Fumigation room temperature and humidity control method and system based on data analysis

Abstract

The invention relates to the technical field of environment control, in particular to a fumigation room temperature and humidity control method and system based on data analysis, and the method comprises the following steps: based on temperature and humidity data collected by a multi-point sensor in a fumigation room, defining fuzzy variables of the temperature and humidity data, mapping the temperature and humidity data to a corresponding fuzzy set according to the fuzzy variables, and obtaining a humiture data fuzzy processing result. According to the invention, by introducing the fuzzy logic, mapping the collected temperature and humidity data to the fuzzy set, calculating the membership value, and performing flexible processing on the temperature and humidity data in the complex environment, compared with direct parameter comparison, the adaptability to the nonlinear dynamic environment is improved. In combination with the dynamic control threshold adjustment process, the current operation condition and the equipment capability range are comprehensively considered, and error accumulation caused by a traditional fixed threshold is avoided. And a gradient calculation mode of multi-point data is combined with distributed processing, so that the operation parameter optimization of the equipment has better real-time performance and regional pertinence.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental control, and in particular to a fumigation warehouse temperature and humidity control method and system based on data analysis. Background Art

[0002] The field of environmental control technology includes the regulation and control of parameters such as temperature, humidity, air pressure and air flow in the environment. Its core content is to accurately control the physical parameters in a specific environment through sensor monitoring, data collection and feedback adjustment mechanism to meet the needs of specific purposes.

[0003] Among them, the temperature and humidity control method of the fumigation warehouse refers to the precise control of the temperature and humidity parameters in the warehouse by designing a dedicated temperature and humidity monitoring and adjustment system for the specific environment of the fumigation warehouse. It covers the layout and collection method of temperature and humidity sensors, the data analysis and processing flow, and the linkage operation of temperature and humidity adjustment equipment. Specifically, the sensor equipment is used to monitor the environmental parameters in the warehouse in real time, combined with the analysis system to make data judgments, and then the humidification and dehumidification device and the heating and cooling device are used to complete the temperature and humidity adjustment.

[0004] The traditional fumigation warehouse temperature and humidity control method is based on fixed threshold judgment, which is difficult to cope with dynamic changes in the environment. It may cause the equipment to start and stop frequently or respond insufficiently, affecting the operating efficiency. The fixed threshold design lacks comprehensive consideration of the equipment capabilities and the dynamic characteristics of the environment, and the threshold may fail due to changes in conditions. In addition, relying solely on single-point monitoring of temperature and humidity data cannot reflect the spatial distribution characteristics of the environment in the fumigation warehouse, resulting in reduced control accuracy. In traditional methods, the adjustment process of the equipment usually lacks regional linkage, and it is impossible to dynamically adjust the operating parameters according to the needs of different regions, which can easily lead to overshoot or undershoot problems in some areas. At the same time, the ability to predict environmental fluctuations is limited, and it is difficult to respond to sudden changes in a timely manner, which may affect the overall operational stability of the fumigation warehouse. These shortcomings limit the adaptability of traditional methods in complex environments and reduce the reliability and energy saving of the control system. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a fumigation warehouse temperature and humidity control method and system based on data analysis.

[0006] In order to achieve the above object, the present invention adopts the following technical scheme: a fumigation warehouse temperature and humidity control method based on data analysis, comprising the following steps:

[0007] S1: Based on the temperature and humidity data collected by multi-point sensors in the fumigation warehouse, the fuzzy variables of the temperature and humidity data are defined, and the temperature and humidity data are mapped to the corresponding fuzzy sets according to the fuzzy variables to obtain the fuzzy processing results of the temperature and humidity data;

[0008] S2: Based on the affiliation status of each temperature and humidity data in the fuzzy processing result of the temperature and humidity data, combined with the affiliation status and the current operating conditions of the fumigation warehouse, dynamically adjust the control threshold of the fumigation warehouse, integrate the adjusted control threshold, and output a dynamic control threshold list;

[0009] S3: Based on the control threshold range in the dynamic control threshold list, the control threshold range is matched with the real-time temperature and humidity data of multiple areas in the fumigation warehouse, the environmental state of each area of ​​the fumigation warehouse and the operating parameters of the regulating equipment are extracted, the operating parameters are distributed to multiple computing nodes, and the gradient calculation of the regulating equipment operating parameters is performed. The sensitivity of the regulating equipment operating parameters to environmental changes is determined according to the gradient of the equipment operating parameters, the output of the node is adjusted, and the equipment operation optimization parameters are generated;

[0010] S4: Perform frequency domain decomposition on all temperature and humidity data of the fumigation warehouse and identify significant frequency characteristics, predict future temperature and humidity change trends based on the significant frequency characteristics, match the future temperature and humidity change trends with the current equipment operation optimization parameters, adjust the fumigation warehouse equipment parameters based on the matching information, and generate the fumigation warehouse temperature and humidity control results.

[0011] As a further solution of the present invention, the steps for obtaining the fuzzy processing result of the temperature and humidity data are specifically as follows:

[0012] S111: Based on the temperature and humidity data collected by the multi-point sensors in the fumigation warehouse, the collected data is cleaned and normalized, the processed data is mapped to the fuzzy set according to the preset temperature and humidity classification standard, and the data is classified and sorted to generate the temperature and humidity data classification result;

[0013] S112: Based on the classification result of the temperature and humidity data, the formula is used:

[0014]

[0015] Calculate the membership value μ(x) of the target temperature and humidity value x in the fuzzy set. The range of the membership value is [0,1]. Integrate the membership values ​​to obtain the fuzzy processing results of the temperature and humidity data.

[0016] Among them, x represents the normalized temperature and humidity data point, c represents the center point of the fuzzy membership function, and σ represents the width of the membership function, which defines the extension range of the fuzzy set.

[0017] As a further solution of the present invention, the step of obtaining the dynamic control threshold list is specifically:

[0018] S211: Based on the affiliation status of each temperature and humidity data in the fuzzy processing result of the temperature and humidity data, combined with the affiliation status and the current operating conditions of the fumigation warehouse, including the humidification capacity range, dehumidification capacity range, heating power and equipment response time, dynamically adjust the control threshold of the fumigation warehouse, the control threshold is the threshold range for starting and stopping humidification, dehumidification and heating, and obtain the control threshold adjustment result of the fumigation warehouse;

[0019] S212: Based on the control threshold adjustment result of the fumigation warehouse, control parameters of each equipment type and regional distribution of the fumigation warehouse are classified and integrated, including humidification equipment, dehumidification equipment and heating equipment, to obtain a dynamic control threshold list.

