execution feedback module; the data receiving module receives the target environmental parameters output by the machine learning model; the model parsing module generates corresponding adjustment commands based on the target environmental parameters, including temperature adjustment commands, humidity adjustment commands, and carbon dioxide concentration adjustment commands; the control signal generation module outputs control signals to the temperature adjustment components, humidity adjustment components, and carbon dioxide concentration adjustment components based on the adjustment commands to drive the adjustment components to perform adjustment operations; the execution feedback module calculates the error between the current environmental parameters and the target environmental parameters based on the current environmental parameter sensors deployed inside the premixing chamber, including the current temperature value, current humidity value, and current carbon dioxide concentration value, and corrects the control signals in real time based on the error, forming a closed-loop adjustment control logic to stabilize the environmental parameters inside the premixing chamber to the target environmental parameters.

[0086] Specifically, the machine learning model is embedded in the core control unit of the control system. This core control unit serves as the central module for environmental regulation execution, integrating a data receiving module, a model parsing module, a control signal generation module, and an execution feedback module. The machine learning model, acting as the target environmental parameter calculation unit, resides in the computational resource area where the model parsing module is located. The control system triggers the model inference process through an embedded logic invocation mechanism to obtain dynamic environmental regulation decision outputs.

[0087] Secondly, the data receiving module is responsible for receiving the target environmental parameters output by the machine learning model. These parameters are generated by the model based on plant species, growth stage, historical environmental parameter sequences, and plant phenotypic indicators. Specifically, they include target temperature, target humidity, and target carbon dioxide concentration. The data receiving module parses and verifies the format of these target environmental parameters to ensure their structure is valid and their values ​​fall within the set control range, serving as the basic input for subsequent instruction generation.

[0088] Next, the model analysis module constructs standardized adjustment instructions based on the received target environmental parameters. The adjustment instructions are clearly divided into three categories: temperature adjustment instructions, humidity adjustment instructions, and carbon dioxide concentration adjustment instructions. Each type of adjustment instruction encapsulates the control target, execution cycle, and response level of the corresponding environmental adjustment component, which is used to drive the control signal generation module to generate specific execution control quantities.

[0089] Then, the control signal generation module converts the adjustment command into a control signal that can directly drive the physical device. The control signal is sent to the temperature regulation component, humidity regulation component and carbon dioxide concentration regulation component in the premixing box in the form of analog voltage, PWM pulse width modulation or digital communication protocol, driving the refrigeration module, humidification module and gas injection module and other equipment to perform precise environmental adjustment operations according to the target parameters.

[0090] Meanwhile, the execution feedback module is responsible for receiving real-time environmental parameters collected by environmental parameter sensors deployed inside the premixing chamber. These parameters include the current temperature, humidity, and carbon dioxide concentration. To ensure the accuracy of the environmental status, the environmental parameter sensors are arranged in a multi-point layout and equipped with high-precision samplers. The feedback data is transmitted back to the execution feedback module in real time for error analysis.

[0091] Finally, the execution feedback module performs item-by-item error calculation between the current environmental parameters and the target environmental parameters, generating temperature error, humidity error, and carbon dioxide concentration error, and corrects the original control signal in real time based on the proportional-integral-derivative control algorithm; the corrected control signal is re-output to the regulating component to form a continuously iterative closed-loop regulating control logic until the environmental parameters in the premixing chamber are stably maintained within the error tolerance range of the target environmental parameters, thereby achieving precise initial regulation of the environment.

[0092] S130 acquires first real-time environmental data through sensors deployed in the premixing tank and feeds it back to the control system to form a closed-loop regulation.

[0093] In one possible implementation, a closed-loop regulation is formed by acquiring first real-time environmental data through sensors deployed within the premixing chamber and feeding it back to the control system. Specifically, this includes: deploying environmental parameter sensor modules at multiple spatial locations within the premixing chamber; each environmental parameter sensor module including a temperature sensor, a humidity sensor, and a carbon dioxide concentration sensor; the environmental parameter sensor modules collecting first real-time environmental parameters at a preset sampling period; the first real-time environmental parameters including a first real-time temperature value, a first real-time humidity value, and a first real-time carbon dioxide concentration value; packaging the first real-time environmental parameters into environmental parameter data frames and sending them to the control system through a data transmission module; the control system calculating the errors between the environmental parameter data and the target environmental parameters to obtain temperature error, humidity error, and carbon dioxide concentration error, and generating control signals based on a proportional-integral-derivative (PID) control algorithm to drive the temperature regulation component, humidity regulation component, and carbon dioxide concentration regulation component to perform regulation operations, thus forming a closed-loop regulation path with the environmental parameter sensor modules as the sensing end, the control system as the decision-making end, and the regulation components as the execution end.

[0094] Specifically, environmental parameter sensor modules, consisting of temperature, humidity, and carbon dioxide concentration sensors, are deployed at multiple representative locations within the premixing chamber. The deployment locations need to be optimized based on the airflow path, mixing uniformity, and heat exchange efficiency within the premixing chamber to ensure that the collected environmental parameters are representative, spatially comprehensive, and responsive, thus supporting high-precision control decisions.

