Multi-parameter environment closed-loop regulation and control method and system based on FPGA

Abstract

The invention relates to the technical field of environment regulation and control, and discloses a multi-parameter environment closed-loop regulation and control method and system based on an FPGA, and the method comprises the steps: unifying monitoring data to a same space coordinate system through the FPGA according to the monitoring data collected by a sensor network disposed in an environment, analyzing the coupling relation between different parameter data, and obtaining fusion data; based on the fusion data, analyzing the environment crop state through a preset crop state recognition model, simulating the influence condition of different parameter adjustment processes on the crop state, and generating a first control instruction set; analyzing the degree of deviation between the environmental parameters in the fused data and a preset target value, calculating the priority of each control instruction in the first control instruction set, and merging and optimizing the control instructions; the environmental parameters are corrected; according to the application, the regulation and control efficiency can be improved, the environmental disturbance and the crop demand change can be quickly responded through FPGA parallel processing, and the environmental regulation and control efficiency and the regulation and control effect are improved.

Description

Technical Field

[0001] This application relates to the field of environmental control technology, and more specifically to a multi-parameter environmental closed-loop control method and system based on FPGA. Background Technology

[0002] Currently, crop growth and development are highly dependent on the precise coordination of four major environmental factors: temperature, humidity, CO2 concentration, and light. The current temperature and humidity control in greenhouses mainly relies on manual operation based on experience, which has inherent defects such as control lag, inaccurate environmental control, and poor stability. This leads to high production risks, inconsistent quality, and low resource utilization, hindering the industry's upgrade towards standardization and large-scale production, and failing to meet the environmental requirements of crops at different growth stages. Furthermore, environmental monitoring is relatively limited and cannot effectively address complex situations such as uneven sensor distribution and numerous interference factors within the environment, resulting in insufficient accuracy and reliability of monitoring results, and failing to provide a solid data foundation for control decisions.

[0003] The existing technology has the following problems: it adopts single-parameter independent control, ignores the correlation between environmental parameters, which will cause conflicts between parameters, making it difficult to achieve a globally optimal stable state and affecting crop growth; it generates control commands based on simple threshold judgment, lacks perception and predictive regulation of the overall environmental state, and has a large calculation delay, which cannot meet the real-time closed-loop response requirements of highly dynamic environments; in order to solve at least one of the above problems, this application proposes a multi-parameter environmental closed-loop control method and system based on FPGA. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the purpose of this application is to provide a multi-parameter environmental closed-loop control method and system based on FPGA, which can effectively solve the problems in the background technology. The specific technical solution of this application is as follows:

[0005] FPGA-based multi-parameter environmental closed-loop control methods include:

[0006] Based on the monitoring data collected by the sensor network set up in the environment, the monitoring data is unified into the same spatial coordinate system through FPGA, the coupling relationship between different parameter data is analyzed, a coupling matrix is ​​constructed, and multi-parameter data is coupled to obtain fused data;

[0007] Based on fused data, the environmental crop status is analyzed through a preset crop status identification model, the impact of different parameter adjustment processes on crop status is simulated, and a first set of control instructions is generated.

[0008] The deviation between environmental parameters and preset target values ​​in the fused data is analyzed, the priority of each control command in the first control command set is calculated, and the control commands are merged and optimized to obtain the second control command set.

[0009] The environmental parameters are adjusted and corrected according to the second set of control instructions in order to achieve closed-loop control of the environment.

[0010] Specifically, the monitoring data collected by the sensor network set up in the environment is unified into the same spatial coordinate system through FPGA, the coupling relationship between different parameter data is analyzed, a coupling matrix is ​​constructed, and multi-parameter data is coupled to obtain fused data, including:

[0011] Based on the monitoring data collected by the sensor network set up in the environment, the monitoring data is uniformly mapped to the same spatial coordinate system through FPGA to obtain the first data point cloud set;

[0012] Based on the first data point cloud set, the correlation between data with different parameters is analyzed, the parameter coupling strength is calculated, the impact of data changes on the first data point cloud set is analyzed, and the spatial influence tensor is calculated.

[0013] By combining the parameter coupling strength and the spatial influence tensor, the time delay between different parameter data is analyzed, a coupling matrix is ​​constructed, and multi-parameter data is coupled to obtain fused data.

[0014] Specifically, the step of analyzing the correlation between different parameter data based on the first data point cloud set, calculating the parameter coupling strength, analyzing the impact of data changes on the first data point cloud set, and calculating the spatial influence tensor includes:

[0015] According to the preset spatial grid, analyze the correlation between different parameter data in each spatial grid of the first data point cloud set, filter out data with correlation greater than the preset correlation threshold, establish connection edges, and obtain the first connection edge set;

[0016] Based on the first set of connecting edges, simulate the impact of parameter changes within a spatial grid on the parameters of adjacent spatial grids, calculate the parameter coupling strength between spatial grids, and construct a grid coupling matrix;

[0017] By combining the spatial distribution and parameter coupling strength of the first data point cloud set, we analyze the propagation of parameter changes between spatial grids, calculate the corresponding propagation distance and medium attenuation coefficient, and construct a spatial propagation weight matrix.

[0018] Based on the grid coupling matrix and the spatial propagation weight matrix, the coupling transformation and total spatial propagation effect of the parameters within each spatial grid are calculated, and the spatial influence tensor is encoded.

[0019] Specifically, by combining the parameter coupling strength and the spatial influence tensor, the time delay between different parameter data is analyzed, a coupling matrix is ​​constructed, and multi-parameter data is coupled to obtain fused data, including:

[0020] Based on the pre-acquired historical control command data and the first data point cloud set, analyze the time difference distribution between control commands and parameter change responses, and construct the first time delay set;

[0021] Based on the parameter coupling strength, the propagation speed between corresponding spatial grids in the grid coupling matrix and the spatial propagation weight matrix is ​​analyzed, the first time delay set is optimized to obtain the second time delay set, and the time delay tensor is extracted.

[0022] The spatial influence tensor and the time delay tensor are correlated and calculated to construct a coupling matrix;

[0023] The first data point cloud set is time-delayed and coupled according to the coupling matrix to obtain fused data.

[0024] Specifically, based on fused data, the environmental crop state is analyzed through a preset crop state recognition model, and the impact of different parameter adjustment processes on the crop state is simulated to generate a first set of control instructions, including:

[0025] Based on fused data, the environmental crop status is analyzed through a pre-set crop status identification model, and the corresponding crop status index is calculated.

[0026] By combining the crop state index and the preset expected growth state, the trajectory of environmental parameter changes corresponding to the state change process is analyzed, and a first set of environmental parameters is constructed.

[0027] Based on the first set of environmental parameters, the impact of different parameter adjustment processes on crop status is simulated to generate a first set of control commands.

[0028] Specifically, the analysis and fusion of data includes determining the deviation between environmental parameters and preset target values, calculating the priority of each control command in the first control command set, merging and optimizing the control commands to obtain a second control command set, including:

[0029] Analyze the degree of deviation between environmental parameters and preset target values ​​in the fused data, analyze the deviation spread range according to the degree of deviation, analyze the deviation threat situation, and calculate the priority of each control command in the first control command set;

[0030] Based on priority from high to low, control commands with overlapping resources and coupled effects are identified, and the control commands are merged and optimized to obtain a second set of control commands.