[0020] As a further solution of the present invention, the step of calculating the gradient of the operating parameters of the regulating device is specifically:

[0021] S311: Based on the control threshold range in the dynamic control threshold list, the real-time temperature and humidity data of multiple areas of the fumigation warehouse are processed in segments according to the time window, the environmental state of each area and the operating parameters of the regulating equipment are extracted and matched, and the equipment operating parameters are matched with the regional environmental state and then allocated to multiple computing nodes to obtain the regional environmental state and equipment operating parameter allocation results;

[0022] S312: Based on the regional environmental status and the equipment operation parameter allocation result, the formula is used:

[0023]

[0024] Calculate the updated equipment operating parameters, which represents the optimized value θ of the equipment operating parameters in the t+1th iteration (t+1) ;

[0025] Among them, J(θ) is the objective function in the fumigation warehouse control, and its complete form is:

[0026] θ (t) is the value of the corresponding device operating parameter θ in the tth iteration, θ (t+1) By the current gradient value and the learning rate η is updated from the current iteration θ (t) Update to the next iteration θ (t+1) , η is the learning rate, which indicates the step size of parameter update, is the gradient of the objective function J(θ) in the fumigation warehouse control with respect to the operating parameter θ of the corresponding equipment, indicating the sensitivity of the current parameter to the objective function in the fumigation warehouse control, m is the number of samples, j is the sample number currently being calculated, and y j represents the expected device output parameter value, f(x j,θ) is the device output value, x j is the input environmental state data point.

[0027] As a further solution of the present invention, the step of obtaining the equipment operation optimization parameters is specifically as follows:

[0028] S321: Based on the updated device operating parameters and the current environmental data change range, the difference value of the operating parameters is compared with the preset sensitivity range, the sensitivity state of the device operating parameters is determined, and the devices are graded and adjusted in combination with the global device priority to generate the device sensitivity state and priority information to the environmental change;

[0029] S322: Based on the sensitivity status and priority information of the device to environmental changes, the adjustment range of each priority device is classified and sorted according to the response time and power range to generate device operation optimization parameters.

[0030] As a further solution of the present invention, the step of predicting the future temperature and humidity change trend is specifically:

[0031] S411: Collect all temperature and humidity data of the fumigation warehouse, process the monitoring data in time series and perform frequency domain decomposition, identify significant frequency characteristics in environmental fluctuations and prioritize them by analyzing the frequency components and amplitude distribution after decomposition, and generate a frequency characteristic distribution table with significant impact;

[0032] S412: Based on the frequency characteristic distribution table with significant influence, the formula is adopted:

[0033] y(t′)=Asin(2πft′+φ)+B;

[0034] Calculate the predicted temperature and humidity value y(t′) at time t′ to obtain the future temperature and humidity change trend data;

[0035] Among them, A represents the maximum fluctuation amplitude of temperature and humidity changes, sin(2πft′+φ) is the sine function part, which describes the periodic changes of temperature and humidity over time, 2π represents the radian of a complete cycle, f is the significant frequency, which represents the dominant frequency characteristics in environmental fluctuations and is inversely proportional to the period T, that is, f=1 / T, t′ is the time variable, which represents the time point of the predicted target, φ represents the initial offset of the temperature and humidity fluctuation, and B represents the average value of the temperature and humidity data.

[0036] As a further solution of the present invention, the steps for obtaining the temperature and humidity control results of the fumigation warehouse are specifically as follows:

[0037] S421: Based on the future temperature and humidity change trend data, the predicted temperature and humidity values ​​are compared with the operation optimization parameters of the current equipment one by one, and whether they exceed the operation power adjustment range and response time range is analyzed, and the deviation time point and amplitude difference are recorded to generate the equipment operation matching result;

[0038] S422: Based on the equipment operation matching result, the power output range and response time parameters of the equipment that do not meet the fluctuation demand are adjusted, the adjusted equipment operation parameters are summarized and the adjustment range is marked, and the temperature and humidity control result of the fumigation warehouse is generated.

[0039] A fumigation warehouse temperature and humidity control system based on data analysis, the fumigation warehouse temperature and humidity control system based on data analysis is used to execute the above-mentioned fumigation warehouse temperature and humidity control method based on data analysis, the system comprises:

[0040] The fuzzy data processing module defines temperature and humidity fuzzy variables based on the temperature and humidity data collected by multi-point sensors in the fumigation warehouse, uses membership functions to map the temperature and humidity data to fuzzy sets, and generates fuzzy processing results for the temperature and humidity data;

[0041] The dynamic threshold adjustment module dynamically adjusts the control threshold of the fumigation warehouse based on the fuzzy processing result of the temperature and humidity data, combined with the subordinate status and the current operating conditions of the fumigation warehouse, and outputs a dynamic control threshold list after integrating the control thresholds;

[0042] The equipment operation optimization module matches the control threshold range with the real-time temperature and humidity data of each area based on the dynamic control threshold list, extracts the regional environmental status and adjusts the equipment operation parameters, performs gradient optimization on the operation parameters through the calculation nodes, and adjusts the node output to generate the equipment operation optimization parameters;

[0043] The temperature and humidity trend prediction module performs frequency domain decomposition and identifies significant frequency characteristics based on all temperature and humidity data of the fumigation warehouse. It predicts future temperature and humidity changes through the trend prediction model, matches the predicted trend with the equipment operation optimization parameters, adjusts the equipment parameters according to the matching information, and generates the temperature and humidity control results of the fumigation warehouse.

