Smart mushroom planting environment regulation and control system and method based on multi-parameter fusion control

Abstract

The embodiment of the invention provides an intelligent mushroom planting environment regulation and control system and method based on multi-parameter fusion control. The system comprises a sensor assembly, an MPC decision module, a control module and execution equipment. The sensor assembly collects environmental parameters such as temperature, humidity, carbon dioxide concentration and illumination intensity in the mushroom house in real time. And the control module transmits the environmental parameters to the MPC decision module. And the MPC decision module predicts a future parameter change trend based on the mushroom growth dynamic model, and generates a cooperative control instruction through rolling optimization solution. The control module drives ventilation, lighting, air conditioning, spraying and other execution devices to cooperatively act according to the instruction, and precise regulation and control of the mushroom house environment are achieved. The multi-variable strong coupling system is globally optimized through the model predictive control algorithm, the problems of control loop conflict, system oscillation and poor adaptability in a traditional method are solved, and high-precision, high-stability and energy-saving control over the environment is achieved.

Description

Technical Field

[0001] This invention relates to the field of agricultural intelligent control technology, and in particular to a smart mushroom cultivation environment regulation system and method based on multi-parameter fusion control. Background Technology

[0002] The multi-parameter fusion control system for mushroom cultivation environment regulation is a distributed control system based on the Internet of Things, sensor technology, and intelligent algorithms. It monitors environmental parameters within the mushroom house in real time, compares them with a preset ideal growth model, and automatically drives various environmental devices to achieve precise, dynamic, and stable control of key factors such as temperature, humidity, light, and air. Its core objective is to transform the traditional experience-based, manual, and extensive control methods into a data-driven, automated, and precise industrial management model.

[0003] There are two main types of existing related technical solutions. One type is the edible fungus growth environment control system based on fuzzy control, which uses a programmable logic controller (PLC) or embedded industrial computer as the lower-level control core, responsible for highly reliable data acquisition and equipment driving. Its core innovation lies in the introduction of a fuzzy logic controller. This controller converts the collected precise quantities such as temperature, humidity, and carbon dioxide concentration into fuzzy linguistic variables such as "temperature is too high," "humidity is moderate," and "carbon dioxide concentration is very high" through a fuzzification process. Subsequently, a pre-set fuzzy rule base based on expert experience performs inference. For example, if the temperature is too high and the humidity is too low, the cooling is moderately increased and the ventilation is slightly reduced. Finally, the inference result is converted into precise control quantities through defuzzification and output to the various actuators. However, the fuzzy control-based edible fungus growth environment control system suffers from the problem of decreased control effect when multiple parameters are coupled. Parameters such as temperature, humidity, carbon dioxide concentration, and light in the edible fungus growth environment are strongly coupled. For example, increasing the temperature may lead to a decrease in humidity, and increasing ventilation to reduce carbon dioxide may also lead to fluctuations in temperature and humidity. If traditional fuzzy controllers are designed independently for each parameter, it is difficult to effectively handle this strong coupling relationship, which can easily lead to system oscillation, mutual interference between parameters, and difficulty in achieving simultaneous stability. Meanwhile, fuzzy control is essentially a nonlinear control, which does not have an integral element. Therefore, for certain specific inputs, the system may not be able to completely eliminate steady-state errors. That is, the controlled variable, such as temperature and humidity, may stabilize at a value with a slight deviation from the set value.

[0004] Another type is a distributed monitoring system based on IoT and PID control. This system collects environmental parameters from multiple points within the mushroom house via wireless sensor networks (such as ZigBee and LoRa), aggregates them at an IoT gateway, and finally uploads them to a cloud platform. The cloud platform or local controller independently applies a PID (Proportional-Integral-Derivative) control algorithm for each environmental parameter. For example, the temperature control loop independently controls the air conditioner based on the deviation between the set temperature and the measured temperature using PID calculations; the humidity control loop independently controls the humidifier. There is virtually no information exchange between the loops. However, the distributed architecture of the distributed monitoring system based on IoT and PID control may lead to coordination failures. Although IoT enables centralized data monitoring, if the control layer remains distributed—that is, temperature PID, humidity PID, CO2 PID, etc., operate independently—the aforementioned parameter coupling problem will reappear. Each controller will fight to achieve its own goal; for example, the humidifier is working hard while the fan is simultaneously removing moisture, resulting in energy waste and system inefficiency. Furthermore, PID parameter tuning is difficult and lacks adaptability. The performance of the PID controller is highly dependent on the tuning of the proportional, integral, and derivative parameters. The growth of edible fungi is a complex process characterized by high inertia, nonlinearity, and time-varying characteristics. A fixed set of PID parameters may only perform well during a specific growth stage or under specific external conditions. When the environment changes abruptly, fixed PID parameters can lead to slow system response, excessive overshoot, or even instability. Summary of the Invention

[0005] In view of this, it is necessary to provide a smart mushroom cultivation environment control system and method based on multi-parameter fusion control to solve the above-mentioned problems of the existing technology.