[0095] Secondly, the environmental parameter sensor module periodically collects the first real-time environmental parameters according to the sampling period set by the control system. The first real-time environmental parameters include the first real-time temperature value, the first real-time humidity value, and the first real-time carbon dioxide concentration value. Each first real-time environmental parameter is accompanied by timestamp information to form a continuous time series for subsequent error calculation and trend assessment. All first real-time environmental parameters are submitted to the data transmission module immediately after being collected.

[0096] Next, the data transmission module packages the first real-time environmental parameters output by each sensor module into a structured environmental parameter data frame. This frame includes fields such as sensor identifier, environmental parameter type, parameter value, and acquisition timestamp. The environmental parameter data frame is then sent to the control system's data receiving channel using a standard communication protocol, ensuring complete, accurate, and high-speed transmission of the data frame even in complex communication environments.

[0097] Then, the control system parses the received environmental parameter data frames and extracts the first real-time temperature value, first real-time humidity value, and first real-time carbon dioxide concentration value. It then performs item-by-item error calculations against the target temperature parameter, target humidity parameter, and target carbon dioxide concentration parameter output by the machine learning model. The error calculation uses a difference model to calculate the temperature error, humidity error, and carbon dioxide concentration error, which serve as the basic input for generating the control quantity.

[0098] Finally, based on the proportional-integral-derivative (PID) control algorithm, the control system generates control signals using the calculated temperature error, humidity error, and carbon dioxide concentration error. These control signals are then output in real-time by the control signal generation module to the temperature regulation component, humidity regulation component, and carbon dioxide concentration regulation component, driving each component to correct and adjust the environmental parameters inside the premixing chamber. This process constitutes a complete closed-loop control path. The closed-loop control path uses the environmental parameter sensor module as the sensing end, the control system as the decision-making end, and the regulation components as the execution end. Through continuous high-frequency sensing, dynamic decision-making, and precise execution, it achieves dynamic and accurate control of the premixing chamber's environmental parameters, ensuring that they are stably maintained within the set target environmental parameter range.

[0099] S140: When the environmental parameters in the premix box are stable within the target range, the air duct is opened to deliver the stable environment corresponding to the environmental parameters to the plant box at a constant flow rate. At the same time, the changes in environmental parameters in the delivery path are compensated and corrected in real time based on the machine learning model.

[0100] In one possible implementation, when the environmental parameters within the premixing chamber stabilize within the target range, the air duct is opened to deliver the stable environment corresponding to the environmental parameters to the plant box at a constant flow rate. Simultaneously, a machine learning model is used to compensate and correct changes in environmental parameters along the delivery path in real time. Specifically, when the control system determines that the first real-time temperature value, the first real-time humidity value, and the first real-time carbon dioxide concentration value are all within the error range of the target temperature parameter, target humidity parameter, and target carbon dioxide concentration parameter, it outputs an air duct opening command to drive the air duct gas delivery mechanism to start. The air duct gas delivery mechanism includes a blower, an airflow control valve, and a pressure stabilization module. An environmental parameter monitoring module along the delivery path is installed inside the air duct. The environmental parameter monitoring module includes a delivery path temperature sensor, a delivery path humidity sensor, and a delivery path carbon dioxide concentration sensor to acquire environmental change data within the delivery path. The environmental change data is input into the environmental transmission prediction sub-model in the machine learning model. The environmental transmission prediction sub-model generates an environmental compensation command. The control system dynamically adjusts the output amplitude of the temperature regulation component, humidity regulation component, and carbon dioxide concentration regulation component according to the environmental compensation command to achieve real-time compensation and correction of the environmental parameters at the end of the delivery path.

[0101] Specifically, the control system continuously monitors the first real-time environmental parameters, including the first real-time temperature, first real-time humidity, and first real-time carbon dioxide concentration, and compares them item by item with the target temperature, target humidity, and target carbon dioxide concentration parameters. An absolute error calculation method is used to determine whether all the current environmental parameters inside the premixing chamber are within the set error tolerance range. When the control system determines that all the first real-time environmental parameters meet the tolerance conditions, it considers the current environment inside the premixing chamber to have reached a stable state, and the system then issues a duct opening command to enter the conveying stage.

[0102] Secondly, the gas delivery mechanism in the duct starts upon receiving the duct opening command. This mechanism includes a supply fan, an airflow control valve, and a pressure stabilization module. The supply fan provides constant power to drive the flow of ambient gas. The airflow control valve dynamically adjusts its opening according to the constant flow parameters set by the system to control the stable output of gas velocity. The pressure stabilization module suppresses transient pressure fluctuations caused by long-distance pipeline transport or gas disturbances, ensuring that the stable ambient gas forms a controllable and orderly laminar flow delivery state in the duct.

[0103] Next, to accurately acquire information on environmental parameter changes during gas transport within the duct, environmental parameter monitoring modules are deployed at key locations inside the duct. These modules include temperature, humidity, and carbon dioxide concentration sensors along the transport path. They are used to collect real-time data on temperature, humidity, and carbon dioxide concentration changes along the transport path, forming environmental change data. All sensor outputs are organized in time-series format and transmitted to the control system for unified processing via a data channel.

[0104] Subsequently, the control system inputs environmental change data into the environmental transport prediction sub-model within the machine learning model. This sub-model compares the environmental parameters at the premixed container outlet with the real-time environmental parameters at the end of the transport path, and incorporates factors such as transport time delay, pipeline length, thermal conductivity, and gas diffusion patterns to calculate the environmental degradation trend of the gas during transport. The environmental transport prediction sub-model generates environmental compensation commands through a feedforward compensation mechanism, including temperature compensation, humidity correction, and carbon dioxide concentration correction targets.