[0044] Compared with the prior art, the advantages and positive effects of the present invention are:

[0045] In the present invention, by introducing fuzzy logic, the collected temperature and humidity data are mapped to fuzzy sets, and the membership value is calculated, so that the temperature and humidity data in complex environments are flexibly processed, and the adaptability to nonlinear dynamic environments is improved compared with direct parameter comparison. Combined with the dynamic control threshold adjustment process, the current operating conditions and the equipment capacity range are comprehensively considered to avoid the error accumulation caused by the traditional fixed threshold. The gradient calculation method of multi-point data is combined with distributed processing to make the equipment operation parameter optimization more real-time and regionally targeted, and reduce the time lag of global adjustment. By identifying significant frequency characteristics through frequency domain decomposition, combined with a periodic prediction model, the future temperature and humidity change trend is analyzed in advance, and the equipment operation parameters are accurately adjusted, which enhances the system's predictability and responsiveness to environmental fluctuations. The optimized equipment operating parameters are matched with the future temperature and humidity trends, and the equipment parameters that do not meet the needs are dynamically adjusted to reduce the operating power consumption while ensuring the stability of the environment in the warehouse. The overall processing logic combines dynamic adjustment, real-time optimization and forward-looking prediction to make the temperature and humidity control of the fumigation warehouse more intelligent and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0047] Figure 2 A flow chart of obtaining the fuzzy processing result of temperature and humidity data in the present invention;

[0048] Figure 3 A flowchart of obtaining a dynamic control threshold list for the present invention;

[0049] Figure 4 A flow chart for calculating the gradient of the operating parameters of the regulating device according to the present invention;

[0050] Figure 5 A flow chart for obtaining equipment operation optimization parameters for the present invention;

[0051] Figure 6 A flow chart for predicting future temperature and humidity change trends of the present invention;

[0052] Figure 7 The present invention is a flow chart for obtaining the temperature and humidity control results of the fumigation warehouse. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0054] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0055] See also Figure 1 The present invention provides a technical solution: a method for controlling temperature and humidity in a fumigation warehouse based on data analysis, comprising the following steps:

[0056] S1: Based on the temperature and humidity data collected by multi-point sensors in the fumigation warehouse, the fuzzy variables of the temperature and humidity data are defined, and the temperature and humidity data are mapped to the corresponding fuzzy sets according to the fuzzy variables to obtain the fuzzy processing results of the temperature and humidity data;

[0057] S2: Based on the affiliation status of each temperature and humidity data in the fuzzy processing result of the temperature and humidity data, combined with the affiliation status and the current operating conditions of the fumigation warehouse, the control threshold of the fumigation warehouse is dynamically adjusted, the adjusted control threshold is integrated, and a dynamic control threshold list is output;

[0058] S3: Based on the control threshold range in the dynamic control threshold list, the control threshold range is matched with the real-time temperature and humidity data of multiple areas in the fumigation warehouse, the environmental status of each area of ​​the fumigation warehouse and the operating parameters of the regulating equipment are extracted, the operating parameters are distributed to multiple computing nodes, and the gradient calculation of the regulating equipment operating parameters is performed. The sensitivity of the regulating equipment operating parameters to environmental changes is determined according to the gradient of the equipment operating parameters, the output of the node is adjusted, and the equipment operation optimization parameters are generated;

[0059] S4: Perform frequency domain decomposition on all temperature and humidity data of the fumigation warehouse and identify significant frequency characteristics. Predict future temperature and humidity change trends based on significant frequency characteristics. Match future temperature and humidity change trends with current equipment operation optimization parameters. Adjust fumigation warehouse equipment parameters based on matching information to generate fumigation warehouse temperature and humidity control results.

[0060] The results of fuzzy processing of temperature and humidity data include membership value, fuzzy variable classification, and temperature and humidity fuzzy set; the dynamic control threshold list includes humidification start threshold range, dehumidification start threshold range, heating start threshold range, and response time parameters; equipment operation optimization parameters specifically include equipment start time, operating power allocation, and regional priority parameters; the results of temperature and humidity control in the fumigation warehouse specifically include future temperature and humidity change trends, adjusted equipment operating parameters, and fluctuation demand matching strategies.

[0061] See also Figure 2 , the specific steps for obtaining the fuzzy processing results of temperature and humidity data are as follows:

[0062] S111: Based on the temperature and humidity data collected by the multi-point sensors in the fumigation warehouse, the collected data is cleaned and normalized, the processed data is mapped to the fuzzy set according to the preset temperature and humidity classification standard, and the data is classified and sorted to generate the temperature and humidity data classification result;

[0063] First, check the effectiveness of the sensor. Perform time consistency detection on the data collected by each sensor, eliminate abnormal data that is lost or jumps, and convert the remaining data into a unified unit. Group the temperature and humidity data according to time granularity such as hours and days to eliminate the interference of short-term fluctuations on the overall data. The normalization process uses the maximum and minimum value normalization method to calculate the minimum and maximum values ​​of the temperature and humidity data. Map all the data to the [0,1] interval in a normalized manner. According to the temperature range (for example, low temperature is set to [0,0.3], medium temperature is [0.3,0.7], and high temperature is [0.7,1]) and humidity range (for example, low humidity is [0,0.4], medium humidity is [0.4,0.6], and high humidity is [0.6,1]), the normalized data is classified according to the variables defined by fuzzy logic. The classified temperature and humidity data are assigned to different fuzzy sets to form corresponding classification results.

[0064] S112: Based on the classification results of temperature and humidity data, the formula is used:

[0065]

[0066] Calculate the membership value μ(x) of the target temperature and humidity value x in the fuzzy set. The range of the membership value is [0,1]. Integrate the membership values ​​to obtain the fuzzy processing results of the temperature and humidity data.

[0067] Among them, x represents the temperature and humidity data point after normalization. The data is collected by sensors and directly obtained after outliers are removed and normalized. c represents the center point of the fuzzy membership function, which corresponds to the most likely value of a specific fuzzy set. It is determined by the statistical value of historical environmental data. All records within a certain time period are extracted. The period can be set to one month or one year. According to the actual environmental requirements, the data is preliminarily classified according to the predefined temperature and humidity ranges. For example, the low temperature range is [below 15°C], the medium temperature range is [15°C~30°C], and the high temperature range is [above 30°C], which are classified into different temperature groups respectively. σ represents the width of the membership function, which defines the extension range of the fuzzy set and is obtained by calculating the standard deviation of the corresponding data.

[0068] If a set of normalized temperature data collected within the low temperature range is: x1 = 0.15, x2 = 0.2, x3 = 0.18, x4 = 0.22, x5 = 0.19

[0069] Calculate the mean of the data is the cumulative value of all data points, x i represents the i-th data point, and n represents the total number of data points.