[0006] To address the aforementioned problems, in a first aspect, embodiments of the present invention provide a smart mushroom cultivation environment control system based on multi-parameter fusion control, comprising: A sensor assembly is used to collect environmental parameters inside the mushroom house in real time; wherein, the environmental parameters include temperature, humidity, carbon dioxide concentration and light intensity; The MPC decision module receives environmental parameters transmitted by the control module, predicts the parameter evolution trend based on the mushroom growth dynamic model, generates collaborative control commands through rolling optimization and outputs them to the control module; wherein, the mushroom growth dynamic model represents the dynamic relationship between the environmental parameter state and the control input and output. The control module is connected to the sensor components, the execution device, and the MPC decision module respectively. It is used to transmit the environmental parameters to the MPC decision module and to output the corresponding drive signal to the execution device according to the cooperative control command. An actuator is used to perform control actions on the mushroom house environment in response to the drive signal.

[0007] Preferably, the sensor assembly includes: Temperature sensors are deployed at multiple locations inside and outside the mushroom house to collect temperature data. Humidity sensors are deployed at multiple locations inside and outside the mushroom house to collect humidity data. Carbon dioxide concentration sensors are deployed at multiple locations within the mushroom house to collect carbon dioxide concentration data. A light intensity sensor is deployed in the light-exposed area of ​​the mushroom house to collect light intensity data.

[0008] Preferably, the control module is a PLC controller, and the control module is also used to receive parameters collected by the sensor components and perform data integration and status judgment; the control module has a built-in programmable logic control program to automatically match environmental control strategies according to the growth stage of the mushrooms.

[0009] Preferably, the execution device includes: Ventilation system is used to regulate the carbon dioxide concentration in the mushroom house and help balance the temperature and humidity. LED lighting system is used to adjust the light intensity inside the mushroom house; An air conditioning system is used to regulate the temperature inside the mushroom house; The spray system is used to regulate the humidity inside the mushroom house.

[0010] Preferably, the mushroom growth dynamic model is as follows: x(k+1) = A·x(k) + B·u(k) y(k) = C·x(k) In the formula, x is the state variable, x = [temperature, humidity, carbon dioxide concentration, light intensity]. T ;u is the control input, u = [ventilation volume, lighting power, air conditioning power, sprinkler volume] T ; y is the output variable, y = [temperature, humidity, carbon dioxide concentration, light intensity] T A, B, and C are the state transition matrix, control input matrix, and output matrix, respectively.

[0011] Preferably, the step of predicting the evolution trend of parameters based on the mushroom growth dynamic model, generating collaborative control commands through rolling optimization and outputting them to the control module includes: Obtain the current state x(k) and the setpoint sequence r(k+1), ..., r(k+Np), where Np is the prediction time domain; Based on the aforementioned mushroom growth dynamic model, the system output Y for the next Np time steps is predicted by the following expression: Y=F·x(k)+Φ·U In the formula, Y is the vector of future output, U is the vector of future control input, and F and Φ are matrices derived from matrices A, B, and C. With the goal of minimizing tracking error and control action variation, an objective function J is constructed and the optimization problem is solved, expressed as: minJ=(YR) · ·(YR)+ΔU · ·ΔU In the formula, R is the set value vector. The output error weight matrix is... The control increment weight matrix is ​​ΔU, which is the control increment vector. The objective function is solved under constraints to obtain the optimal control sequence. The constraints include upper and lower limits of the system state at future time points, upper and lower limits of the control input, and upper and lower limits of the control increment. The first control variable in the optimal control sequence is output as the cooperative control command.

[0012] Preferably, the MPC decision module is further used for: After each control cycle, the measured values ​​of environmental parameters collected by the sensor components are compared with the prediction results of the mushroom growth dynamic model to obtain the prediction error; based on the prediction error, the initial state of the prediction for the next control cycle is corrected, or the parameters of the mushroom growth dynamic model are adjusted.

[0013] Secondly, embodiments of the present invention provide a method for regulating the intelligent mushroom cultivation environment based on multi-parameter fusion control, comprising: The sensor assembly collects environmental parameters inside the mushroom house in real time and transmits them to the control module. The environmental parameters include temperature, humidity, carbon dioxide concentration, and light intensity. The control module transmits the environmental parameters to the MPC decision module; The MPC decision module predicts the evolution trend of environmental parameters based on the mushroom growth dynamic model and generates collaborative control commands through rolling optimization. The control module generates corresponding drive signals based on the coordinated control instructions and outputs them to the execution device; The device responds to the drive signal and performs control actions on the mushroom house environment.