[0105] Finally, the control system adjusts the output amplitudes of the temperature, humidity, and carbon dioxide concentration adjustment components in real time according to environmental compensation commands, enabling the output environmental gases in the next cycle to have advanced adjustment characteristics. This achieves precise reconstruction of the target environmental parameters at the end of the air duct, i.e., the inlet of the plant box. This compensation and correction process dynamically responds to environmental changes within the transport path, ensuring accurate restoration of the final environmental parameters of the gas after transport. This guarantees that the environment received within the plant box remains consistent with the target of the premixed chamber, forming a high-precision, highly consistent environmental control mechanism.

[0106] The S150 inputs the disturbance data monitored in real time by the sensors inside the plant box into the disturbance adaptation submodule in the machine learning model to predict the disturbance trend and output environmental disturbance compensation data.

[0107] In one possible implementation, disturbance data monitored in real time by sensors inside the plant box is input into a disturbance adaptation submodule in a machine learning model for disturbance trend prediction, and environmental disturbance compensation data is output. Specifically, this includes: deploying environmental disturbance monitoring sensor modules at multiple locations inside the plant box; these modules include plant box temperature sensors, plant box humidity sensors, and plant box carbon dioxide concentration sensors, used to collect real-time temperature disturbance values, real-time humidity disturbance values, and real-time carbon dioxide concentration disturbance values ​​inside the plant box, forming a disturbance data sequence; packaging the disturbance data sequence into a disturbance data input frame and transmitting it to the disturbance adaptation submodule of the machine learning model; the disturbance adaptation submodule performs disturbance trend prediction based on a time-series modeled neural network structure and outputs the disturbance trend prediction result; generating environmental disturbance compensation data based on the disturbance trend prediction result; and generating environmental disturbance compensation data, which includes the pre-adjustment amplitude, adjustment rate, and duration for controlling the temperature adjustment component, humidity adjustment component, and carbon dioxide concentration adjustment component to achieve feedforward compensation correction of the disturbance.

[0108] Specifically, firstly, environmental disturbance monitoring sensor modules are deployed at multiple spatial locations inside the plant enclosure. These modules consist of plant enclosure temperature sensors, humidity sensors, and carbon dioxide concentration sensors, used to collect real-time temperature disturbance values, humidity disturbance values, and carbon dioxide concentration disturbance values ​​within the plant enclosure at high frequency. The deployment locations of the environmental disturbance monitoring sensor modules need to cover different vertical levels within the enclosure and areas where airflow paths intersect, to ensure that the collected environmental disturbance data has spatial representativeness and dynamic response capability, comprehensively reflecting the local environmental disturbance characteristics caused by plant metabolic behavior.

[0109] Secondly, the control system organizes the real-time temperature disturbance values, real-time humidity disturbance values, and real-time carbon dioxide concentration disturbance values ​​continuously collected by the environmental disturbance monitoring sensor module into a disturbance data sequence in chronological order, and packages it into a structured disturbance data input frame. The disturbance data input frame contains fields such as disturbance type identifier, disturbance amplitude information, sampling timestamp, and corresponding sensor number. The disturbance data input frame serves as the time series input to the neural network model and, after data verification, is transmitted to the disturbance adaptation submodule in the machine learning model.

[0110] Next, the disturbance adaptation submodule predicts the disturbance trend of the disturbance data input frame based on a time-series modeling neural network structure. The neural network structure employs a gated recurrent unit network or a long short-term memory network, possessing the ability to model the time-dependent characteristics of disturbances and accurately identify disturbance development patterns and potential change paths. After receiving the disturbance data input frame, the model generates the disturbance trend prediction result through multi-layer recursive operations. The prediction result includes the changing trend of the environmental disturbance amplitude, the slope of the disturbance intensity change, and the possible threshold range of disturbance values ​​to be reached over a future period.

[0111] Then, based on the disturbance trend prediction results, the disturbance adaptation submodule outputs corresponding environmental disturbance compensation data. This data includes three control dimensions: pre-adjustment amplitude, adjustment rate, and duration. These dimensions define the advance response strategies of the temperature control component, humidity control component, and carbon dioxide concentration control component before the disturbance arrives, respectively. The pre-adjustment amplitude sets the preliminary control target for the control components, the adjustment rate controls the execution rhythm of the adjustment process, and the duration defines the effective window for the control action, thereby achieving disturbance feedforward suppression.

[0112] Ultimately, the control system parses the environmental disturbance compensation data into control commands, driving the adjustment components into a pre-response state. This allows for parameter correction before the actual arrival of the environmental disturbance, forming a control chain of "disturbance identification—trend prediction—feedback adjustment." This control chain achieves dynamic perception and proactive response to environmental disturbances induced by plant metabolism, significantly improving the stability and predictability of environmental regulation within the plant enclosure, ensuring that the plant remains in a highly consistent and suitable target environment throughout its entire growth cycle.

[0113] S160, when the second real-time environmental parameter inside the plant box exceeds the set threshold, the control system drives the exhaust mechanism to discharge the air from the plant box, and at the same time controls the premix box to output a new round of stable environmental gas, and adaptively adjusts the control strategy of the premix box based on the environmental disturbance compensation data.