[0031] Specifically, the analysis of the deviation between environmental parameters and preset target values ​​in the fused data includes: analyzing the deviation spread range based on the degree of deviation, analyzing the deviation threat situation, and calculating the priority of each control command in the first control command set, including:

[0032] Analyze the degree of deviation between environmental parameters and preset target values ​​in the fused data, calculate the deviation value, analyze the corresponding deviation impact intensity and diffusion range, and construct a deviation descriptor;

[0033] Based on the deviation descriptor, analyze the deviation threat situation and construct a deviation threat vector;

[0034] Based on the deviation threat vector, the threat index and urgency index are analyzed using a preset threat analysis model to obtain the deviation analysis results;

[0035] Based on the deviation analysis results, the priority of each control instruction in the first control instruction set is calculated.

[0036] Specifically, the control commands that overlap in resources and have coupled effects are identified in descending order of priority, and the control commands are merged and optimized to obtain a second set of control commands, including:

[0037] Analyze the spatial influence domain of each control instruction in the first set of control instructions, identify control instructions with overlapping resources and coupled effects in the spatial influence domain, and obtain the set of conflicting instructions;

[0038] Instructions with priority differences less than a preset priority threshold in the conflict instruction set are merged and optimized in descending order of priority to obtain an optimized instruction set.

[0039] Based on the optimized instruction set, the corresponding control instructions in the first control instruction set are replaced to obtain the second control instruction set.

[0040] Specifically, the step of adjusting environmental parameters according to the second set of control instructions and correcting the environmental parameters to achieve closed-loop control of the environment includes:

[0041] The environmental parameters of the second set of control commands are analyzed to obtain the corresponding drive commands.

[0042] According to the driving instructions, the environmental parameters are corrected to achieve closed-loop control of the environment.

[0043] An FPGA-based multi-parameter environmental closed-loop control system is used to implement the aforementioned FPGA-based multi-parameter environmental closed-loop control method, including:

[0044] The data fusion module uses an FPGA to unify the monitoring data collected by the sensor network set up in the environment into the same spatial coordinate system, analyzes the coupling relationship between different parameter data, constructs a coupling matrix, couples multiple parameter data, and obtains fused data.

[0045] The instruction generation module, based on fused data, analyzes the environmental crop status through a preset crop status recognition model, simulates the impact of different parameter adjustment processes on the crop status, and generates the first set of control instructions.

[0046] The instruction optimization module analyzes the deviation between environmental parameters and preset target values ​​in the fused data, calculates the priority of each control instruction in the first control instruction set, and merges and optimizes the control instructions to obtain the second control instruction set.

[0047] The environmental control module adjusts and corrects environmental parameters according to the second set of control instructions to achieve closed-loop control of the environment.

[0048] The beneficial effects of this application are as follows: By leveraging the parallel computing capabilities of FPGA, multi-source, heterogeneous sensor data can be mapped in real time to a unified spatial coordinate system, forming a data point cloud. This provides a data foundation for parameter analysis. By analyzing the parameter coupling strength and constructing a coupling matrix, the spatiotemporal correlation between environmental parameters can be incorporated into the analysis process to improve the joint regulation effect of parameters. Based on environmental parameters, the potential impact of different parameter adjustment processes on crop status can be simulated to generate control commands, thereby improving the foresight and target orientation of control commands. By analyzing the severity of environmental deviations, calculating the dynamic priority of control commands, identifying conflicting commands, and merging and optimizing them, the control effect and synergy of control commands can be improved. By compensating for the coupling effect and time delay between parameters through multi-parameter coupling, regulation conflicts can be avoided, and regulation efficiency can be improved. Parallel processing via FPGA can quickly respond to environmental disturbances and changes in crop demand, improving the efficiency and effect of environmental regulation. Attached Figure Description

[0049] Figure 1 This is a flowchart illustrating the process of the FPGA-based multi-parameter environmental closed-loop control method in the embodiments of this application.

[0050] Figure 2 This is a schematic diagram of the system hardware structure in an embodiment of this application;

[0051] Figure 3 This is a schematic diagram of the first connecting edge in an embodiment of this application;

[0052] Figure 4 This is a schematic diagram of the structure of the FPGA-based multi-parameter environmental closed-loop control system in the embodiments of this application. Detailed Implementation

[0053] The present application will be further described in detail below with reference to the accompanying drawings and embodiments.

[0054] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0055] Hereinafter, the terms "first," "second," and other generic terms are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0056] refer to Figure 1 The diagram illustrates a specific implementation of the FPGA-based multi-parameter environmental closed-loop control method of this application, including:

[0057] S101. Based on the monitoring data collected by the sensor network set in the environment, the monitoring data is unified into the same spatial coordinate system through FPGA, the coupling relationship between different parameter data is analyzed, a coupling matrix is ​​constructed, and multi-parameter data is coupled to obtain fused data.

[0058] S102. Based on the fused data, analyze the environmental crop status through a preset crop status identification model, simulate the impact of different parameter adjustment processes on the crop status, and generate a first set of control instructions.

[0059] S103. Analyze the degree of deviation between environmental parameters and preset target values ​​in the fused data, calculate the priority of each control command in the first control command set, merge and optimize the control commands to obtain the second control command set;

[0060] S104. Adjust the environmental parameters according to the second set of control instructions, and correct the environmental parameters to perform closed-loop control of the environment.

[0061] In this embodiment, based on the monitoring data collected by the sensor network set up in the environment, the monitoring data is unified into the same spatial coordinate system through the parallel I / O and high-speed processing capabilities of the FPGA. Inside the FPGA, parallel computing units analyze the coupling relationship between different parameter data, construct a coupling matrix, and couple multi-parameter data to obtain fused data. Through the real-time processing of the FPGA, the problem of difficulty in unifying and utilizing multi-source heterogeneous data can be effectively solved. By constructing the coupling matrix, the real interaction relationship and spatiotemporal delay characteristics between environmental parameters can be dynamically reflected. Environmental regulation analysis based on environmental state data containing rich spatiotemporal correlation information can avoid control command conflicts caused by ignoring parameter coupling, thereby improving the accuracy and coordination of the environmental regulation process.

[0062] Specifically, based on fused data, the system analyzes the environmental crop state through a pre-defined crop state identification model, simulates the impact of different parameter adjustment processes on crop state, analyzes various parameter adjustment schemes to predict the future trend of crop state indices under each scheme, and compares the simulation results to select the adjustment scheme that can make the crop state approach the desired state most quickly and smoothly, generating a first set of control instructions. By simulating the effects of adjustment schemes and generating control instructions, the foresight and predictability of control instructions can be improved. Before actually issuing instructions, the potential effects of different regulation strategies can be evaluated, and the scheme most beneficial to crop growth can be selected. This can effectively avoid the problem of ineffective regulation caused by focusing only on the parameters themselves, and improve the accuracy and bioeffectiveness of the regulation process.