[0014] Preferably, after the sensor components collect environmental parameters inside the mushroom house in real time and transmit them to the control module, the method further includes: The control module receives the parameters collected by the sensor components and performs data integration and status judgment. The control module has a built-in programmable logic control program that automatically matches environmental control strategies according to the growth stage of the mushrooms.

[0015] Preferably, after the device performs the control action, the method further includes: After each control cycle, the measured values ​​of environmental parameters collected by the sensor components are compared with the prediction results of the mushroom growth dynamic model to obtain the prediction error; based on the prediction error, the initial state of the prediction for the next control cycle is corrected, or the parameters of the mushroom growth dynamic model are adjusted.

[0016] The intelligent mushroom cultivation environment control system and method based on multi-parameter fusion control provided by this invention has the following advantages compared with the prior art: 1) This invention introduces Model Predictive Control (MPC) algorithm as the core of multi-parameter fusion decision-making, achieving coordinated optimization control of multiple strongly coupled environmental variables such as temperature, humidity, carbon dioxide concentration, and light intensity within the mushroom house. This method utilizes a dynamic model of mushroom growth to predict the evolution trend of each parameter over a future period and calculates the optimal coordinated action sequence of all actuators in one go through rolling optimization. This avoids the problems of conflicting control loops and system oscillations caused by traditional fuzzy control or independent PID control, thus improving the overall control stability of the multivariable coupled system.

[0017] 2) The rolling optimization mechanism of this invention can perform forward-looking solutions within each control cycle with the comprehensive objective of minimizing the overall deviation between the environmental setpoint and the predicted value, as well as the system energy consumption. This model prediction-based optimization process enables the system to anticipate the slow dynamic characteristics of environmental parameters and plan smooth and precise control actions in advance, achieving fast and accurate dynamic tracking and adjustment.

[0018] 3) This invention, through its built-in feedback correction mechanism, can compare the actual measured values ​​with the model prediction values ​​and correct errors after each control cycle. This mechanism enables the system to continuously adapt to the dynamic changes at different stages of mushroom growth and external environmental disturbances, automatically adjusting the prediction model or initial state values, thereby ensuring the long-term adaptability and robustness of the control algorithm throughout the entire production cycle. Attached Figure Description

[0019] Figure 1 The structural block diagram of the intelligent mushroom cultivation environment control system based on multi-parameter fusion control provided by the present invention; Figure 2 The MPC-PLC collaborative control flowchart provided by this invention; Figure 3 A flowchart illustrating the predictive control process of the mushroom growth dynamic model provided by this invention; Figure 4The flowchart shows the intelligent mushroom cultivation environment regulation method based on multi-parameter fusion control provided by the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] The naming or numbering of steps in the embodiments of the present invention does not mean that the steps in the method flow must be executed in the time / logical order indicated by the naming or numbering. The execution order of the named or numbered process steps can be changed according to the technical purpose to be achieved, as long as the same or similar technical effect can be achieved.

[0022] This invention aims to construct a smart planting closed loop capable of self-perception, self-decision-making, and self-optimization through a technical approach that integrates multi-source data perception, model-driven prediction and decision-making, and collaborative global optimization. This will transform edible mushroom cultivation from a traditional experience-based model into a modern, precision industry driven by data. Its core objective is to systematically address the uncertainties, inefficiencies, and low quality issues in edible mushroom cultivation, promoting the process towards precision, automation, optimization, and replicability, ultimately achieving the goals of high-yield, high-quality, high-efficiency, safe, and labor-saving modern agricultural production.

[0023] Figure 1 This is a structural block diagram of a smart mushroom cultivation environment control system based on multi-parameter fusion control provided in an embodiment of the present invention; see reference. Figure 1 The intelligent mushroom cultivation environment control system based on multi-parameter fusion control provided by this invention includes: A sensor assembly is used to collect environmental parameters inside the mushroom house in real time; wherein, the environmental parameters include temperature, humidity, carbon dioxide concentration and light intensity; The MPC decision module receives environmental parameters transmitted by the control module, predicts the parameter evolution trend based on the mushroom growth dynamic model, generates collaborative control commands through rolling optimization and outputs them to the control module; wherein, the mushroom growth dynamic model represents the dynamic relationship between the environmental parameter state and the control input and output. The control module is connected to the sensor components, the execution device, and the MPC decision module respectively. It is used to transmit the environmental parameters to the MPC decision module and to output the corresponding drive signal to the execution device according to the cooperative control command. An actuator is used to perform control actions on the mushroom house environment in response to the drive signal.

[0024] Traditional methods of controlling the environment in mushroom cultivation often rely on single-parameter control or simple linkage, which struggles to handle the strong coupling relationships between multiple parameters, leading to problems such as control oscillations and energy waste. To address this issue, this invention introduces the concept of multivariable model predictive control into the field of agricultural facility environmental control. By establishing an intelligent control system capable of "foreseeing" the future and making collaborative decisions, it fundamentally achieves decoupling and global optimization of multiple parameters.