[0114] In one possible implementation, when the second real-time environmental parameter inside the plant box exceeds a set threshold, the control system drives the exhaust mechanism to expel air from the plant box, simultaneously controlling the premixing chamber to output a new round of stable environmental gas. Based on environmental disturbance compensation data, the control strategy of the premixing chamber is adaptively adjusted. Specifically, the environmental parameter judgment module receives the second real-time environmental parameters collected by the plant box temperature sensor, plant box humidity sensor, and plant box carbon dioxide concentration sensor. The second real-time environmental parameters include a second real-time temperature value, a second real-time humidity value, and a second real-time carbon dioxide concentration value. The module then compares the real-time temperature value, real-time humidity value, and real-time carbon dioxide concentration value with target temperature parameters and target humidity parameters. The system compares the target carbon dioxide concentration parameter with an error threshold. When any second real-time environmental parameter exceeds the corresponding error threshold range, the control system outputs an exhaust command, driving the exhaust fan to expel non-standard environmental gases from the plant box through the exhaust duct. Simultaneously, the control system outputs a gas replenishment command, driving the premix box to generate a new round of stable environmental gases based on the current target temperature, target humidity, and target carbon dioxide concentration parameters, and delivers them to the plant box through the air duct. The control system further inputs environmental disturbance compensation data into the strategy adjustment module to adaptively optimize parameters such as the current premix box's adjustment cycle, adjustment amplitude, sensor sampling density, and control signal gain, forming a control strategy with enhanced dynamic response.

[0115] Specifically, firstly, the environmental parameter judgment module receives second real-time environmental parameters collected by the plant box temperature sensor, plant box humidity sensor, and plant box carbon dioxide concentration sensor. These second real-time environmental parameters include the second real-time temperature value, second real-time humidity value, and second real-time carbon dioxide concentration value. The environmental parameter judgment module performs data integrity verification on these second real-time environmental parameters and compares them one-to-one with the target temperature parameter, target humidity parameter, and target carbon dioxide concentration parameter output by the machine learning model. It uses absolute error threshold judgment logic to identify whether any environmental parameter deviates beyond the tolerance range.

[0116] When the deviation of any second real-time environmental parameter exceeds the corresponding error threshold, the system determines that the current environmental state inside the plant box is unstable. The control system immediately sends an exhaust command to the exhaust control unit, driving the exhaust fan to start. The exhaust fan is connected to the exhaust duct via a preset path, quickly expelling non-standard environmental gases from inside the plant box to the outside. A one-way valve is installed in the exhaust path to prevent backflow of external gases, ensuring the unidirectional and efficient exhaust operation and preventing the accumulation of disturbed gases within the plant box.

[0117] Simultaneously, the control system sends a gas replenishment command to the premixing chamber control module, which includes the current target temperature, target humidity, and target carbon dioxide concentration parameters. The temperature, humidity, and carbon dioxide concentration adjustment components within the premixing chamber quickly complete a new round of environmental adjustment tasks based on the target environmental parameters, generating a stable environmental gas with the characteristics of the set parameters. Subsequently, this stable environmental gas is delivered into the plant container at a constant flow rate through the air duct gas delivery mechanism, completing the environmental reconstruction process within the container.

[0118] To prevent repetitive disturbances and improve the response sensitivity and adaptability of the environmental control system, the control system further inputs the environmental disturbance compensation data previously generated by the disturbance adaptation submodule into the strategy adjustment module. The strategy adjustment module dynamically optimizes and adjusts the key control parameters in the current control strategy. These parameters include the adjustment period, adjustment amplitude, sensor sampling density, and control signal gain. The adjustment period affects the adjustment response frequency, the adjustment amplitude controls the amplitude boundary of the adjustment behavior, the sensor sampling density determines the environmental perception accuracy, and the control signal gain affects the adjustment rate and control stability. The strategy adjustment module performs differential-level adaptive reconfiguration of these parameters based on the current disturbance type and predicted trends to form an enhanced adaptive control strategy.

[0119] Finally, through the above-mentioned process of judging environmental parameter errors, emitting non-standard environmental gases, reconstructing standard environmental gases, and adaptively adjusting control strategies, a closed-loop disturbance response mechanism with environmental parameters as the core control objective is constructed. This significantly improves the environmental stability of the plant box and the overall disturbance bearing capacity of the system, ensuring that plants can maintain a stable and consistent growth environment under fluctuating disturbance conditions.

[0120] This embodiment also discloses a machine learning-based intelligent premixed control device, referring to... Figure 2 The device includes an acquisition module 201, a processing module 202, and an output module 203. It is used to execute any of the machine learning-based intelligent premixed control methods described above, wherein:

[0121] The acquisition module 201 is used to build a machine learning model based on historical environmental control data, plant growth data, and environmental disturbance response data.

[0122] The processing module 202 is used to embed the machine learning model into the control system, and the control system controls the adjustment component to make preliminary adjustments to the environment inside the premixing box according to the target environmental parameters.

[0123] The processing module 202 is used to acquire first real-time environmental data through sensors deployed in the premixing chamber and feed it back to the control system to form closed-loop regulation.

[0124] The processing module 202 is used to open the air duct when the environmental parameters in the premix box are stable within the target range, and deliver the stable environment corresponding to the environmental parameters to the plant box at a constant flow rate. At the same time, it performs real-time compensation and correction for changes in environmental parameters in the delivery path based on a machine learning model.