[0063] Specifically, the deviation between environmental parameters and preset target values ​​in the fused data is analyzed to assess the threat level of the deviation, including but not limited to the intensity of the deviation, the spatiotemporal diffusion range of the impact, and the degree of harm to crop status. The priority of each control command in the first control command set is calculated, and conflicting commands are identified and merged to optimize the control commands, resulting in a second control command set. Through the dynamic prioritization mechanism of threat assessment and command conflict resolution, frequent oscillations or invalid actions of actuators caused by receiving conflicting commands can be avoided, reducing equipment wear. By merging coordinating commands, the number of actuator actions can be reduced, energy consumption can be reduced, and it can be ensured that in the case of limited resources or sudden interference, the problems that pose the greatest threat to crop growth can be addressed first, thereby improving the control effect and environmental adaptability of the entire control system.

[0064] Specifically, environmental parameters are adjusted according to the second set of control instructions, which are then parsed into drive instructions for actuators, including but not limited to humidifiers, fans, shading curtain motors, and LED driver circuits, to correct the environmental parameters. Simultaneously, the sensor network continuously collects the latest monitoring data and immediately feeds it back to the input, initiating a new round of data fusion, state analysis, decision optimization, and execution loops for closed-loop environmental control. Utilizing the hardware determinism and multi-interface parallel drive capabilities of the FPGA ensures that the optimized control strategy is executed promptly and accurately. By highly integrating the entire closed-loop process within a single hardware platform, the uncertainty and latency of communication between different hardware modules are reduced, improving the response speed and reliability of the control system. This allows for rapid, stable, and coordinated adjustment to the optimal state in the event of external disturbances or internal changes, enhancing the accuracy of environmental parameters and creating a continuous and suitable growth environment for crops.

[0065] This application leverages the parallel computing capabilities of FPGAs to map multi-source, heterogeneous sensor data to a unified spatial coordinate system in real time, forming a data point cloud. This provides a data foundation for parameter analysis. By analyzing parameter coupling strength and constructing a coupling matrix, the spatiotemporal correlation between environmental parameters can be incorporated into the analysis process to improve the joint regulation effect. Simulating the potential impact of different parameter adjustment processes on crop status based on environmental parameters and generating control commands enhances the foresight and target orientation of these commands. Analyzing the severity of environmental deviations, calculating the dynamic priority of control commands, identifying conflicting commands, and merging and optimizing them improves the control effect and synergy of the commands. By compensating for coupling effects and time delays between parameters through multi-parameter coupling, regulation conflicts can be avoided, and regulation efficiency can be improved. FPGA parallel processing enables rapid response to environmental disturbances and changes in crop demand, enhancing environmental regulation efficiency and effectiveness.

[0066] like Figure 2 As shown, the system is based on the XC7Z020-2CLG484I and integrates a dual-core ARM Cortex-A9 processor, whose performance far exceeds that of traditional microcontrollers. It can easily run complex control algorithms and integrates DDR controller, SDIO, UART, SPI, I2C, etc., allowing a complete embedded computer system to be built without external chips. It can simultaneously process massive amounts of data from 14 temperature channels, 10 sets of RS485 devices, and multiple valve statuses, and reserves ample performance margin for future algorithm upgrades.

[0067] Specifically, it employs 14 sets of PT1000 RTDs (high-precision temperature sensors) and combines them with an 8-channel, 12-bit, high-speed, low-power ADC (ADC128S102) for digitization. This chip consumes only 2.3mW at a sampling rate of up to 1MSPS with a 3V power supply, and can reduce power consumption to 0.06µW in low-power mode. This enables millivolt-level monitoring of multi-point temperatures in the environment, ensuring data accuracy.

[0068] The system utilizes the ADC128S102 and AD5676 to monitor and control valves for hot water, cold water, and humidification. The AD5676 chip's core advantage is its 16-bit precision and integrated eight independent channels, greatly simplifying the design of multi-channel analog output systems. The design incorporates ten RS485 interfaces, which benefit from differential signal transmission, offering strong resistance to common-mode noise. This allows for stable data transmission over distances of hundreds to thousands of meters, ensuring reliable communication even in complex environments. Five RS485 interfaces are connected to the Stevenson screen sensor, differential pressure sensor, fan inverter, carbon dioxide monitoring device, and intelligent serial port display, respectively, with three RS485 interfaces reserved for future system expansion.

[0069] Specifically, the external devices connected to the RS485 side include a Stevenson screen sensor (JXBS-J001-0XZX), a differential pressure sensor, a fan frequency converter (CHF100), a carbon dioxide detection device (IDG100-D-CO2), and a smart serial port display (HYW121F01). The Stevenson screen sensor (JXBS-J001-0XZX) is highly integrated internally, capable of simultaneously detecting meteorological elements such as wind direction, wind speed, air temperature, humidity, atmospheric pressure, illuminance, carbon dioxide concentration, PM2.5, PM10, oxygen concentration, ammonia concentration, hydrogen sulfide concentration, and noise. It employs a high-sensitivity digital probe, ensuring stable signal and high accuracy. It features a wide measurement range, good linearity, excellent waterproof performance, ease of use, convenient installation, and long transmission distance. The overall structure is a waterproof weather Stevenson screen, suitable for various environments and possessing excellent corrosion resistance. Differential Pressure Sensor: The 984A / Q series automatically and periodically performs zero-point calibration, eliminating measurement errors caused by zero-point drift, significantly reducing maintenance costs and workload, and ensuring long-term data accuracy and reliability. It can monitor pressures from extremely low pressure (±25Pa) to medium to high pressure (250kPa), with linearity error ≤±1% FS and repeatability accuracy ≤±0.2%FS. Furthermore, the integrated LED display allows direct reading of pressure values, facilitating debugging and inspection. Fan Inverter (CHF100): Enables smooth, shock-free starting of rotating motors, with no shutdown during momentary power outages. Supports programmable digital and analog inputs, and programmable open-collector, relay, and analog outputs. Features 16-speed control and provides overcurrent, overvoltage, and overload protection. Carbon Dioxide Detector (IDG100-D-CO2): Internally integrated, it simultaneously detects temperature, humidity, and CO2 concentration over a wide measurement range. The HYW121F01 intelligent serial port display uses a 600MHz RISC-V architecture processor, whose performance far exceeds that of traditional serial port display 8 / 32-bit MCUs, enabling smooth processing of complex UIs, images, and 1080P videos. It supports firmware and UI upgrades via USB flash drive, UART, and WiFi, making it very convenient for function updates or maintenance after device deployment.

[0070] The system is designed with an SD card interface to continuously record all raw sensor data, device operation logs, alarm information, etc., for later fault diagnosis, performance analysis, and algorithm optimization. The Type-C interface enables serial port debugging and program burning, simplifying the development and testing process.