[0070] Bring in data:

[0071] Calculate the standard deviation (σ): The cumulative value of the squared deviations of all data points from the mean, calculating the squared difference item by item:

[0072] (0.15-0.188) 2 =(-0.038) 2 =0.001444;

[0073] (0.2-0.188) 2 =(0.012) 2 =0.000144;

[0074] (0.18-0.188) 2 =(-0.008) 2 =0.000064;

[0075] (0.22-0.188) 2 =(0.032) 2 =0.001024;

[0076] (0.19-0.188) 2 =(0.002) 2 =0.000004;

[0077] Sum:

[0078]

[0079] Calculate the standard deviation:

[0080] Calculate the membership value (μ(x)):

[0081] Assume the center point of the fuzzy set c = 0.2, and take the data x = 0.18 as an example:

[0082] The membership value μ(0.18)≈0.5715 indicates that the degree to which the data point x=0.18 belongs to the fuzzy set is 0.5715, which is closer to the center point of the set c=0.2.

[0083] See also Figure 3 , the specific steps for obtaining the dynamic control threshold list are:

[0084] S211: Based on the affiliation status of each temperature and humidity data in the fuzzy processing result of the temperature and humidity data, combined with the affiliation status and the current operating conditions of the fumigation warehouse, including the humidification capacity range, dehumidification capacity range, heating power and equipment response time, the control threshold of the fumigation warehouse is dynamically adjusted. The control threshold is the threshold range for starting and stopping humidification, dehumidification and heating, and the control threshold adjustment result of the fumigation warehouse is obtained;

[0085] Based on the analysis of the membership value μ(x) = 0.5715 and the fuzzy set center point c = 0.2, the membership value is compared with the current operating conditions. According to the preset range, the membership value is divided into three states: low humidity, medium humidity, and high humidity. The low humidity range is the membership value [0, 0.4], the medium humidity range is the membership value (0.4, 0.7], and the high humidity range is the membership value (0.7, 1]. According to the membership value μ(x) = 0.5715 falling in the medium humidity range, the current state is the medium humidity state, and the equipment needs to work within the preset start and stop range. Combined with the capacity range of the humidification equipment, according to the position of the membership value falling in the medium humidity range, it is judged that the current start and stop thresholds of the equipment need to be respectively located at Near the lower and upper limits of the medium humidity range. Match the operating conditions of the humidification equipment, confirm the specific control thresholds for starting and stopping, and adjust the specific operating time and power according to the operating characteristics of the equipment to ensure that the equipment's membership adjustment in the medium humidity range is dynamic and consistent with the current environmental status. Further combined with the operating conditions of the heating equipment, match the equipment response time and power range with the upper and lower limits of the current membership state, analyze the compatibility of the performance curve of the heating equipment with the membership range, and confirm that the equipment is operating within the preset response range. By dynamically adjusting the upper and lower limits of the preset threshold range, the operating characteristics and current membership status of each device are combined to generate a dynamically adjusted control threshold range for starting and stopping humidification, dehumidification, and heating.

[0086] S212: Based on the control threshold adjustment result of the fumigation warehouse, the control parameters of each equipment type and regional distribution of the fumigation warehouse are classified and integrated, including humidification equipment, dehumidification equipment and heating equipment, to obtain a dynamic control threshold list;

[0087] By classifying the control parameters of different equipment types and regional distributions, a list of operating threshold ranges for humidification equipment, dehumidification equipment and heating equipment is formed, and the equipment control parameters in all areas are grouped to correspond to their respective operating logic and control modes. For example, the start-up conditions of the humidification equipment are refined into several intervals according to the distribution of different membership values. By classifying and integrating the equipment power and time response characteristics, a unified dynamic control threshold list is output, which includes the start time, stop time, operating power range and adjustment conditions of each device.

[0088] See also Figure 4 , the steps for calculating the gradient of the operating parameters of the adjustment equipment are as follows:

[0089] S311: Based on the control threshold range in the dynamic control threshold list, the real-time temperature and humidity data of multiple areas of the fumigation warehouse are processed in segments according to the time window, the environmental state of each area and the operating parameters of the regulating equipment are extracted and matched, and the equipment operating parameters are matched with the regional environmental state and then allocated to multiple computing nodes to obtain the regional environmental state and equipment operating parameter allocation results;

[0090] First, the temperature and humidity data collected in each area are segmented and processed in chronological order. The data are grouped according to the set time window (such as every 5 minutes), and the temperature and humidity values ​​in the grouped data are extracted. According to the preset range, the temperature is divided into low temperature (below 15°C), medium temperature (15°C to 25°C), and high temperature (above 25°C). The humidity is divided into low humidity (below 40%), medium humidity (40% to 60%), and high humidity (above 60%). In each interval, the specific state of the data points is determined according to their distribution. For example, when the temperature is 22°C, it falls into the medium temperature interval, and when the humidity is 65%, it falls into the high humidity interval. Subsequently, the environmental status of each area is analyzed, and the temperature and humidity data that exceeds the threshold range is marked as abnormal data points. For example, when the threshold in the medium temperature range (15℃ to 25℃) is 23℃ and the monitoring data is 26℃, it is marked as a high temperature state; extract the equipment operation parameters, including the equipment's startup power, response delay time, and operation duration. By consulting the equipment's historical operation records and operation logs, the characteristics of each device in different states are organized into available data tables, such as the equipment's startup power of 3kW and response time of 10 seconds under low humidity conditions; associate these equipment operation parameters with the matching results of the regional environmental status to form a set of environmental status and equipment parameters for each area. Finally, according to the environmental status and equipment operation parameters in the area, the data is allocated to the corresponding computing nodes. The allocation is based on the physical location of each area and the equipment operation priority to ensure that each node receives data evenly. For example, the equipment operation parameters in the high humidity area are preferentially allocated to the nodes with shorter delay times, and all temperature and humidity data points in the area are allocated at the same time for the nodes to perform subsequent gradient calculations, completing the regional allocation of equipment operation parameters and environmental status.