[0125] The processing module 202 is used to input the disturbance data monitored in real time by the sensors inside the plant box into the disturbance adaptation submodule in the machine learning model to predict the disturbance trend and output environmental disturbance compensation data.

[0126] The output module 203 is used to control the ventilation mechanism to exhaust the air from the plant box when the second real-time environmental parameter inside the plant box exceeds the set threshold, and at the same time control the premix box to output a new round of stable environmental gas, and adaptively adjust the control strategy of the premix box based on the environmental disturbance compensation data.

[0127] In one possible implementation, the acquisition module 201 is used to collect historical environmental control data of plants over multiple growth cycles. The historical environmental control data includes environmental parameter setpoints, actual environmental parameter values, environmental parameter adjustment ranges, environmental parameter adjustment response times, environmental parameter adjustment frequencies, and environmental parameter stability indicators.

[0128] The acquisition module 201 is used to collect plant growth data within the corresponding time period of historical environmental control data. The plant growth data includes plant species, growth stage, plant height, leaf area index, chlorophyll content, dry matter and growth rate.

[0129] The acquisition module 201 is used to collect environmental disturbance response data recorded during plant growth. The environmental disturbance response data includes disturbance type, disturbance amplitude, disturbance duration, disturbance recovery time, and disturbance regulation feedback characteristics.

[0130] Processing module 202 is used to time-align historical environmental control data, plant growth data and environmental disturbance response data to construct a unified training sample dataset.

[0131] Output module 203 is used to train a deep neural network on the training sample dataset to form a machine learning model for outputting target environmental parameters and environmental disturbance compensation data. It constructs a multi-layer feedforward neural network containing an input layer, multiple hidden layers, and an output layer. The input layer receives plant species, growth stages, historical sequences of environmental parameters, and phenotypic features. The output layer generates target environmental parameters and disturbance compensation strategies. The training process of the machine learning model adopts a supervised learning mechanism. The objective function is set to minimize the mean square error between the predicted environmental parameters and the actual adjustment results. A regularization term is introduced to constrain the complexity of the controller parameters and prevent overfitting. Cross-validation is used to evaluate the model's generalization ability during training, and hyperparameter grid search is used to optimize the model structure and control parameters.

[0132] In one possible implementation, the processing module 202 is used to deploy the machine learning model in the core control unit of the control system. The core control unit includes a data receiving module, a model parsing module, a control signal generation module, and an execution feedback module.

[0133] The processing module 202 is used to determine the target environmental parameters that the data receiving module receives from the machine learning model output. The model parsing module generates corresponding adjustment instructions based on the target environmental parameters. The adjustment instructions include temperature adjustment instructions, humidity adjustment instructions, and carbon dioxide concentration adjustment instructions.

[0134] The output module 203 is used to control the control signal generation module to output control signals to the temperature control component, humidity control component and carbon dioxide concentration control component according to the adjustment command, so as to drive the control components to perform adjustment operations.

[0135] The processing module 202 is used to control the execution feedback module to collect the current environmental parameters based on the environmental parameter sensors deployed inside the premixing box. The current environmental parameters include the current temperature value, the current humidity value, and the current carbon dioxide concentration value.

[0136] The processing module 202 is used to calculate the error between the current environmental parameters and the target environmental parameters, and to correct the control signal in real time based on the error, forming a closed-loop regulation control logic to stabilize the environmental parameters in the premixing box to the target environmental parameters.

[0137] In one possible implementation, the processing module 202 is configured to deploy environmental parameter sensor modules at multiple spatial locations within the premixing chamber. The environmental parameter sensor modules include a temperature sensor, a humidity sensor, and a carbon dioxide concentration sensor.

[0138] The acquisition module 201 is used to control the environmental parameter sensor module to collect first real-time environmental parameters at a preset sampling period. The first real-time environmental parameters include first real-time temperature value, first real-time humidity value and first real-time carbon dioxide concentration value.

[0139] The output module 203 is used to package the first real-time environmental parameters into an environmental parameter data frame and send it to the control system through the data transmission module.

[0140] The processing module 202 is used to control the control system to calculate the error between the environmental parameter data and the target environmental parameters, obtain the temperature error, humidity error and carbon dioxide concentration error, and generate control signals based on the proportional-integral-derivative control algorithm to drive the temperature regulation component, humidity regulation component and carbon dioxide concentration regulation component to perform regulation operations, forming a closed-loop regulation path with the environmental parameter sensor module as the sensing end, the control system as the decision end and the regulation component as the execution end.

[0141] In one possible implementation, the processing module 202 is used to output a duct opening command when the control system determines that the first real-time temperature value, the first real-time humidity value, and the first real-time carbon dioxide concentration value are all within the error range of the target temperature parameter, the target humidity parameter, and the target carbon dioxide concentration parameter, thereby driving the duct gas delivery mechanism to start. The duct gas delivery mechanism includes a blower, an airflow control valve, and a pressure stabilization module.

[0142] The processing module 202 is used to install environmental parameter monitoring modules along the inside of the air duct. The environmental parameter monitoring modules include a temperature sensor, a humidity sensor, and a carbon dioxide concentration sensor along the transport path, which are used to acquire environmental change data within the transport path.