[0025] Reference Figure 1 The system consists of sensor components, a control module, an MPC (Model Predictive Control) decision module, actuators, and a power management unit. The sensor components are connected to the control module via a RS-485 communication bus, transmitting the collected environmental parameter data to the control module. The control module, acting as the local control hub, engages in bidirectional data exchange with the MPC decision module via industrial communication protocols and connects to and drives the various actuators via control signal lines. The MPC decision module receives data, runs algorithms, and issues coordinated commands. Actuators include lighting, ventilation systems, sprinkler systems, and air conditioning systems, all controlled by specific drive signals output by the control module. The power management unit provides a stable 24V DC power supply to the entire system.

[0026] Specifically, temperature and humidity sensors deployed at key locations inside and outside the mushroom house, carbon dioxide concentration sensors deployed at multiple points inside the mushroom house, and light intensity sensors deployed in the light-exposed areas together constitute the sensor assembly, responsible for real-time and comprehensive collection of environmental parameters such as temperature, humidity, carbon dioxide concentration, and light intensity within the mushroom house. This raw data is transmitted to the local core control module via a fieldbus.

[0027] The control module performs preliminary integration, filtering, and reliability assessment on the received multi-source data to form a snapshot of the system's current state. Subsequently, it sends this integrated environmental state data to the upper-level MPC decision module via the industrial communication network.

[0028] Once the MPC decision module is activated, its core process begins. This module invokes its built-in mushroom growth dynamic model, which mathematically encodes the interrelationships between environmental parameters and their response characteristics to control inputs. The MPC decision module uses this model to predict future parameter evolution trends based on the received environmental parameters, and then generates coordinated control commands through rolling optimization, which are fed back to the control module. Rolling optimization is the core operating mechanism of the MPC algorithm. In each control cycle, the algorithm re-predicts future environmental changes based on the latest system state and solves an optimization problem aimed at minimizing tracking error and energy consumption. However, only the first step of the calculated optimal control sequence is implemented. This process is repeated in the next cycle, thus achieving dynamic and adaptive optimal control.

[0029] The control module parses the coordinated control commands and converts them into specific, executable drive signals, which are then sent to the execution devices, including the ventilation system, LED lighting system, air conditioning system, and sprinkler system. These devices work together to precisely intervene in the mushroom house environment. After completing a control cycle, the system collects actual environmental data again. The MPC decision module compares this measured value with the predicted value from the previous cycle, uses the deviation for feedback correction, and updates the model or state estimate. This provides a more accurate starting point for the rolling optimization of the next cycle, forming a continuously self-optimizing closed loop.

[0030] This invention introduces Model Predictive Control (MPC) algorithm as the core of multi-parameter fusion decision-making, achieving coordinated optimization control of multiple strongly coupled environmental variables such as temperature, humidity, carbon dioxide concentration, and light intensity within the mushroom house. This method utilizes a dynamic model of mushroom growth to predict the evolution trend of each parameter over a future period and calculates the optimal coordinated action sequence of all actuators in one go through rolling optimization. This avoids the problems of conflicting control loops and system oscillations caused by traditional fuzzy control or independent PID control, thus improving the overall control stability of the multivariable coupled system.

[0031] In some embodiments of this application, the control module is a PLC (Programmable Logic Controller) controller. The control module is also used to receive parameters collected by the sensor components and to integrate the data and determine the status. The control module has a built-in programmable logic control program that automatically matches environmental control strategies according to the growth stage of the mushrooms.

[0032] It is understood that using a PLC controller as the control module is just one example of this application. An embedded dedicated controller or industrial computer based on ARM, DSP or FPGA can also be used as the control module.

[0033] Specifically, the PLC controller directly connects to and receives real-time signals from various sensors in the sensor module through its digital and analog input interfaces. These signals include raw environmental data such as temperature, humidity, carbon dioxide concentration, and light intensity inside and outside the mushroom house. After receiving the data, the program built into the PLC controller first performs preliminary integration and status judgment on this multi-source and heterogeneous data. For example, it removes obvious outliers, calculates multi-point averages to represent the regional status, and determines whether the current environmental parameters are within the preset safety threshold range.

[0034] The PLC controller internally stores and runs pre-written programmable logic control programs. One of the core logics of this program is the ability to identify the current growth stage of the mushrooms, for example, by determining whether it is in the mycelial stage, bud induction stage, or fruiting stage through operator settings, timers, or instructions from the host system. Once the stage is determined, the PLC automatically retrieves and matches the corresponding environmental control basic strategies from memory. These strategies include the ideal environmental parameter target range for that stage and basic equipment linkage rules.