[0071] Furthermore, based on the monitoring data collected by the sensor network set up in the environment, the monitoring data is unified into the same spatial coordinate system through the FPGA, the coupling relationship between different parameter data is analyzed, a coupling matrix is ​​constructed, and multi-parameter data is coupled to obtain fused data, including:

[0072] S201. Based on the monitoring data collected by the sensor network set in the environment, the monitoring data is uniformly mapped to the same spatial coordinate system through the FPGA to obtain the first data point cloud set;

[0073] S202. Based on the first data point cloud set, analyze the correlation between data with different parameters, calculate the parameter coupling strength, analyze the impact of data changes on the first data point cloud set, and calculate the spatial influence tensor.

[0074] S203. Combining parameter coupling strength and spatial influence tensor, analyze the time delay between different parameter data, construct a coupling matrix, couple multi-parameter data, and obtain fused data.

[0075] In this embodiment, based on the monitoring data collected by the sensor network set up in the environment, including but not limited to temperature and humidity sensors, light sensors, and CO2 sensors, the physical location of the sensor network deployed in the environment is calibrated to determine the three-dimensional spatial coordinates of each sensor node. When the FPGA receives the real-time monitoring data packets uploaded by each sensor node via wired or wireless means, it parses the node identifier carried in the data packet and maps it to the accurate spatial coordinates. Through the parallel processing architecture of the FPGA, the spatial coordinates are mapped to the same spatial coordinate system to obtain data points with spatial location and parameter attributes. The data point information is integrated to construct the first data point cloud set.

[0076] It should be noted that by constructing a data point cloud through spatial mapping, independent sensor readings with only timestamps are converted into a data point cloud with unified spatial coordinates. This provides an accurate data foundation for spatial analysis and collaborative control, avoids logical errors caused by coordinate confusion, and improves the accuracy of the parameter analysis process.

[0077] Specifically, based on the first data point cloud set, the first data point cloud set is divided into a preset three-dimensional grid according to its spatial location. The correlation between different parameter data is analyzed, the parameter coupling strength is calculated, and the impact of data changes on the first data point cloud set is analyzed. The spatial influence tensor is calculated. By calculating the spatial influence tensor, the strength and spatial coupling relationship between environmental parameters can be reflected. The analyzed environmental disturbances not only change the local temperature, but also spread spatially and affect other parameters such as humidity in distant places, thus enabling coupled regulation of internal environmental parameters. The spatial influence tensor is dynamically calculated based on real-time data. The spatial influence tensor can be adjusted according to the coupling relationship corresponding to changes in environmental structure, improving the robustness of the calculation process and the synergy of the regulation process.

[0078] Specifically, by combining parameter coupling strength and spatial influence tensor, the time delay between different parameter data is analyzed, the time delay tensor is calculated, and a coupling matrix is ​​constructed. Multi-parameter data is then coupled through time delay compensation to obtain fused data. By constructing a coupling matrix to compensate for and fuse the time delay of the original data, the accuracy of the fused data and the environmental response speed can be improved, enhancing the speed and realism of environmental state perception. Parameter fusion through time delay compensation can avoid making premature or repetitive control decisions based on parameters with delays, reducing overshoot and oscillations in the control system, and improving the stability, response speed, and control effect of the control system.

[0079] Furthermore, based on the first data point cloud set, the correlation between data with different parameters is analyzed, the parameter coupling strength is calculated, the impact of data changes on the first data point cloud set is analyzed, and the spatial influence tensor is calculated, including:

[0080] S301. According to the preset spatial grid, analyze the correlation between different parameter data in each spatial grid of the first data point cloud set, filter out data with correlation greater than the preset correlation threshold, establish connection edges, and obtain the first connection edge set.

[0081] S302. Based on the first set of connecting edges, simulate the influence of parameter changes within a spatial grid on the parameters of adjacent spatial grids, calculate the parameter coupling strength between spatial grids, and construct a grid coupling matrix.

[0082] S303. Combining the spatial distribution and parameter coupling strength of the first data point cloud set, analyze the propagation of parameter changes between spatial grids, calculate the corresponding propagation distance and medium attenuation coefficient, and construct a spatial propagation weight matrix.

[0083] S304. Based on the grid coupling matrix and the spatial propagation weight matrix, calculate the coupling transformation and total spatial propagation effect of the parameters within each spatial grid, and encode the spatial influence tensor.

[0084] In this embodiment, according to a preset spatial grid, the correlation between different parameter data within each spatial grid of the first data point cloud set is analyzed. The spatial grid can be set automatically according to the environmental control precision. For each grid, the FPGA extracts all data points falling within that grid in parallel. Data values ​​of the same parameter at different timestamps are grouped into their respective time series according to parameter type. For each pair of different parameter combinations within the grid, the Pearson correlation coefficient of the corresponding time series is calculated. A correlation threshold is set according to the environmental control precision, and each correlation coefficient is compared with a preset correlation threshold. Figure 3 As shown, parameter pairs with correlation coefficients greater than the correlation threshold are selected, and an undirected connection edge is created between the corresponding parameter data points of the parameter pairs. All connection edges constructed within the grid are integrated to obtain the first set of connection edges.

[0085] It should be noted that by using grid partitioning and correlation threshold filtering, environmental monitoring data can be quickly filtered, effectively filtering out spurious correlations caused by random noise, ensuring that the set of connecting edges can reflect the real coupling relationship between environmental parameters, providing a high-quality correlation network foundation for subsequent analysis, avoiding model overfitting and computational redundancy caused by analyzing all statistical fluctuations, and improving the accuracy and efficiency of the analysis process.

[0086] Furthermore, based on the first set of connecting edges, each spatial grid is traversed to simulate the impact of parameter changes within a spatial grid on the parameters of adjacent spatial grids. For the current spatial grid, the FPGA assumes a unit step change in a parameter within the source grid based on the connecting edges within the grid. By utilizing a large amount of historical change impact data and a pre-trained neural network model, the impact of the unit change on other related parameters within the spatial grid through the connecting edge relationships within the grid is analyzed. Simultaneously, the impact of the unit change on each parameter within adjacent spatial grids is analyzed, and the parameter coupling strength is calculated. The parameter coupling strength within the spatial grid and the parameter coupling strength between adjacent grids are combined according to the grid and parameter order to construct a grid coupling matrix.

[0087] It is important to emphasize that by constructing a grid coupling matrix, we can accurately predict the effect of a local control action on parameters within and between adjacent grids. Combined with spatial adjacent grid analysis, this can improve the accuracy of spatial influence judgment results, provide accurate direct interaction data for spatial propagation analysis, and enhance the accuracy of spatial coordinated control.

[0088] Specifically, by combining the spatial distribution and parameter coupling strength of the first data point cloud set, the propagation of parameter changes between spatial grids is analyzed. The Euclidean distance between the geometric centers of the spatial grids is calculated as the basic propagation distance. By using a nonlinear correlation model pre-trained with a large amount of historical parameter change data corresponding to the disturbance attenuation ratio, the corresponding medium attenuation coefficient is calculated. Weights are set according to the degree of influence of the numerical value on the parameter change. The propagation distance and the medium attenuation coefficient are weighted and summed to calculate the corresponding spatial propagation weights. The spatial propagation weights are integrated to construct a spatial propagation weight matrix.