[0143] The processing module 202 is used to input environmental change data into the environmental transmission prediction sub-model in the machine learning model. The environmental transmission prediction sub-model generates environmental compensation instructions. The control system dynamically adjusts the output amplitude of the temperature regulation component, humidity regulation component and carbon dioxide concentration regulation component according to the environmental compensation instructions, so as to realize real-time compensation and correction of the environmental parameters at the end of the transmission path.

[0144] In one possible implementation, the processing module 202 is used to deploy environmental disturbance monitoring sensor modules at multiple locations inside the plant box. The environmental disturbance monitoring sensor modules include a plant box temperature sensor, a plant box humidity sensor, and a plant box carbon dioxide concentration sensor, which are used to collect real-time temperature disturbance values, real-time humidity disturbance values, and real-time carbon dioxide concentration disturbance values ​​inside the plant box to form a disturbance data sequence.

[0145] The processing module 202 is used to package the perturbation data sequence into a perturbation data input frame and transmit it to the perturbation adaptation submodule of the machine learning model. The perturbation adaptation submodule performs perturbation trend prediction based on the neural network structure of time series modeling and outputs the perturbation trend prediction result.

[0146] The processing module 202 is used to generate environmental disturbance compensation data based on the disturbance trend prediction results. The environmental disturbance compensation data includes the pre-adjustment amplitude, adjustment rate and duration for controlling the temperature adjustment component, humidity adjustment component and carbon dioxide concentration adjustment component, so as to achieve feedforward compensation correction of the disturbance.

[0147] In one possible implementation, the processing module 202 is used to control the environmental parameter judgment module to receive second real-time environmental parameters collected by the plant box temperature sensor, the plant box humidity sensor, and the plant box carbon dioxide concentration sensor. The second real-time environmental parameters include a second real-time temperature value, a second real-time humidity value, and a second real-time carbon dioxide concentration value. The module then compares the real-time temperature value, the real-time humidity value, and the real-time carbon dioxide concentration value with error thresholds for the target temperature parameter, the target humidity parameter, and the target carbon dioxide concentration parameter.

[0148] The processing module 202 is used to control the system to output an exhaust command when any second real-time environmental parameter exceeds the corresponding error threshold range, and drive the exhaust fan to discharge non-standard environmental gases from the plant box through the exhaust channel.

[0149] The processing module 202 is used to control the output of the gas replenishment command of the control system, drive the premix box to generate a new round of stable environmental gas based on the current target temperature parameter, target humidity parameter and target carbon dioxide concentration parameter, and deliver it to the plant box through the air duct.

[0150] The processing module 202 is used to control the control system to further input environmental disturbance compensation data into the strategy adjustment module, and to adaptively optimize parameters such as the adjustment cycle, adjustment amplitude, sensor sampling density and control signal gain of the current premix box to form a control strategy with enhanced dynamic response.

[0151] It should be noted that the above embodiments of the apparatus are only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0152] This embodiment also discloses an electronic device, referring to... Figure 3The electronic device may include: at least one processor 301, at least one communication bus 302, user interface 303, network interface 304, and at least one memory 305.

[0153] The communication bus 302 is used to enable communication between these components.

[0154] The user interface 303 may include a display screen and a camera. Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0155] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0156] The processor 301 may include one or more processing cores. The processor 301 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory 305, and by calling data stored in memory 305. Optionally, the processor 301 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 301 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications. The GPU is responsible for rendering and drawing the content required for display. The modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 301 and may be implemented as a separate chip.

[0157] The memory 305 may include random access memory (RAM) or read-only memory. Optionally, the memory may include a non-transitory computer-readable storage medium. The memory 305 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the various method embodiments described above, etc. The data storage area may store data involved in the various method embodiments described above. Optionally, the memory 305 may also be at least one storage device located remotely from the aforementioned processor 301. As a computer storage medium, the memory 305 may include an operating system, a network communication module, a user interface 303 module, and an application program for a machine learning-based intelligent premixed control method.

[0158] exist Figure 3 In the illustrated electronic device, the user interface 303 is primarily used to provide an input interface for the user and to acquire user input data. The processor 301 can be used to call an application program stored in the memory 305 that represents a machine learning-based intelligent premixed control method. When executed by one or more processors 301, this causes the electronic device to perform one or more methods as described in the above embodiments.

[0159] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0160] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0161] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the shown or discussed mutual couplings or direct couplings or communication connections may be through some service interfaces; indirect couplings or communication connections between apparatuses or units may be electrical or other forms.

[0162] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0163] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0164] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory 305 and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory 305 includes various media capable of storing program code, such as a USB flash drive, external hard drive, magnetic disk, or optical disk.

[0165] This application also discloses a computer-readable storage medium storing instructions. When executed by one or more processors 301, these instructions cause an electronic device to perform one or more methods as described in the above embodiments.