[0091] S312: Based on the regional environmental status and the equipment operation parameter allocation results, the formula is used:

[0092]

[0093] Calculate the updated equipment operating parameters, which represents the optimized value θ of the equipment operating parameters in the t+1th iteration (t+1) ;

[0094] Among them, J(θ) is the objective function in the fumigation warehouse control, and its complete form is:

[0095] θ (t) It is the value of the device operating parameter θ corresponding to the current iteration. It is obtained from the initial device operating parameters through the distributed computing node and represents the specific value of the device operating parameter in the tth iteration. The initial value of the basic operating parameter is obtained by reading the actual power, response time and other characteristics of the device. (t+1) By the current gradient value And the learning rate η is updated. It is obtained from the current iteration θ (t) Update to the next iteration θ (t+1) , optimize the equipment parameters. t+1 represents the iteration time number, which is used to track the number of steps in the equipment parameter optimization process, and increases in sequence. Each time a gradient calculation is performed, the iteration number t increases by 1. η is the learning rate, which represents the step size of the parameter update and is used to control the amplitude of the parameter adjustment. By analyzing the equipment adjustment sensitivity range and combining the historical operation data, the appropriate learning rate is determined. The learning rate range is evaluated according to the sensitivity of the equipment to environmental changes. If the device responds quickly to changes in the input environment, the learning rate can be set larger. If the device responds slowly or the adjustment is complex, the learning rate needs to be set smaller to ensure the stability of the parameter update. By analyzing the historical operation data of the equipment, the amplitude and range of parameter adjustment are evaluated. According to the adjustment step size of the equipment in similar environments in the past, the average step size is evaluated as the preliminary setting value of the learning rate. The effect of different learning rates on parameter update is tested in the experiment, and a learning rate value that can steadily reduce the objective function in the fumigation warehouse control is selected through multiple experiments. It is the gradient of the objective function J(θ) in fumigation control with respect to the operating parameter θ of the corresponding equipment, indicating the sensitivity of the current parameter to the objective function in fumigation control. The larger the value, the more significant the error effect of the current parameter on the environmental adjustment. It is derived by calculating the partial derivative of the objective function in fumigation control, combining the equipment's operating curve and actual output data. Indicates that the square error of each sample in the objective function of the fumigation warehouse control is summed up to calculate the overall objective function value of the fumigation warehouse control. m is the number of samples, which indicates the total number of samples of equipment operating parameters and environmental data in all computing nodes. j indicates the sample number currently being calculated, from 1 to m. y j Indicates the expected output parameter value of the device. It is obtained through laboratory verification or device design standards. The optimal output power of the device under specific temperature and humidity conditions is determined by the experiment as the target value. j ,θ) is the actual output value of the device, which is determined by the device operating parameters and input environment status. It is calculated based on the device performance curve and real-time environmental input data. It represents the relationship between the actual effect of the current device operation and the input conditions. j It is the input environmental status data point, which represents the real-time temperature and humidity data of a certain area and corresponds to number j.

[0096] If the current operating parameters of the device are θ (t) =50, input environmental data point x j =0.6, target output y j =1.0, the actual output value f(x j ,θ)=θ (t) ·x j =50·0.6=30.0, learning rate η=0.01.

[0097] Calculate the objective function J(θ) in fumigation warehouse control:

[0098]

[0099] Computing Gradients

[0100] Assume that the partial derivative of the device response function with respect to the parameter is but:

[0101] Update parameters:

[0102] Submit data to calculate: θ (t+1) =50-0.01·17.4=50-0.174=49.826;

[0103] The final calculation result θ (t+1) =49.826 shows that after optimization by the gradient descent algorithm, the equipment operating parameters are further adjusted to the global optimal value, and the objective function value in the fumigation warehouse control gradually decreases.

[0104] See also Figure 5 , the specific steps for obtaining the equipment operation optimization parameters are as follows:

[0105] S321: Based on the updated device operating parameters and the current environmental data change range, the difference value of the operating parameters is compared with the preset sensitivity range to determine the sensitivity state of the device operating parameters, and the devices are graded and adjusted in combination with the global device priority to generate the device sensitivity state and priority information to environmental changes;

[0106] First, update the value θ according to the equipment operating parameters calculated above. (t+1)=49.826, compare the benchmark parameter values ​​in the equipment operation history data, assume that the benchmark value is 50, calculate the difference between the current parameter and the benchmark value, the difference is 50-49.826=0.174, compare this difference with the equipment parameter change range, set the high sensitivity range to be greater than 0.3, the medium sensitivity range to be between 0.1 and 0.3, and the low sensitivity range to be less than 0.1. The difference of 0.174 determines that the sensitivity of the current equipment operation parameters to environmental changes is in the medium sensitivity range. Further combined with the historical response records of the equipment, the amplitude of environmental changes is matched with the amplitude of equipment operation parameter adjustment. For example, the current temperature rises from 22℃ to 24℃, and the humidity rises from 50% to 55%. When the temperature change is greater than 3℃ or the humidity change is greater than 10%, it is a high environmental change range, 1℃ to 3℃ or 5% to 10% is a medium environmental change range, and less than 1℃ or 5% is a low environmental change range. The current environmental change is classified as a medium environmental change range. Combined with the sensitivity status of the equipment operating parameters, it is determined that the equipment is a medium sensitive device within the medium environmental change range. The adjustment strategy is to keep the power output range within the current value within 5%. Finally, combined with the operation priority information of the global fumigation warehouse equipment, the adjustment strategy of high-priority equipment will give priority to responding to significant changes in temperature and humidity, and low-priority equipment will be adjusted later. By analyzing the matching degree between the equipment's operating parameters and environmental changes, the sensitivity status of the equipment to environmental changes is further clarified, the matching relationship between the equipment priority and the adjustment strategy is recorded, and the mapping relationship between environmental changes and equipment sensitivity is completed.