[0089] It should be noted that by constructing a spatial propagation weight matrix, the effective range and influence gradient of regulatory actions can be realistically simulated, which can improve the accuracy and realism of the simulation results. By quantitatively analyzing the decay process, regions with high spatial influence weights can be prioritized for adjustment during collaborative optimization, thereby improving energy utilization efficiency and regulatory effectiveness during environmental regulation.

[0090] Specifically, based on the mesh coupling matrix and the spatial propagation weight matrix, the mesh coupling matrix is ​​used as a linear transformation operator, and the spatial propagation weight matrix is ​​used as a spatial attenuation modulator for the output of the linear transformation operator. In the FPGA, matrix power series calculations are performed using a fixed-point iterative circuit to simulate the multiple transmission and superposition processes of influence, resulting in a spatial influence tensor. This spatial influence tensor has multiple dimensions, including but not limited to mesh index and parameter type. Each element in the spatial influence tensor quantifies the combined effect on other parameters when a specific parameter of the quantized mesh undergoes a unit change. By calculating the spatial influence tensor, the combined impact of control operations on every location and parameter in the environment can be quickly predicted, providing accurate data support for the control process, multi-objective optimization, and conflict resolution. This avoids conflicting effects between strategies, improves the synergy of the control process, and enhances the control effect and accuracy.

[0091] Furthermore, by combining the parameter coupling strength and the spatial influence tensor, the time delay between different parameter data is analyzed, a coupling matrix is ​​constructed, and multi-parameter data are coupled to obtain fused data, including:

[0092] S401. Based on the pre-acquired historical control command data and the first data point cloud set, analyze the time difference distribution between the control command and the parameter change response, and construct the first time delay set;

[0093] S402. Based on the parameter coupling strength, analyze the propagation speed between corresponding spatial grids in the grid coupling matrix and the spatial propagation weight matrix, optimize the first time delay set to obtain the second time delay set, and extract the time delay tensor.

[0094] S403. Perform correlation calculations on the spatial influence tensor and the time delay tensor to construct a coupling matrix;

[0095] S404. Perform time delay compensation and coupling on the first data point cloud set according to the coupling matrix to obtain fused data.

[0096] In this embodiment, control commands and corresponding timestamps issued within a historical time period are acquired, along with point clouds of data with spatial grid labels reported by all sensors within the corresponding time period. Each type of control command is analyzed independently, and the target parameters and spatial grids expected to be affected by the control command are selected. The first-order difference of the corresponding parameter time series is calculated, and the moment when the parameter changes significantly due to the control command is identified. The difference between the change moment and the command issuance moment is calculated to obtain the time difference distribution between the control command and the parameter change response. The calculated time differences are integrated to obtain the first time delay set. By calculating the time difference, the dynamic delay characteristics of the system execution link can be quantified, providing accurate data support for time delay compensation. By constructing the first time delay set, accurate timing coordinated control can be performed, avoiding control overshoot problems caused by excessive time delay.

[0097] Specifically, based on the parameter coupling strength, the propagation speed between corresponding spatial grids in the grid coupling matrix and spatial propagation weight matrix is ​​analyzed. Physical propagation information is extracted from the grid coupling matrix and spatial propagation weight matrix. The parameter coupling strength reflects the strength of the mutual influence of parameters, and the spatial propagation weight reflects the propagation speed. By using a heat conduction model pre-trained with a large amount of historical grid coupling matrix and spatial propagation weight matrix data, the theoretical propagation time and speed range of disturbances between each pair of grids are analyzed. The empirical time delay values ​​in the first time delay set are compared with the corresponding theoretical propagation time to obtain the time difference. According to the environmental control accuracy requirements, a difference threshold is set, and abnormal data points with time differences greater than the difference threshold are screened out. Spatial interpolation estimation is performed using the known time delay of neighboring grids and the calculated theoretical propagation speed. The optimized and spatially interpolated data is used as the second time delay set. The time delay values ​​are integrated according to the corresponding grid and parameters to obtain a time delay tensor that matches the dimensional structure of the spatial influence tensor. Each element in the time delay tensor represents the delay time required for the corresponding influence. By combining internal coupling and propagation within the environment, noise in time-delay data can be effectively removed, improving the reliability and consistency of the time-delay estimation process, and enhancing the response speed and environmental adaptability of the control process.

[0098] Specifically, spatial influence tensors and time delay tensors are correlated and calculated through tensor operations. The spatial influence tensor is used as a column vector, and the time delay tensor is used as a row vector. After multiplication, a coupling matrix is ​​constructed. By constructing the coupling matrix, the inertia of the environmental response can be dynamically reflected. Combining spatial coupling and time delay can improve the realism of the analysis process and predict the chain effects of control actions at different points in the future, providing accurate data support for predictive control and the formulation of multi-time series coordinated strategies.

[0099] Specifically, the first data point cloud set is time-delay compensated and coupled according to the coupling matrix. For each data point in the point cloud, the corresponding influence weight is located according to the coupling matrix, the weight and time-delay compensation are calculated, and the data is stripped from or superimposed in reverse from the current observation value to reconstruct the system state. After parallel processing of all data points, fused data is obtained. By fusing data, the confusion of historical action delays can be eliminated, the instantaneous coupling interference between parameters can be reduced, the accuracy and foresight of control decisions can be improved, and misjudgments and invalid operations caused by sensor reading lag or coupling interference can be avoided, thereby improving the stability of the closed-loop control system.

[0100] Furthermore, based on the fused data, the environmental crop state is analyzed through a pre-defined crop state identification model, simulating the impact of different parameter adjustment processes on the crop state, and generating a first set of control instructions, including:

[0101] S501. Based on fused data, analyze the environmental crop status through a preset crop status identification model and calculate the corresponding crop status index.

[0102] S502. Combining the crop state index and the preset expected growth state, analyze the trajectory of environmental parameter changes corresponding to the state change process, and construct a first set of environmental parameters.

[0103] S503. Based on the first set of environmental parameters, simulate the impact of different parameter adjustment processes on crop status and generate a first set of control commands.

[0104] In this embodiment, based on fused data, the crop status in the environment is analyzed through a preset crop status recognition model. The crop status recognition model includes, but is not limited to, a convolutional neural network model pre-trained using a large amount of historical fused data. The model analyzes the fused data and outputs a calculated crop status index. The crop status recognition model can quickly calculate the corresponding crop status index, analyze the real-time growth status of crops, provide accurate data support for the regulation decision-making process, and improve the pertinence and biological effectiveness of the regulation process.

[0105] Specifically, by combining crop state indices and preset desired growth states, and analyzing historical data to learn crop growth mechanisms, the system determines the correspondence between typical state deviations and recommended environmental parameter adjustment schemes, constructs a state-environment mapping table, calculates the deviation between the crop state index and the desired growth state, and searches for corresponding environmental parameters in the state-environment mapping table according to the deviation. The selected environmental parameters are then combined to obtain the environmental parameter change trajectory, constructing the first set of environmental parameters. By comparing crop growth states, the system can quickly identify the environmental path from the current environmental state to the target crop state, avoiding the blindness of relying on fixed environmental parameters based on human experience. Based on the dynamic control strategy of the current crop state, the system improves the directionality and timing of control commands, enhancing the accuracy and effectiveness of the environmental control process.