[0035] The programmable logic controller (PLC) program can implement multi-parameter coupled control logic. For example, when the program determines that the carbon dioxide concentration is too high and ventilation is needed, it will simultaneously detect the current humidity level. If the humidity is found to be at the upper limit of the allowable range, the program will not immediately trigger a high-volume ventilation command, because direct ventilation may cause a sudden drop in humidity. Instead, it will first wait for the ongoing spray humidification cycle to end, or try to first activate the dehumidification function of the air conditioner to appropriately reduce the humidity before starting the ventilation equipment for air exchange. This logic effectively avoids drastic fluctuations in other environmental parameters caused by single-parameter adjustment.

[0036] Furthermore, this PLC controller supports switching between manual and automatic operating modes. In automatic mode, it primarily executes optimized instructions from the host MPC decision module. When a network interruption prevents the receipt of host instructions, the system can automatically or manually switch to local control mode. In this mode, the PLC will rely entirely on its built-in programmable logic control program and growth stage strategy to independently drive the execution equipment, maintaining a basic but safe and reliable closed-loop environment control, thereby ensuring production safety.

[0037] At the communication level, the PLC controller interacts with the data acquisition and MPC control unit, which serves as the upper-level monitoring and decision-making unit, through industrial Ethernet or serial communication interfaces and using standard industrial communication protocols such as Modbus TCP or MQTT.

[0038] This embodiment employs a PLC controller as the control module. The PLC controller's built-in intelligent program automatically matches and adjusts control strategies according to the mushroom growth stage and executes advanced control logic that considers multi-parameter coupling relationships, effectively preventing drastic environmental fluctuations. The ability to switch between manual and automatic modes and the local closed-loop control capability in case of network interruption ensures the continuity and safety of the production process.

[0039] Figure 2 This demonstrates the collaborative workflow between the MPC decision module and the PLC controller within a control cycle in the system of this invention. At the start of each control cycle, for example, after 5 minutes, the sensor components collect real-time data of the mushroom house environment, namely the process variable (PV), representing the actual measured values ​​of controlled variables such as temperature, humidity, carbon dioxide concentration, and light intensity. The process variable PV data is then uploaded to the MPC decision module via the PLC.

[0040] The MPC decision module then initiates its core calculations. The first step is prediction, using the built-in mushroom growth dynamic model and the received current process variable PV to predict the environmental state at multiple future time points. The second step is optimization, aiming to minimize tracking error and system cost, solving for the optimal control sequence for a future time period. The first value of this sequence is the optimal setpoint SV (Set Value) for the next cycle. The MPC decision module then sends this optimized SV to the PLC.

[0041] After receiving the SV (Signal Variable), the PLC converts it into specific drive signals to control the actions of equipment such as ventilation, lighting, air conditioning, and sprinklers. The equipment's actions cause changes in the mushroom house environment, which are then detected by sensors, creating a new process variable PV, thus initiating the next control cycle. This cyclical process, combined with the MPC's internal mechanism of comparing the actual process variable PV with the predicted value and performing feedback correction, constitutes a dynamic, continuously self-optimizing closed-loop control system.

[0042] This invention achieves a high degree of coordination between perception, decision-making, and execution through an MPC-PLC collaborative control process. The MPC decision module is responsible for forward-looking multivariate optimization calculations, generating a globally optimal setpoint sequence; the PLC module is responsible for reliably and quickly translating optimization instructions into equipment actions, and ensuring the continuity and safety of the underlying control. This architecture effectively solves the control problem of multi-parameter strongly coupled systems, avoids system oscillations caused by control loop conflicts in traditional methods, and improves the overall stability and accuracy of environmental control.

[0043] In some embodiments of this application, the mushroom growth dynamic model built into the MPC decision module is as follows: x(k+1) = A·x(k) + B·u(k) y(k) = C·x(k) In the formula, x is the state variable, x = [temperature, humidity, carbon dioxide concentration, light intensity]. T ;u is the control input, u = [ventilation volume, lighting power, air conditioning power, sprinkler volume] T ; y is the output variable, y = [temperature, humidity, carbon dioxide concentration, light intensity]T; A, B, and C are the state transition matrix, control input matrix, and output matrix, respectively.

[0044] In this embodiment, matrices A, B, and C are respectively (calibrated as needed): Coefficient A=

[0045] Coefficient B=

[0046] coefficient C=

[0047] The growth environment of mushrooms is complex, with multiple environmental parameters such as temperature, humidity, carbon dioxide concentration, and light intensity interacting with each other. Traditional control methods struggle to comprehensively consider the dynamic changes and interrelationships of these parameters, making precise regulation impossible. To address this issue, this invention incorporates a dynamic mushroom growth model into its MPC decision module. This model can predict environmental parameter trends in advance and make optimization decisions based on the prediction results, thereby achieving precise control of the mushroom growth environment and meeting the needs of mushrooms at different growth stages.