[0107] S322: Based on the sensitivity status and priority information of the device to environmental changes, the adjustment range of each priority device is classified and sorted according to the response time and power range, and the device operation optimization parameters are generated;

[0108] Generate equipment operation optimization parameters based on the adjustment results, summarize the sensitivity results of equipment operation parameters and environmental changes by priority, first set a dynamic adjustment range for high-priority equipment, for example, the power adjustment range of high-priority equipment is set to the current power fluctuation of no more than 3%, and the response time adjustment range is the current response time fluctuation of no more than 1 second; for medium-priority equipment, the adjustment range is appropriately relaxed, the power fluctuation range is adjusted to 5%, and the response time is adjusted to the fluctuation of no more than 3 seconds; the adjustment range of low-priority equipment is further relaxed, the power fluctuation range is adjusted to 10%, and the response time is adjusted to the fluctuation of no more than 5 seconds. The optimization parameters of all equipment are uniformly sorted out to form an optimization data table, which includes the real-time operating power range, response time range and adjustment range of each equipment, and clarifies the specific optimization goals of each equipment, such as the target power range of high-priority equipment is 45kW to 46.5kW, the medium-priority equipment is 40kW to 42kW, and the low-priority equipment is 35kW to 38.5kW. The final output of the equipment operation optimization parameter set provides an accurate parameter basis for the implementation of the subsequent fumigation warehouse environment control strategy.

[0109] See also Figure 6 ,The steps to predict future temperature and humidity trends are as follows:

[0110] S411: Collect all temperature and humidity data of the fumigation warehouse, process the monitoring data in time series and perform frequency domain decomposition, identify significant frequency characteristics in environmental fluctuations and prioritize them by analyzing the frequency components and amplitude distribution after decomposition, and generate a frequency characteristic distribution table with significant impact;

[0111] All temperature and humidity data of the fumigation warehouse are decomposed in the frequency domain. First, multiple data segments of continuous monitoring are extracted from the time series data of temperature and humidity. Each data segment is converted into the frequency domain using discrete Fourier transform to generate a distribution table of frequency and corresponding amplitude. The frequency range is divided into three intervals: low frequency (below 0.01 Hz), medium frequency (0.01 Hz to 0.1 Hz) and high frequency (above 0.1 Hz). The amplitude proportion in each frequency interval is obtained by calculating the ratio of the sum of the amplitudes of each frequency interval to the sum of the amplitudes of all frequencies. For example, the sum of the amplitudes of the low frequency interval is 300, and the sum of the amplitudes of all frequencies is 500. The proportion is calculated by (300 / 500)×100%, which is 60%, the medium frequency interval is (150 / 500)×100%=30%, and the high frequency interval is (50 / 500)×100%=10%. Subsequently, the amplitude contribution rate of each frequency interval is analyzed to identify frequency characteristics with high amplitude proportion and significant fluctuation. For example, if a certain frequency is around 0.05Hz and the amplitude increases significantly to 25% of the total amplitude, it is determined that the frequency has a significant impact on environmental fluctuations and is marked as a key frequency. Combined with the frequency division results of each region, the frequency characteristics are prioritized according to the amplitude significance, and the frequency characteristics with the highest proportion are recorded as key frequencies, which are used to further predict environmental fluctuation trends and evaluate the needs of equipment response, and finally generate a distribution table of frequency characteristics with significant impact.

[0112] S412: Based on the frequency characteristic distribution table with significant influence, the formula is adopted:

[0113] y(t′)=Asin(2πft′+φ)+B;

[0114] Calculate the predicted temperature and humidity value y(t′) at time t′ to obtain the future temperature and humidity change trend data;

[0115] Among them, A represents the maximum fluctuation amplitude of temperature and humidity changes, which is directly related to the amplitude of the significant frequency component. The amplitude corresponding to the significant frequency is extracted from the amplitude after frequency domain decomposition. For example, if the amplitude of the significant frequency is 10, then the amplitude is 10. sin(2πft′+φ) is the sine function part, which describes the periodic changes of temperature and humidity over time: 2π represents the radian of a complete cycle, which is a mathematical constant. f is the significant frequency, which represents the dominant frequency characteristics in environmental fluctuations and is inversely proportional to the period T, that is, f=1 / T. The significant frequency is extracted through the frequency distribution in the frequency domain analysis. For example, the significant frequency is 0.0 5Hz, indicating a fluctuation with a cycle of 20 seconds. t′ is a time variable, indicating the time point of the forecast target, in seconds, and is set according to specific forecast requirements. For example, t=10 indicates the forecast of the temperature and humidity changes at the 10th second. φ indicates the initial offset of the temperature and humidity fluctuation, which is related to the initial time point of the significant frequency and is calibrated by the initial waveform of the significant frequency characteristics in the historical time series data. For example, the fluctuation starts from the lowest point and the phase is -π / 2. B indicates the average value of the temperature and humidity data, reflecting the baseline level of the overall temperature and humidity, and is calculated by the mean value of the historical temperature and humidity data. For example, the historical data is

[0116] [20,21,22,23,24], then the offset is 22.

[0117] Significant frequency f = 0.05Hz, amplitude A = 10, phase The offset B = 22, and the predicted time point is t′ = 10s. Substitute into the formula for calculation:

[0118] y(t′)=Asin(2πft′+φ)+B;

[0119]

[0120] Calculate the frequency part: 2πft′=2π·0.05·10=π;

[0121] Calculate the angle after phase superposition:

[0122] Take the sine value:

[0123] Calculate the final value:

[0124] The results show that the predicted value y(10) = 14.93 means that at the time point t′ = 10s, the temperature and humidity of the environment will reach 14.93 (the unit is set according to the data source). By comparing the predicted value with the current operating parameters of the device, it is possible to determine whether the device operating parameters meet the requirements of future temperature and humidity change trends. For example, the device may need to increase power or delay response time under this temperature and humidity, thereby providing a basis for the device operation adjustment strategy.

[0125] See also Figure 7 The specific steps for obtaining the temperature and humidity control results of the fumigation warehouse are as follows:

[0126] S421: Based on the future temperature and humidity change trend data, the predicted temperature and humidity values ​​are compared with the current equipment operation optimization parameters one by one, and whether they exceed the operating power adjustment range and response time range is analyzed, and the deviation time point and amplitude difference are recorded to generate the equipment operation matching result;

[0127] First, the trend data of future temperature and humidity changes are calculated according to the prediction formula. For example, the predicted temperature and humidity values ​​in the next 10 seconds are [14.93, 15.10, 15.25, 15.40]. These predicted values ​​are compared with the operation optimization parameters of the current equipment. For example, the operating power adjustment range of the equipment is [14.5, 15.5], and the response time is 2 seconds. Through comparison, it is determined whether the predicted value at each time point is within the operating power adjustment range of the equipment. If it exceeds the range, the specific time point and the corresponding amplitude difference are recorded. For example, at the 4th second, the predicted value 15.40 exceeds the upper limit of the operating range 15.5, and the excess amplitude is 0.10. Subsequently, the operation priority information of each device is comprehensively analyzed. For high-priority devices, the predicted temperature and humidity values ​​are matched with their power output range, response time range, and environmental sensitivity level to clarify whether the current device needs to adjust the operating parameters. For example, when the response time of the high-priority device is not enough to meet the future temperature and humidity change trend, it is necessary to record the insufficient time point and amplitude deviation to provide data support for subsequent adjustments. Finally, a set of operation matching result data tables are generated based on the above matching process, which include the matching status of each device, the deviation amount, and a mark indicating whether adjustment is required.