[0106] Specifically, based on the first set of environmental parameters, a series of control commands are transformed. Using the current fused data as the initial state, these control commands are input into a pre-trained environmental dynamic model. This model includes, but is not limited to, a deep neural network model. The model simulates the changes in environmental parameters during the execution of virtual commands, predicts the evolution of crop state indices, and selects the control commands corresponding to the scheme that leads to the fastest attainment of the target crop state, thus obtaining the first set of control commands. By simulating the effects of different schemes and crop states, the optimal strategy can be proactively selected, improving the overall performance of the control system and ensuring that regulatory actions are always guided by the principle of improving crop state, thereby enhancing the regulatory effect.

[0107] Furthermore, the deviation between environmental parameters and preset target values ​​in the fused data is analyzed, the priority of each control command in the first control command set is calculated, and the control commands are merged and optimized to obtain the second control command set, including:

[0108] S601. Analyze the degree of deviation between environmental parameters and preset target values ​​in the fused data, analyze the deviation spread range according to the degree of deviation, analyze the deviation threat situation, and calculate the priority of each control command in the first control command set;

[0109] S602. Identify control commands with overlapping resources and coupled effects in descending order of priority, merge and optimize the control commands to obtain a second set of control commands.

[0110] In this embodiment, for each environmental parameter in the fused data, the degree of deviation between the environmental parameter in the fused data and the preset target value is analyzed. The deviation diffusion range is analyzed according to the degree of deviation, the deviation threat situation is analyzed, and the priority of each control command in the first control command set is calculated. By analyzing the spatial diffusion of the deviation and the impact threat situation, the most critical and harmful threats can be quickly screened out, resources can be prioritized for regulation, and the corresponding priority is calculated to provide a sequence reference for the regulation process. This ensures that when control resources are limited or there are many concurrent demands, the decision most conducive to global stability can be made, thereby improving the response speed and regulation effect of the regulation system.

[0111] Specifically, control commands with overlapping resources and coupled effects are identified in descending order of priority. These commands are then merged and optimized to obtain a second set of control commands. By identifying and optimizing the resource conflicts and effect coupling among control commands, ineffective oscillations caused by the actuator receiving conflicting commands can be avoided, improving the overall execution efficiency and stability of the control process. Command merging allows the same regulatory effect to be achieved with fewer execution actions and shorter running time, reducing system energy consumption and improving the efficiency and effectiveness of the regulatory process.

[0112] Furthermore, the deviation between environmental parameters and preset target values ​​in the fused data is analyzed, the deviation propagation range is analyzed according to the degree of deviation, the deviation threat situation is analyzed, and the priority of each control command in the first control command set is calculated, including:

[0113] S701. Analyze the degree of deviation between environmental parameters and preset target values ​​in the fused data, calculate the deviation value, analyze the corresponding deviation impact intensity and diffusion range, and construct a deviation descriptor;

[0114] S702. Based on the deviation descriptor, analyze the deviation threat situation and construct a deviation threat vector;

[0115] S703. Based on the deviation threat vector, the threat index and urgency index are analyzed using a preset threat analysis model to obtain the deviation analysis results;

[0116] S704. Based on the deviation analysis results, calculate the priority of each control instruction in the first control instruction set.

[0117] In this embodiment, for each environmental parameter in the fused data, the degree of deviation between the environmental parameter and the preset target value in the corresponding spatial grid is analyzed, and the deviation value is calculated. The deviation influence intensity is obtained by multiplying the deviation value as an input vector by the corresponding influence weight in the coupling matrix, reflecting the magnitude of the influence of the deviation on its own parameter in other locations and other related parameters. By analyzing the decay characteristics of the weights related to the deviation source in the spatial influence tensor with distance, and combining the homogeneity of the environmental medium, the effective transmission spatial boundary is calculated. The deviation diffusion range is determined by combining the spatial boundary, reflecting the influence range of the deviation. The calculated deviation value, deviation influence intensity, and deviation diffusion range are combined in sequence to construct a deviation descriptor. By simulating the impact of deviation, the chain effects caused by local deviations can be analyzed, and the local problems and system problems can be accurately distinguished, providing accurate data support for determining priorities.

[0118] Specifically, based on deviation descriptors, a random forest model pre-trained with a large number of historical deviation descriptors is used to analyze deviation threats, including but not limited to heat stress, water stress, and photosynthetic inhibition. The analysis results are quantified and combined to construct a deviation threat vector. By analyzing deviation threats and constructing a deviation threat vector, the deviation situation can be quantified, ensuring that subsequent prioritization is crop-centric, prioritizing the treatment of environmental deviations that are most sensitive to and most destructive to the crop's current growth, thus improving the biological rationality and protective efficacy of regulatory decisions.

[0119] Specifically, based on the deviation threat vector, a pre-set threat analysis model is used to analyze the threat index and urgency. The threat analysis model includes, but is not limited to, a neural network model pre-trained using a large number of historical deviation threat vectors. The model normalizes the threat scores of each dimension in the deviation threat vector, calculates the threat index by weighted summation, and reflects the overall severity of the expected damage to crops caused by the deviation. The time dimension information corresponding to the threat is analyzed, and the urgency index is calculated. The deviation analysis results are obtained by combining the threat index and the urgency index. By analyzing the threat situation, accurate data support can be provided for the priority calculation process, ensuring that the system can quickly handle serious problems when resources are scarce, and improving the overall decision-making quality and efficiency of the system in complex multi-task scenarios.

[0120] Specifically, based on the deviation analysis results, the priority of each control command in the first control command set is calculated. For each control command in the first control command set, the deviation of the corresponding environmental parameters is determined. The threat index and urgency index in the deviation analysis results are weighted and normalized, and the corresponding priority is set according to the normalized result. By combining the threat analysis with the corresponding priority calculation, the most critical threat situations can be dealt with first, improving the security and timeliness of the generated control command sequence.

[0121] Furthermore, according to priority from high to low, control instructions with overlapping resources and coupled effects are identified, and these control instructions are merged and optimized to obtain a second set of control instructions, including:

[0122] S801. Analyze the spatial influence domain of each control instruction in the first set of control instructions, identify the control instructions with overlapping resources and coupled effects in the spatial influence domain, and obtain the set of conflicting instructions.

[0123] S802. Merge and optimize instructions in the conflict instruction set whose priority difference is less than a preset priority threshold in descending order of priority to obtain an optimized instruction set.

[0124] S803. Based on the optimized instruction set, the corresponding control instruction in the first control instruction set is replaced to obtain the second control instruction set.