[0048] Specifically, the mushroom growth dynamic model built into the model predictive control decision module is constructed using a linear discrete-time state-space equation. The system needs to determine the model's structure and parameters. The state variable *x* is defined as a column vector consisting of four parameters: temperature, humidity, carbon dioxide concentration, and light intensity. The control input *u* is defined as a column vector consisting of four operational quantities: ventilation volume, lighting power, air conditioning power, and sprinkler volume. The output variable *y* is usually consistent with the state variable of interest and is therefore also defined as the same four-parameter column vector. The specific element values ​​of coefficient matrices A, B, and C need to be calibrated through system identification or mechanism analysis combined with historical operating data. For example, the elements in matrix B quantify the impact of a unit change in ventilation volume on temperature, humidity, and carbon dioxide concentration, respectively.

[0049] In some embodiments of this application, the rolling optimization process for the MPC decision module to generate collaborative control instructions is specifically implemented as follows: First, initialization is performed. At the beginning of each discrete control cycle k, the MPC decision module obtains the estimated value x(k) of the current mushroom house environment state after integration and judgment from the control module, and at the same time obtains or calculates the sequence of environmental parameter setpoints r(k+1) to r(k+Np) for the next Np time steps.

[0050] Next, the future output is predicted. The MPC decision module calls the built-in mushroom growth dynamic model, which is defined by the state transition matrix A, the control input matrix B, and the output matrix C. Based on this model and the current state x(k), the system output prediction vector Y for the next Np time steps is calculated: Y=F·x(k)+Φ·U Where Y is the vector of future output, U is the vector of future control input, and F and Φ are constant matrices derived from matrices A, B, and C, which are related to the prediction time domain Np and the control time domain Nc.

[0051] To ensure that the environmental conditions are as close as possible to the ideal growth conditions, an objective function J is defined. This function has two components: first, minimizing the deviation between the actual state and the ideal state; and second, controlling the change in input or energy consumption. The constructed objective function J is as follows: In the formula, Np is the prediction time domain (e.g., Np=10). Nc is the control time domain (e.g., Nc=5). r(k+i) is the setpoint at a future time. y(k+i|k) is the predicted future output at time k. Δu(k+i|k) is the control increment, Δu(k)=u(k)-u(k-1). Q is the output error weight matrix. R is the control increment weight matrix.

[0052] The objective function J is an optimization problem aimed at minimizing tracking error and control action variation. Its expression is: minJ=(YR) · ·(YR)+ΔU · ·ΔU In the formula, R is the set value vector. The output error weight matrix is... The control increment weight matrix is ​​ΔU, which is the control increment vector. The objective function is solved under constraints to obtain the optimal control sequence. A rolling optimization algorithm is used. In each control cycle, based on the current state estimate, the system behavior over a future period is predicted using the above model, and a series of optimal control actions are calculated according to the objective function. Then, only the first control action is implemented, and the rest are used as a reference for the next cycle.

[0053] The constraints include upper and lower limits for the system state at future time points, upper and lower limits for the control input, and upper and lower limits for the control increment; wherein: Control input constraints: u min ≤u≤u max Control Incremental Constraint: Δu min ≤Δu≤Δu max Output constraint: y min ≤y≤y max Based on the principle of rolling optimization, the MPC decision module extracts and outputs only the first control variable u(k|k) in the sequence as the cooperative control instruction at the current time k, and sends it to the lower-level control module for execution.

[0054] The rolling optimization mechanism of this invention can perform forward-looking solutions within each control cycle with the comprehensive objective of minimizing the overall deviation between the environmental setpoint and the predicted value, as well as system energy consumption. This model-predictive-based optimization process enables the system to anticipate the slow-dynamic characteristics of environmental parameters and plan smooth and precise control actions in advance, achieving fast and accurate dynamic tracking and adjustment.

[0055] In some embodiments of this application, the MPC decision module is further configured to: After each control cycle, the measured values ​​of environmental parameters collected by the sensor components are compared with the prediction results of the mushroom growth dynamic model to obtain the prediction error; based on the prediction error, the initial state of the prediction for the next control cycle is corrected, or the parameters of the mushroom growth dynamic model are adjusted.

[0056] Specifically, at the end of each control cycle, the MPC decision module first acquires the measured values ​​of environmental parameters collected by the sensor components, and simultaneously retrieves the prediction results output by the mushroom growth dynamic model within this cycle. The two are then compared numerically to calculate the prediction error. Subsequently, a correction operation is performed based on this prediction error: if the error stems from initial state deviation, the initial prediction state for the next control cycle is corrected to make the prediction starting point of the new cycle more closely match the actual environment; if the error stems from insufficient adaptability of the model itself, the parameters of the mushroom growth dynamic model are adjusted to enhance the model's adaptability to changes in the actual environment. This process corresponds to... Figure 3 The feedback correction and model update process provides a more accurate basis for the system state estimation and prediction model calculation in the next control cycle.