[0166] The above are merely exemplary embodiments of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Those skilled in the art will readily conceive of other embodiments of this disclosure upon considering the specification and the disclosure of practical truth. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure. The specification and embodiments are considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. A method of intelligent premix control based on machine learning, characterized in that, The method includes: A machine learning model was constructed based on historical environmental control data, plant growth data, and environmental disturbance response data. The machine learning model is embedded in the control system, and the control system controls the adjustment components to make preliminary adjustments to the environment inside the premixing box according to the target environmental parameters. The first real-time environmental data is acquired by sensors deployed in the premixing tank and fed back to the control system to form a closed-loop regulation; When the environmental parameters in the premix box are stable within the target range, the air duct is opened to deliver the stable environment corresponding to the environmental parameters to the plant box at a constant flow rate. At the same time, the changes in environmental parameters in the delivery path are compensated and corrected in real time based on the machine learning model. The disturbance data monitored in real time by the sensors inside the plant box is input into the disturbance adaptation submodule in the machine learning model to predict the disturbance trend and output environmental disturbance compensation data. When the second real-time environmental parameter inside the plant box exceeds the set threshold, the control system drives the exhaust mechanism to discharge the air from the plant box, and at the same time controls the premix box to output a new round of stable environmental gas, and adaptively adjusts the control strategy of the premix box based on the environmental disturbance compensation data. The process of acquiring first real-time environmental data through sensors deployed within the premixing chamber and feeding it back to the control system to form a closed-loop regulation specifically includes: An environmental parameter sensor module is deployed in multiple spatial locations inside the premixing chamber. The environmental parameter sensor module includes a temperature sensor, a humidity sensor, and a carbon dioxide concentration sensor. The environmental parameter sensor module collects first real-time environmental parameters at a preset sampling period. The first real-time environmental parameters include a first real-time temperature value, a first real-time humidity value, and a first real-time carbon dioxide concentration value. The first real-time environmental parameters are packaged into an environmental parameter data frame and sent to the control system through the data transmission module; The control system calculates the error between the environmental parameter data and the target environmental parameters to obtain the temperature error, humidity error and carbon dioxide concentration error, and generates control signals based on the proportional-integral-derivative control algorithm to drive the temperature regulation component, humidity regulation component and carbon dioxide concentration regulation component to perform regulation operations, forming a closed-loop regulation path with the environmental parameter sensor module as the sensing end, the control system as the decision end and the regulation components as the execution end. When the environmental parameters within the premixing chamber stabilize within the target range, the air duct is opened to deliver the stable environment corresponding to the environmental parameters to the plant container at a constant flow rate. Simultaneously, based on the machine learning model, real-time compensation and correction are performed on changes in environmental parameters along the delivery path, specifically including: When the control system determines that the first real-time temperature value, the first real-time humidity value, and the first real-time carbon dioxide concentration value are all within the error range of the target temperature parameter, the target humidity parameter, and the target carbon dioxide concentration parameter, it outputs a duct opening command to drive the duct gas delivery mechanism to start. The duct gas delivery mechanism includes a blower, an airflow control valve, and a pressure stabilization module. An environmental parameter monitoring module for the transport path is installed along the inside of the air duct. The environmental parameter monitoring module for the transport path includes a temperature sensor, a humidity sensor, and a carbon dioxide concentration sensor for the transport path, which are used to acquire environmental change data within the transport path. The environmental change data is input into the environmental transmission prediction sub-model in the machine learning model. The environmental transmission prediction sub-model generates an environmental compensation instruction. The control system dynamically adjusts the output amplitude of the temperature regulation component, humidity regulation component, and carbon dioxide concentration regulation component according to the environmental compensation instruction, so as to realize real-time compensation and correction of the environmental parameters at the terminal of the transport path.

2. The intelligent premix control method based on machine learning according to claim 1, wherein, The machine learning model constructed based on historical environmental control data, plant growth data, and environmental disturbance response data specifically includes: Historical environmental control data of plants were collected over multiple growth cycles. The historical environmental control data included environmental parameter setpoints, actual environmental parameter values, environmental parameter adjustment ranges, environmental parameter adjustment response times, environmental parameter adjustment frequencies, and environmental parameter stability indicators. Collect plant growth data within the time period corresponding to the historical environmental control data. The plant growth data includes plant species, growth stage, plant height, leaf area index, chlorophyll content, dry matter and growth rate. Collect environmental disturbance response data recorded during plant growth, including disturbance type, disturbance amplitude, disturbance duration, disturbance recovery time, and disturbance regulation feedback characteristics; The historical environmental control data, plant growth data and environmental disturbance response data are time-aligned to construct a unified training sample dataset. A deep neural network is used to train the training sample dataset to form a machine learning model for outputting target environmental parameters and environmental disturbance compensation data. This model comprises a multi-layer feedforward neural network with an input layer, multiple hidden layers, and an output layer. The input layer receives plant species, growth stages, historical sequences of environmental parameters, and phenotypic features. The output layer generates the target environmental parameters and disturbance compensation strategies. The training process of the machine learning model employs a supervised learning mechanism. The objective function is set to minimize the mean square error between the predicted environmental parameters and the actual adjustment results. A regularization term is introduced to constrain the complexity of the controller parameters and prevent overfitting. Cross-validation is used to evaluate the model's generalization ability during training, and hyperparameter grid search is used to optimize the model structure and control parameters. 3.The intelligent premixing control method based on machine learning of claim 1, wherein, The step of embedding the machine learning model into the control system, wherein the control system controls the adjustment components to initially adjust the environment inside the premixing chamber according to the target environmental parameters, specifically includes: The machine learning model is deployed in the core control unit of the control system, which includes a data receiving module, a model parsing module, a control signal generation module, and an execution feedback module. The data receiving module is used to receive the target environmental parameters output by the machine learning model, and the model parsing module generates corresponding adjustment instructions based on the target environmental parameters. The adjustment instructions include temperature adjustment instructions, humidity adjustment instructions and carbon dioxide concentration adjustment instructions. The control signal generation module outputs control signals to the temperature regulation component, humidity regulation component, and carbon dioxide concentration regulation component according to the regulation command, so as to drive the regulation components to perform regulation operations; The execution feedback module collects current environmental parameters based on environmental parameter sensors deployed inside the premixing box. The current environmental parameters include the current temperature value, the current humidity value, and the current carbon dioxide concentration value. The error between the current environmental parameters and the target environmental parameters is calculated, and the control signal is corrected in real time based on the error to form a closed-loop regulation control logic, so that the environmental parameters in the premixing box are stabilized to the target environmental parameters.