[0125] In this embodiment, the spatial influence domain of each control instruction in the first set of control instructions is analyzed. Based on the actuator position, action parameter type, and adjustment amount corresponding to the control instruction, combined with the coupling matrix and spatial influence tensor, the changes that will occur in various positions and parameters in the environment after the control instruction is executed are analyzed to filter out the spatial influence domains. By checking whether the spatial influence domains of different control instructions require the same physical actuator to perform logically mutually exclusive actions or actions exceeding its physical capabilities within the same time period, resource overlap is identified. By comparing the spatial influence domains of different instructions, the effect coupling is identified by analyzing whether the superposition of overlapping areas will produce undesirable effects. All instruction pairs with resource competition or negative effect coupling are filtered out to obtain a set of conflicting instructions. By extrapolating and comparing the effects of the instructions, instruction combinations that will lead to equipment conflicts and energy waste during execution can be quickly filtered out, avoiding the loss problems caused by uncoordinated parallel execution between instructions. By identifying conflicting instructions in advance, clear goals and basis are provided for the optimization and adjustment process, improving the overall execution efficiency and stability of the system.

[0126] Specifically, the instructions in the conflict instruction set are arranged in descending order of priority. Starting with the highest priority instruction, the system searches downwards for conflicting instructions whose priority difference is within a preset priority threshold. The priority threshold is set according to the environmental control precision requirements. For the selected instructions, the control target values ​​of the instructions are weighted and summed to output a new instruction. Through merging and optimization, an optimized instruction set is obtained. By merging and optimizing instructions, control instructions can be integrated into fewer, more refined, and more coordinated control operations. This can reduce the frequency and amplitude of actuator actions, reduce equipment mechanical wear and energy consumption, improve the execution performance of execution resources, and improve the efficiency and effectiveness of the control process.

[0127] Specifically, based on the optimized instruction set, the original instructions identified as conflicting and already involved in the optimization process are located, and the corresponding original instructions in the first control instruction set are replaced to obtain the second control instruction set. Through instruction replacement and updates, the coordination between each instruction in the second control instruction set and other instructions is improved, enhancing the stability of the environmental control process, ensuring the accurate execution of each control instruction, and improving the effectiveness of environmental control.

[0128] Furthermore, environmental parameters are adjusted and corrected according to the second set of control instructions to achieve closed-loop control of the environment, including:

[0129] S901. The environmental parameters of the second control instruction set are analyzed to obtain the corresponding drive instructions;

[0130] S902. According to the driving command, the environmental parameters are corrected to perform closed-loop control of the environment.

[0131] In this embodiment, the environmental parameters of the second set of control instructions are parsed, each instruction is broken down, and core elements are extracted, including but not limited to the target object, the control target value, and optional action constraints. Combined with the characteristic parameters of the actuator, the control target value is converted into specific driving parameters. Information such as device address, protocol instructions, driving parameters, and execution timing are encapsulated and constructed into driving instructions. Through instruction parsing and conversion, execution deviations or lags caused by human configuration errors or software parsing delays can be eliminated, ensuring that multiple instructions can be sent to different actuators synchronously or in precise timing, ensuring the accurate execution of control instructions, and improving the execution accuracy and reliability of the system.

[0132] Specifically, according to the drive commands, they are sent to the connected actuators in real time. After receiving the commands, the actuators generate corresponding physical actions, including but not limited to adjusting power, changing opening degree, adjusting angle, influencing the environment, and correcting environmental parameters. Simultaneously, the sensor network continuously collects the latest environmental status data, performing a new round of data processing, status analysis, decision optimization, and command generation loops. This loop runs at high speed, clocked by the FPGA, enabling the system to evaluate the effectiveness of its control actions in near real-time and continuously and dynamically adjust the control strategy based on the evaluation results. By rapidly executing drive commands, environmental disturbances can be suppressed in a timely manner, stabilizing parameters within the target range. The real-time closed-loop mechanism can handle dynamically changing environments and the needs of growing crops, ensuring that regardless of changes in external conditions, the system can intelligently maintain or approach the optimal environmental state through rapid adjustment loops. This provides technical support for achieving stable, efficient, and unattended precision production and management.

[0133] like Figure 4 As shown, an FPGA-based multi-parameter environmental closed-loop control system is used to implement an FPGA-based multi-parameter environmental closed-loop control method, including:

[0134] The data fusion module uses an FPGA to unify the monitoring data collected by the sensor network set up in the environment into the same spatial coordinate system, analyzes the coupling relationship between different parameter data, constructs a coupling matrix, couples multiple parameter data, and obtains fused data.

[0135] The instruction generation module, based on fused data, analyzes the environmental crop status through a preset crop status recognition model, simulates the impact of different parameter adjustment processes on the crop status, and generates the first set of control instructions.

[0136] The instruction optimization module analyzes the deviation between environmental parameters and preset target values ​​in the fused data, calculates the priority of each control instruction in the first control instruction set, and merges and optimizes the control instructions to obtain the second control instruction set.

[0137] The environmental control module adjusts and corrects environmental parameters according to the second set of control instructions to achieve closed-loop control of the environment.

[0138] In this embodiment, the data fusion module uses FPGA to perform hardware-accelerated spatiotemporal alignment, coupling relationship analysis, and dynamic fusion of multi-source, heterogeneous, and dispersed monitoring data collected by the sensor network. This results in fused data that eliminates time delays and cross-interference, accurately reflecting the global spatiotemporal state of the environment. This provides unified and accurate environmental data for environmental analysis, establishes a quantitative model of the mutual influence between environmental parameters, and improves the accuracy and effectiveness of environmental regulation. The instruction generation module uses the fused data as input, evaluates the crop physiological state in real time through a crop state recognition model, and simulates the effects of different regulation strategies based on the state target. It generates a preliminary set of control instructions. By simulating and analyzing crop growth to generate control instructions, the biological target orientation and predictability of the instructions can be improved. This allows for proactive screening of the optimal regulation path for crop growth, enhancing the accuracy and biological effectiveness of control.

[0139] Specifically, the instruction optimization module performs threat assessment on the initially generated instructions, analyzes dynamic priorities, identifies potential conflicts between instructions in terms of device resource consumption and control effects, and obtains a final set of instructions that are internally coordinated and highly efficient through merging and optimization. By anticipating conflicts and integrating resources, the orderliness of control actions can be improved, ineffective actions and energy consumption can be avoided, and the effectiveness of the overall system execution efficiency can be improved while ensuring key control objectives. The environmental control module parses the optimized control instructions into drive signals that can be executed by specific actuators, and controls the actuators to complete environmental correction actions. At the same time, it triggers a new round of sensor data acquisition, closing the entire control loop and forming an adaptive closed loop that can continuously perceive and adjust in real time, ensuring the accurate execution of control strategies and the long-term stable and autonomous operation of the system.

[0140] The above description is merely a preferred embodiment of this application. The scope of protection of this application is not limited to the above embodiments. All technical solutions falling within the scope of this application's concept are within the scope of protection of this application. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of this application should also be considered within the scope of protection of this application.

Claims

1. A multi-parameter environmental closed-loop control method based on FPGA, characterized in that, include: Based on the monitoring data collected by the sensor network set up in the environment, the monitoring data is unified into the same spatial coordinate system through FPGA, the coupling relationship between different parameter data is analyzed, a coupling matrix is ​​constructed, and multi-parameter data is coupled to obtain fused data; Based on fused data, the environmental crop status is analyzed through a preset crop status identification model, the impact of different parameter adjustment processes on crop status is simulated, and a first set of control instructions is generated. The deviation between environmental parameters and preset target values ​​in the fused data is analyzed, the priority of each control command in the first control command set is calculated, and the control commands are merged and optimized to obtain the second control command set. The environmental parameters are adjusted and corrected according to the second set of control instructions in order to achieve closed-loop control of the environment.