[0128] S422: Based on the equipment operation matching result, the power output range and response time parameters of the equipment that does not meet the fluctuation demand are adjusted, the adjusted equipment operation parameters are summarized and the adjustment range is marked, and the fumigation warehouse temperature and humidity control result is generated;

[0129] First, according to the recorded operation deviation data, analyze the priority of equipment power, response time and sensitivity adjustment. For example, for high-priority equipment with large deviation amplitude, give priority to adjusting its power output range to expand the power range to cover the future predicted temperature and humidity change range, such as adjusting the power upper limit from 15.5 to 16.0; for equipment with insufficient response time, increase its response frequency from once every 2 seconds to once every 1 second to quickly respond to changes in temperature and humidity. In specific operations, for low-priority equipment, appropriately relax the adjustment range, such as adjusting the power range from [14.5,15.5] to [14.0,16.0], to reduce resource consumption and maintain overall operation efficiency. The adjusted equipment operation parameters are summarized into a new set of optimized parameters, including the power range, response time adjustment value and adjustment range mark of each device. For example, the power range of high-priority equipment is adjusted to [14.5,16.0], and the response time is shortened to 1 second; the power range of medium-priority equipment is [14.0,15.8], and the response time remains unchanged. Finally, the temperature and humidity control results of the fumigation warehouse are generated, providing an adjustment basis for subsequent equipment operation regulation.

[0130] A fumigation warehouse temperature and humidity control system based on data analysis, the fumigation warehouse temperature and humidity control system based on data analysis is used to execute the above-mentioned fumigation warehouse temperature and humidity control method based on data analysis, the system includes:

[0131] The fuzzy data processing module defines temperature and humidity fuzzy variables based on the temperature and humidity data collected by multi-point sensors in the fumigation warehouse, uses membership functions to map the temperature and humidity data to fuzzy sets, and generates fuzzy processing results for the temperature and humidity data;

[0132] The dynamic threshold adjustment module dynamically adjusts the control threshold of the fumigation warehouse based on the fuzzy processing results of temperature and humidity data, combined with the subordinate status and the current operating conditions of the fumigation warehouse, and outputs a dynamic control threshold list after integrating the control thresholds;

[0133] The equipment operation optimization module is based on the dynamic control threshold list, matches the control threshold range with the real-time temperature and humidity data of each area, extracts the regional environmental status and adjusts the equipment operation parameters, performs gradient optimization on the operation parameters through the calculation nodes, and adjusts the node output to generate the equipment operation optimization parameters;

[0134] The temperature and humidity trend prediction module performs frequency domain decomposition and identifies significant frequency characteristics based on all temperature and humidity data of the fumigation warehouse. It predicts future temperature and humidity changes through the trend prediction model, matches the predicted trend with the equipment operation optimization parameters, adjusts the equipment parameters according to the matching information, and generates the temperature and humidity control results of the fumigation warehouse.

[0135] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A fumigation warehouse temperature and humidity control method based on data analysis, characterized in that: The following steps are involved: S1: Based on the temperature and humidity data collected by multi-point sensors in the fumigation warehouse, the fuzzy variables of the temperature and humidity data are defined, and the temperature and humidity data are mapped to the corresponding fuzzy sets according to the fuzzy variables to obtain the fuzzy processing results of the temperature and humidity data; S2: Based on the affiliation status of each temperature and humidity data in the fuzzy processing result of the temperature and humidity data, combined with the affiliation status and the current operating conditions of the fumigation warehouse, dynamically adjust the control threshold of the fumigation warehouse, integrate the adjusted control threshold, and output a dynamic control threshold list; S3: Based on the control threshold range in the dynamic control threshold list, the control threshold range is matched with the real-time temperature and humidity data of multiple areas in the fumigation warehouse, the environmental state of each area of ​​the fumigation warehouse and the operating parameters of the regulating equipment are extracted, the operating parameters are distributed to multiple computing nodes, and the gradient calculation of the regulating equipment operating parameters is performed. The sensitivity of the regulating equipment operating parameters to environmental changes is determined according to the gradient of the equipment operating parameters, the output of the node is adjusted, and the equipment operation optimization parameters are generated; S4: Perform frequency domain decomposition on all temperature and humidity data of the fumigation warehouse and identify significant frequency characteristics, predict future temperature and humidity change trends based on the significant frequency characteristics, match the future temperature and humidity change trends with the current equipment operation optimization parameters, adjust the fumigation warehouse equipment parameters based on the matching information, and generate the fumigation warehouse temperature and humidity control results.

2. The fumigation warehouse temperature and humidity control method based on data analysis according to claim 1 is characterized in that: The steps for obtaining the fuzzy processing result of the temperature and humidity data are specifically as follows: S111: Based on the temperature and humidity data collected by the multi-point sensors in the fumigation warehouse, the collected data is cleaned and normalized, the processed data is mapped to the fuzzy set according to the preset temperature and humidity classification standard, and the data is classified and sorted to generate the temperature and humidity data classification result; S112: Based on the classification result of the temperature and humidity data, the formula is used: Calculate the membership value μ(x) of the target temperature and humidity value x in the fuzzy set. The range of the membership value is [0,1]. Integrate the membership values ​​to obtain the fuzzy processing results of the temperature and humidity data. Among them, x represents the normalized temperature and humidity data point, c represents the center point of the fuzzy membership function, and σ represents the width of the membership function, which defines the extension range of the fuzzy set.