[0057] Figure 3 This is a flowchart illustrating the predictive control process of the mushroom growth dynamics model provided by the present invention. Figure 3This diagram illustrates the complete workflow of the model predictive control algorithm employed in this invention, representing a closed loop from perception to decision-making, execution, and learning. The feedback correction and model update stage, located at the end of the flowchart, is a crucial step connecting the preceding and following stages. Its inputs come from two upstream stages: first, the actual environmental data collected by sensors after the execution of control actions; and second, the prediction results generated by the predictive model calculation stage, which need to be verified. By comparing these two, this stage calculates the prediction error. Its output is either an updated, more accurate system state estimate or the fine-tuned predictive model itself. These outputs directly serve as inputs for the next control loop, providing an optimized basis for the new round of system state estimation and predictive model calculation.

[0058] This invention, through its built-in feedback correction mechanism, compares the actual measured values ​​with the model's predicted values ​​and corrects errors after each control cycle. This mechanism enables the system to continuously adapt to the dynamic changes at different stages of mushroom growth and external environmental disturbances, automatically adjusting the prediction model or initial state values, thereby ensuring the long-term adaptability and robustness of the control algorithm throughout the entire production cycle.

[0059] Figure 4 The flowchart of the intelligent mushroom cultivation environment control method based on multi-parameter fusion control provided by the present invention is shown below. Figure 4 The method includes: Step S1: The sensor component collects environmental parameters in the mushroom house in real time and transmits them to the control module. The environmental parameters include temperature, humidity, carbon dioxide concentration and light intensity. Step S2: The control module transmits the environmental parameters to the MPC decision module; Step S3: The MPC decision module predicts the evolution trend of environmental parameters based on the mushroom growth dynamic model and generates collaborative control instructions through rolling optimization solution. Step S4: The control module generates a corresponding drive signal according to the cooperative control command and outputs it to the execution device; Step S5: The device responds to the drive signal and performs control actions on the mushroom house environment.

[0060] In some embodiments of this application, after the sensor assembly collects environmental parameters inside the mushroom house in real time and transmits them to the control module, the method further includes: The control module receives the parameters collected by the sensor components and performs data integration and status judgment. The control module has a built-in programmable logic control program that automatically matches environmental control strategies according to the growth stage of the mushrooms.

[0061] In some embodiments of this application, after the device performs the control action in step S5, the method further includes: After each control cycle, the measured values ​​of environmental parameters collected by the sensor components are compared with the prediction results of the mushroom growth dynamic model to obtain the prediction error; based on the prediction error, the initial state of the prediction for the next control cycle is corrected, or the parameters of the mushroom growth dynamic model are adjusted.

[0062] The intelligent mushroom cultivation environment control method based on multi-parameter fusion control provided by the present invention is executed by the intelligent mushroom cultivation environment control system based on multi-parameter fusion control provided in the foregoing embodiments. The intelligent mushroom cultivation environment control system based on multi-parameter fusion control has been described in detail in the foregoing embodiments, and the intelligent mushroom cultivation environment control method based on multi-parameter fusion control will not be described again in this embodiment.

[0063] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise expressly specified and limited, the terms "installed," "connected," or "linked" should be interpreted broadly, for example, as a fixed connection, a detachable connection, or an integral connection; a mechanical connection or an electrical connection; a direct connection or an indirect connection via an intermediate medium; or a connection within two components. Those skilled in the art will understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0064] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

[0065] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A smart mushroom cultivation environment control system based on multi-parameter fusion control, characterized in that, include: A sensor assembly is used to collect environmental parameters inside the mushroom house in real time; wherein, the environmental parameters include temperature, humidity, carbon dioxide concentration and light intensity; The MPC decision module receives environmental parameters transmitted by the control module, predicts the parameter evolution trend based on the mushroom growth dynamic model, generates collaborative control commands through rolling optimization and outputs them to the control module; wherein, the mushroom growth dynamic model represents the dynamic relationship between the environmental parameter state and the control input and output. The control module is connected to the sensor components, the execution device, and the MPC decision module respectively. It is used to transmit the environmental parameters to the MPC decision module and to output the corresponding drive signal to the execution device according to the cooperative control command. An actuator is used to perform control actions on the mushroom house environment in response to the drive signal.

2. The intelligent mushroom cultivation environment control system based on multi-parameter fusion control according to claim 1, characterized in that, The sensor assembly includes: Temperature sensors are deployed at multiple locations inside and outside the mushroom house to collect temperature data. Humidity sensors are deployed at multiple locations inside and outside the mushroom house to collect humidity data. Carbon dioxide concentration sensors are deployed at multiple locations within the mushroom house to collect carbon dioxide concentration data. A light intensity sensor is deployed in the light-exposed area of ​​the mushroom house to collect light intensity data.