2. The FPGA-based multi-parameter environmental closed-loop control method according to claim 1, characterized in that, The monitoring data collected by the sensor network set up in the environment is unified into the same spatial coordinate system through FPGA, the coupling relationship between different parameter data is analyzed, a coupling matrix is ​​constructed, and multi-parameter data is coupled to obtain fused data, including: Based on the monitoring data collected by the sensor network set up in the environment, the monitoring data is uniformly mapped to the same spatial coordinate system through FPGA to obtain the first data point cloud set; Based on the first data point cloud set, the correlation between data with different parameters is analyzed, the parameter coupling strength is calculated, the impact of data changes on the first data point cloud set is analyzed, and the spatial influence tensor is calculated. By combining the parameter coupling strength and the spatial influence tensor, the time delay between different parameter data is analyzed, a coupling matrix is ​​constructed, and multi-parameter data is coupled to obtain fused data.

3. The FPGA-based multi-parameter environmental closed-loop control method according to claim 2, characterized in that, The process of analyzing the correlation between different parameter data based on the first data point cloud set, calculating the parameter coupling strength, analyzing the impact of data changes on the first data point cloud set, and calculating the spatial influence tensor includes: According to the preset spatial grid, analyze the correlation between different parameter data in each spatial grid of the first data point cloud set, filter out data with correlation greater than the preset correlation threshold, establish connection edges, and obtain the first connection edge set; Based on the first set of connecting edges, simulate the impact of parameter changes within a spatial grid on the parameters of adjacent spatial grids, calculate the parameter coupling strength between spatial grids, and construct a grid coupling matrix; By combining the spatial distribution and parameter coupling strength of the first data point cloud set, we analyze the propagation of parameter changes between spatial grids, calculate the corresponding propagation distance and medium attenuation coefficient, and construct a spatial propagation weight matrix. Based on the grid coupling matrix and the spatial propagation weight matrix, the coupling transformation and total spatial propagation effect of the parameters within each spatial grid are calculated, and the spatial influence tensor is encoded.

4. The FPGA-based multi-parameter environmental closed-loop control method according to claim 3, characterized in that, The method combines the parameter coupling strength and spatial influence tensor to analyze the time delay between different parameter data, constructs a coupling matrix, couples multi-parameter data, and obtains fused data, including: Based on the pre-acquired historical control command data and the first data point cloud set, analyze the time difference distribution between control commands and parameter change responses, and construct the first time delay set; Based on the parameter coupling strength, the propagation speed between corresponding spatial grids in the grid coupling matrix and the spatial propagation weight matrix is ​​analyzed, the first time delay set is optimized to obtain the second time delay set, and the time delay tensor is extracted. The spatial influence tensor and the time delay tensor are correlated and calculated to construct a coupling matrix; The first data point cloud set is time-delayed and coupled according to the coupling matrix to obtain fused data.

5. The FPGA-based multi-parameter environmental closed-loop control method according to claim 1, characterized in that, The process involves analyzing the environmental crop state based on fused data using a pre-set crop state identification model, simulating the impact of different parameter adjustment processes on the crop state, and generating a first set of control commands, including: Based on fused data, the environmental crop status is analyzed through a pre-set crop status identification model, and the corresponding crop status index is calculated. By combining the crop state index and the preset expected growth state, the trajectory of environmental parameter changes corresponding to the state change process is analyzed, and a first set of environmental parameters is constructed. Based on the first set of environmental parameters, the impact of different parameter adjustment processes on crop status is simulated to generate a first set of control commands.

6. The FPGA-based multi-parameter environmental closed-loop control method according to claim 1, characterized in that, The analysis quantifies the deviation between environmental parameters and preset target values ​​in the fused data, calculates the priority of each control command in the first control command set, and merges and optimizes the control commands to obtain a second control command set, including: Analyze the degree of deviation between environmental parameters and preset target values ​​in the fused data, analyze the deviation spread range according to the degree of deviation, analyze the deviation threat situation, and calculate the priority of each control command in the first control command set; Based on priority from high to low, control commands with overlapping resources and coupled effects are identified, and the control commands are merged and optimized to obtain a second set of control commands.

7. The FPGA-based multi-parameter environmental closed-loop control method according to claim 6, characterized in that, The analysis assesses the degree of deviation between environmental parameters and preset target values ​​in the fused data, analyzes the deviation spread range based on the degree of deviation, analyzes the deviation threat situation, and calculates the priority of each control command in the first control command set, including: Analyze the degree of deviation between environmental parameters and preset target values ​​in the fused data, calculate the deviation value, analyze the corresponding deviation impact intensity and diffusion range, and construct a deviation descriptor; Based on the deviation descriptor, analyze the deviation threat situation and construct a deviation threat vector; Based on the deviation threat vector, the threat index and urgency index are analyzed using a preset threat analysis model to obtain the deviation analysis results; Based on the deviation analysis results, the priority of each control instruction in the first control instruction set is calculated.

8. The FPGA-based multi-parameter environmental closed-loop control method according to claim 7, characterized in that, The control commands that overlap in resources and have coupled effects are identified in descending order of priority. These control commands are then merged and optimized to obtain a second set of control commands, including: Analyze the spatial influence domain of each control instruction in the first set of control instructions, identify control instructions with overlapping resources and coupled effects in the spatial influence domain, and obtain the set of conflicting instructions; Instructions with priority differences less than a preset priority threshold in the conflict instruction set are merged and optimized in descending order of priority to obtain an optimized instruction set. Based on the optimized instruction set, the corresponding control instructions in the first control instruction set are replaced to obtain the second control instruction set.

9. The FPGA-based multi-parameter environmental closed-loop control method according to claim 1, characterized in that, The step of adjusting environmental parameters according to the second set of control instructions and correcting the environmental parameters to achieve closed-loop control of the environment includes: The environmental parameters of the second set of control commands are analyzed to obtain the corresponding drive commands. According to the driving instructions, the environmental parameters are corrected to achieve closed-loop control of the environment.

10. A multi-parameter environmental closed-loop control system based on FPGA, characterized in that, The method for implementing the FPGA-based multi-parameter environmental closed-loop control method as described in any one of claims 1 to 9 includes: The data fusion module uses an FPGA to unify the monitoring data collected by the sensor network set up in the environment into the same spatial coordinate system, analyzes the coupling relationship between different parameter data, constructs a coupling matrix, couples multiple parameter data, and obtains fused data. The instruction generation module, based on fused data, analyzes the environmental crop status through a preset crop status recognition model, simulates the impact of different parameter adjustment processes on the crop status, and generates the first set of control instructions. The instruction optimization module analyzes the deviation between environmental parameters and preset target values ​​in the fused data, calculates the priority of each control instruction in the first control instruction set, and merges and optimizes the control instructions to obtain the second control instruction set. The environmental control module adjusts and corrects environmental parameters according to the second set of control instructions to achieve closed-loop control of the environment.

Images