3. The fumigation warehouse temperature and humidity control method based on data analysis according to claim 2 is characterized in that: The steps for obtaining the dynamic control threshold list are specifically as follows: S211: Based on the affiliation status of each temperature and humidity data in the fuzzy processing result of the temperature and humidity data, combined with the affiliation status and the current operating conditions of the fumigation warehouse, including the humidification capacity range, dehumidification capacity range, heating power and equipment response time, dynamically adjust the control threshold of the fumigation warehouse, the control threshold is the threshold range for starting and stopping humidification, dehumidification and heating, and obtain the control threshold adjustment result of the fumigation warehouse; S212: Based on the control threshold adjustment result of the fumigation warehouse, control parameters of each equipment type and regional distribution of the fumigation warehouse are classified and integrated, including humidification equipment, dehumidification equipment and heating equipment, to obtain a dynamic control threshold list.

4. The fumigation warehouse temperature and humidity control method based on data analysis according to claim 3 is characterized in that: The steps of calculating the gradient of the operating parameters of the regulating device are specifically as follows: S311: Based on the control threshold range in the dynamic control threshold list, the real-time temperature and humidity data of multiple areas of the fumigation warehouse are processed in segments according to the time window, the environmental state of each area and the operating parameters of the regulating equipment are extracted and matched, and the equipment operating parameters are matched with the regional environmental state and then allocated to multiple computing nodes to obtain the regional environmental state and equipment operating parameter allocation results; S312: Based on the regional environmental status and the equipment operation parameter allocation result, the formula is used: Calculate the updated equipment operating parameters, which represents the optimized value θ of the equipment operating parameters in the t+1th iteration (t+1) ; Among them, J(θ) is the objective function in the fumigation warehouse control, and its complete form is: θ (t) is the value of the corresponding device operating parameter θ in the tth iteration, θ (t+1) By the current gradient value and the learning rate η is updated from the current iteration θ (t) Update to the next iteration θ (t+1) , η is the learning rate, which indicates the step size of parameter update, is the gradient of the objective function J(θ) in the fumigation warehouse control with respect to the operating parameter θ of the corresponding equipment, indicating the sensitivity of the current parameter to the objective function in the fumigation warehouse control, m is the number of samples, j is the sample number currently being calculated, and y j represents the expected device output parameter value, f(x j ,θ) is the device output value, x j is the input environmental state data point.

5. The fumigation warehouse temperature and humidity control method based on data analysis according to claim 4 is characterized in that: The steps for obtaining the equipment operation optimization parameters are specifically as follows: S321: Based on the updated device operating parameters and the current environmental data change range, the difference value of the operating parameters is compared with the preset sensitivity range, the sensitivity state of the device operating parameters is determined, and the devices are graded and adjusted in combination with the global device priority to generate the device sensitivity state and priority information to the environmental change; S322: Based on the sensitivity status and priority information of the device to environmental changes, the adjustment range of each priority device is classified and sorted according to the response time and power range to generate device operation optimization parameters.

6. The method for controlling temperature and humidity in a fumigation warehouse based on data analysis according to claim 1 is characterized in that: The steps of predicting the future temperature and humidity change trend are specifically as follows: S411: Collect all temperature and humidity data of the fumigation warehouse, process the monitoring data in time series and perform frequency domain decomposition, identify significant frequency characteristics in environmental fluctuations and prioritize them by analyzing the frequency components and amplitude distribution after decomposition, and generate a frequency characteristic distribution table with significant impact; S412: Based on the frequency characteristic distribution table with significant influence, the formula is adopted: y(t′)=Asin(2πft′+φ)+B; Calculate the predicted temperature and humidity value y(t′) at time t′ to obtain the future temperature and humidity change trend data; Among them, A represents the maximum fluctuation amplitude of temperature and humidity changes, sin(2πft′+φ) is the sine function part, which describes the periodic changes of temperature and humidity over time, 2π represents the radian of a complete cycle, f is the significant frequency, which represents the dominant frequency characteristics in environmental fluctuations and is inversely proportional to the period T, that is, f=1 / T, t′ is the time variable, which represents the time point of the predicted target, φ represents the initial offset of the temperature and humidity fluctuation, and B represents the average value of the temperature and humidity data.

7. The method for controlling temperature and humidity in a fumigation warehouse based on data analysis according to claim 6 is characterized in that: The steps for obtaining the temperature and humidity control results of the fumigation warehouse are specifically as follows: S421: Based on the future temperature and humidity change trend data, the predicted temperature and humidity values ​​are compared with the operation optimization parameters of the current equipment one by one, and whether they exceed the operation power adjustment range and response time range is analyzed, and the deviation time point and amplitude difference are recorded to generate the equipment operation matching result; S422: Based on the equipment operation matching result, the power output range and response time parameters of the equipment that do not meet the fluctuation demand are adjusted, the adjusted equipment operation parameters are summarized and the adjustment range is marked, and the temperature and humidity control result of the fumigation warehouse is generated.

8. A fumigation warehouse temperature and humidity control system based on data analysis, characterized in that: According to the method for controlling temperature and humidity in a fumigation warehouse based on data analysis according to any one of claims 1 to 7, the system comprises: The fuzzy data processing module defines temperature and humidity fuzzy variables based on the temperature and humidity data collected by multi-point sensors in the fumigation warehouse, uses membership functions to map the temperature and humidity data to fuzzy sets, and generates fuzzy processing results for the temperature and humidity data; The dynamic threshold adjustment module dynamically adjusts the control threshold of the fumigation warehouse based on the fuzzy processing result of the temperature and humidity data, combined with the subordinate status and the current operating conditions of the fumigation warehouse, and outputs a dynamic control threshold list after integrating the control thresholds; The equipment operation optimization module matches the control threshold range with the real-time temperature and humidity data of each area based on the dynamic control threshold list, extracts the regional environmental status and adjusts the equipment operation parameters, performs gradient optimization on the operation parameters through the calculation nodes, and adjusts the node output to generate the equipment operation optimization parameters; The temperature and humidity trend prediction module performs frequency domain decomposition and identifies significant frequency characteristics based on all temperature and humidity data of the fumigation warehouse. It predicts future temperature and humidity changes through the trend prediction model, matches the predicted trend with the equipment operation optimization parameters, adjusts the equipment parameters according to the matching information, and generates the temperature and humidity control results of the fumigation warehouse.

Images