3. The intelligent mushroom cultivation environment control system based on multi-parameter fusion control according to claim 1, characterized in that, The control module is a PLC controller. The control module is also used to receive parameters collected by the sensor components and to integrate the data and determine the status. The control module has a built-in programmable logic control program that automatically matches environmental control strategies according to the growth stage of the mushrooms.

4. The intelligent mushroom cultivation environment control system based on multi-parameter fusion control according to claim 1, characterized in that, The execution device includes: Ventilation system is used to regulate the carbon dioxide concentration in the mushroom house and help balance the temperature and humidity. LED lighting system is used to adjust the light intensity inside the mushroom house; An air conditioning system is used to regulate the temperature inside the mushroom house; The spray system is used to regulate the humidity inside the mushroom house.

5. The intelligent mushroom cultivation environment control system based on multi-parameter fusion control according to claim 1, characterized in that, The mushroom growth dynamic model is as follows: x(k+1) = A·x(k) + B·u(k) y(k) = C·x(k) In the formula, x is the state variable, x = [temperature, humidity, carbon dioxide concentration, light intensity]. T ;u is the control input, u = [ventilation volume, lighting power, air conditioning power, sprinkler volume] T ; y is the output variable, y = [temperature, humidity, carbon dioxide concentration, light intensity] T A, B, and C are the state transition matrix, control input matrix, and output matrix, respectively.

6. The intelligent mushroom cultivation environment control system based on multi-parameter fusion control according to claim 5, characterized in that, The method of predicting parameter evolution trends based on a mushroom growth dynamic model, generating collaborative control commands through rolling optimization and outputting them to the control module includes: Obtain the current state x(k) and the setpoint sequence r(k+1), ..., r(k+Np), where Np is the prediction time domain; Based on the aforementioned mushroom growth dynamic model, the system output Y for the next Np time steps is predicted by the following expression: Y=F·x(k)+Φ·U In the formula, Y is the vector of future output, U is the vector of future control input, and F and Φ are matrices derived from matrices A, B, and C. With the goal of minimizing tracking error and control action variation, an objective function J is constructed and the optimization problem is solved, expressed as: minJ=(YR) · ·(YR)+ΔU · ·ΔU In the formula, R is the set value vector. The output error weight matrix is... The control increment weight matrix is ​​ΔU, which is the control increment vector. The objective function is solved under constraints to obtain the optimal control sequence. The constraints include upper and lower limits of the system state at future time points, upper and lower limits of the control input, and upper and lower limits of the control increment. The first control variable in the optimal control sequence is output as the cooperative control command.

7. The intelligent mushroom cultivation environment control system based on multi-parameter fusion control according to claim 1, characterized in that, The MPC decision module is also used for: After each control cycle, the measured values ​​of environmental parameters collected by the sensor components are compared with the prediction results of the mushroom growth dynamic model to obtain the prediction error; based on the prediction error, the initial state of the prediction for the next control cycle is corrected, or the parameters of the mushroom growth dynamic model are adjusted.

8. A method for regulating the intelligent mushroom cultivation environment based on multi-parameter fusion control, applied to the intelligent mushroom cultivation environment regulation system based on multi-parameter fusion control as described in any one of claims 1-7, characterized in that, The method includes: The sensor assembly collects environmental parameters inside the mushroom house in real time and transmits them to the control module. The environmental parameters include temperature, humidity, carbon dioxide concentration, and light intensity. The control module transmits the environmental parameters to the MPC decision module; The MPC decision module predicts the evolution trend of environmental parameters based on the mushroom growth dynamic model and generates collaborative control commands through rolling optimization. The control module generates corresponding drive signals based on the coordinated control instructions and outputs them to the execution device; The device responds to the drive signal and performs control actions on the mushroom house environment.

9. The intelligent mushroom cultivation environment control method based on multi-parameter fusion control according to claim 8, characterized in that, After the sensor components collect environmental parameters inside the mushroom house in real time and transmit them to the control module, the method further includes: The control module receives the parameters collected by the sensor components and performs data integration and status judgment. The control module has a built-in programmable logic control program that automatically matches environmental control strategies according to the growth stage of the mushrooms.

10. The intelligent mushroom cultivation environment control method based on multi-parameter fusion control according to claim 8, characterized in that, After the device performs the control action, the method further includes: After each control cycle, the measured values ​​of environmental parameters collected by the sensor components are compared with the prediction results of the mushroom growth dynamic model to obtain the prediction error; based on the prediction error, the initial state of the prediction for the next control cycle is corrected, or the parameters of the mushroom growth dynamic model are adjusted.

Images