4. The intelligent premix control method based on machine learning according to claim 1, wherein, The step of inputting the disturbance data monitored in real time by sensors inside the plant box into the disturbance adaptation submodule of the machine learning model for disturbance trend prediction and outputting environmental disturbance compensation data specifically includes: An environmental disturbance monitoring sensor module is deployed at multiple locations inside the plant box. The environmental disturbance monitoring sensor module includes a plant box temperature sensor, a plant box humidity sensor, and a plant box carbon dioxide concentration sensor, which are used to collect real-time temperature disturbance values, real-time humidity disturbance values, and real-time carbon dioxide concentration disturbance values ​​inside the plant box to form a disturbance data sequence. The perturbation data sequence is packaged into a perturbation data input frame and transmitted to the perturbation adaptation submodule of the machine learning model. The perturbation adaptation submodule performs perturbation trend prediction based on the neural network structure of time series modeling and outputs the perturbation trend prediction result. Environmental disturbance compensation data is generated based on the disturbance trend prediction results. The environmental disturbance compensation data includes the pre-adjustment amplitude, adjustment rate and duration for controlling the temperature regulation component, humidity regulation component and carbon dioxide concentration regulation component, so as to achieve feedforward compensation correction of the disturbance.

5. The intelligent premix control method based on machine learning according to claim 1, wherein, When the second real-time environmental parameter inside the plant box exceeds a set threshold, the control system drives the exhaust mechanism to discharge the air from the plant box, simultaneously controlling the premixing chamber to output a new round of stable environmental gas, and adaptively adjusting the control strategy of the premixing chamber based on the environmental disturbance compensation data, specifically including: The environmental parameter judgment module receives second real-time environmental parameters collected by the plant box temperature sensor, the plant box humidity sensor, and the plant box carbon dioxide concentration sensor. The second real-time environmental parameters include a second real-time temperature value, a second real-time humidity value, and a second real-time carbon dioxide concentration value. The module then compares the real-time temperature value, the real-time humidity value, and the real-time carbon dioxide concentration value with the error thresholds of the target temperature parameter, the target humidity parameter, and the target carbon dioxide concentration parameter. When any of the second real-time environmental parameters exceeds the corresponding error threshold range, the control system outputs an exhaust command to drive the exhaust fan to discharge non-standard environmental gases from the plant box through the exhaust channel. At the same time, the control system outputs a gas replenishment command, driving the premix box to generate a new round of stable environmental gas based on the current target temperature parameters, target humidity parameters, and target carbon dioxide concentration parameters, and delivers it to the plant box through the air duct; The control system further inputs environmental disturbance compensation data into the strategy adjustment module to adaptively optimize the current premix box's adjustment cycle, adjustment amplitude, sensor sampling density, and control signal gain, forming a control strategy with enhanced dynamic response.

6. A machine learning-based intelligent premixed control device, characterized in that, The device is used to execute a machine learning-based intelligent premixed control method as described in any one of claims 1-5, the device comprising an acquisition module (201), a processing module (202), and an output module (203), wherein: The acquisition module (201) is used to construct a machine learning model based on historical environmental control data, plant growth data and environmental disturbance response data; The processing module (202) is used to embed the machine learning model into the control system, and the control system controls the adjustment component to make preliminary adjustments to the environment inside the premix box according to the target environmental parameters. The processing module (202) is used to acquire first real-time environmental data through sensors deployed in the premix box and feed it back to the control system to form a closed-loop regulation; The processing module (202) is used to open the air duct when the environmental parameters in the premix box are stable within the target range, and to deliver the stable environment corresponding to the environmental parameters to the plant box at a constant flow rate. At the same time, it performs real-time compensation and correction on the changes in environmental parameters in the delivery path based on the machine learning model. The processing module (202) is used to input the disturbance data monitored in real time by the sensors inside the plant box into the disturbance adaptation submodule in the machine learning model to predict the disturbance trend and output environmental disturbance compensation data. The output module (203) is used to drive the exhaust mechanism to discharge the air from the plant box when the second real-time environmental parameter inside the plant box exceeds the set threshold, and at the same time control the premix box to output a new round of stable environmental gas, and adaptively adjust the control strategy of the premix box based on the environmental disturbance compensation data.

7. An electronic device, characterized in that, The device includes a processor (301), a communication bus (302), a user interface (303), a network interface (304), and a memory (305). The memory (305) is used to store instructions. The user interface (303) and the network interface (304) are both used to communicate with other devices. The communication bus (302) is used to realize the connection and communication between the components within the electronic device. The processor (301) is used to execute the instructions stored in the memory (305) so that the electronic device performs the method described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed, perform the method as described in any one of claims 1-5.

Images