Heat preservation control method and system for horticultural facilities for soilless culture

Abstract

The invention discloses a horticultural facility heat preservation control method for soilless culture, which comprises the following steps: acquiring historical environment parameter data in a horticultural facility and preprocessing the data to obtain a time sequence data set; constructing a temperature prediction model based on a machine learning algorithm; actually measuring temperature data in the gardening facility in a future time period, performing data fusion on the time sequence data set and the actually measured temperature data in the future time period, and inputting the fused data into the temperature prediction model for parameter optimization; the temperature in the gardening facility is predicted in real time based on the optimized temperature prediction model, and when predicted temperature data exceed a preset temperature threshold value, early warning is carried out; and based on the predicted temperature data, an automatic adjustment control strategy is adopted to intervene the temperature in the gardening facility in advance. According to the invention, historical data analysis, real-time monitoring, future forecasting and intelligent control are organically combined, and accurate prediction and active regulation and control of the internal temperature of the facility are realized.

Description

Technical Field

[0001] The invention belongs to the technical field of equipment thermal insulation control, and in particular relates to a thermal insulation control method and system for horticultural facilities for soilless cultivation. Background Art

[0002] The temperature changes inside soilless horticultural facilities are greatly affected by external environmental conditions, especially the fluctuations in meteorological factors, which can cause drastic fluctuations in the temperature inside the facilities. Unstable temperature can affect the normal growth and development of crops, and even lead to reduced production or reduced quality. Currently, commonly used temperature control equipment is mainly adjusted based on real-time monitored temperature data, which has a lag and cannot provide early warning and precise control.

[0003] How to predict the temperature change trend inside the facility in the future based on the historical temperature data inside the facility, combined with the current environmental conditions and future weather forecasts, and adjust the temperature control equipment in advance according to the prediction results to achieve early warning and precise control of the temperature inside the facility is a technical problem that needs to be solved urgently. At the same time, due to the complex factors affecting the temperature change inside the facility, there are certain nonlinearities and uncertainties, and it is difficult for traditional temperature prediction models to accurately predict the temperature inside the facility.

[0004] Therefore, how to establish an intelligent temperature prediction model based on environmental factors and historical data to improve the accuracy of temperature prediction is a technical problem that needs to be solved urgently. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a method and system for heat preservation control of horticultural facilities for soilless cultivation to solve the problems existing in the above prior art.

[0006] To achieve the above object, the present invention provides a method for controlling heat preservation of horticultural facilities for soilless cultivation, comprising the following steps:

[0007] Obtain historical environmental parameter data inside horticultural facilities and preprocess them to obtain a time series data set;

[0008] Build a temperature prediction model based on machine learning algorithms;

[0009] Measure the temperature data of the horticultural facility in the future time period, fuse the time series data set and the measured temperature data in the future time period, and input the data into the temperature prediction model for parameter optimization;

[0010] Based on the optimized temperature prediction model, the temperature inside the horticultural facilities is predicted in real time. When the predicted temperature data exceeds the preset temperature threshold, an early warning is issued.

[0011] Based on the predicted temperature data, an automatic adjustment control strategy is used to intervene in the temperature inside the horticultural facilities in advance.

[0012] Optionally, the process of obtaining historical environmental parameter data inside the soilless cultivation horticultural facility and preprocessing the data to obtain a time series data set includes:

[0013] A sensor network is deployed inside a soilless horticultural facility to obtain historical environmental parameter data; the historical environmental parameter data is filtered, denoised and outliers are removed, and then the processed historical environmental parameter data is subjected to time series analysis and normalization to obtain a time series data set.

[0014] Optionally, the process of measuring the temperature data within the facility in the future time period, fusing the time series data set and the measured temperature data within the future time period, and inputting the data into the temperature prediction model for parameter optimization includes:

[0015] The temperature prediction model predicts temperature data in a future time period based on an input time series data set, compares the predicted temperature data in the future time period with the input measured temperature data in the future time period to obtain a temperature deviation, and dynamically adjusts the parameters of the temperature prediction model based on the temperature deviation to obtain an optimized temperature prediction model.

[0016] Optionally, based on the predicted temperature data, the process of using the automatic adjustment control strategy to intervene in advance the temperature inside the horticultural facility includes:

[0017] If the predicted temperature data is higher than the preset temperature threshold, the shading measures and ventilation system adjustment are activated; if the predicted temperature data is lower than the preset temperature threshold, the heating measures are activated.

[0018] Optionally, it also includes real-time monitoring of temperature data inside the horticultural facilities and comparing it with corresponding temperature prediction data. When the comparison result exceeds a preset threshold, an adaptive optimization mechanism is triggered to update the parameters of the temperature prediction model in real time.

[0019] The present invention also provides a heat preservation control system for horticultural facilities for soilless cultivation, which is used to implement a heat preservation control method for horticultural facilities for soilless cultivation, and includes: a data acquisition module, a model construction module, a model optimization module, a temperature prediction module and a temperature intervention module;

[0020] The data acquisition module is used to acquire historical environmental parameter data inside the horticultural facility and perform preprocessing to obtain a time series data set;

[0021] The model building module is used to build a temperature prediction model based on a machine learning algorithm;

[0022] The model optimization module is used to measure the temperature data within the horticultural facility in the future time period, and after data fusion of the time series data set and the measured temperature data in the future time period, input them into the temperature prediction model for parameter optimization;

[0023] The temperature prediction module is used to predict the temperature inside the horticultural facility in real time based on the optimized temperature prediction model, and issue an early warning when the predicted temperature data exceeds a preset temperature threshold;

[0024] The temperature intervention module is used to intervene in advance the temperature inside the gardening facility by adopting an automatic adjustment control strategy based on the predicted temperature data.

[0025] Optionally, the model optimization module includes a temperature prediction unit and a deviation adjustment unit;

[0026] The temperature prediction unit is used to predict temperature data in a future time period based on an input time series data set;

[0027] The deviation adjustment unit is used to compare the predicted temperature data in the future time period with the input measured temperature data in the future time period to obtain a temperature deviation, and dynamically adjust the parameters of the temperature prediction model based on the temperature deviation to obtain an optimized temperature prediction model.

[0028] Optionally, the temperature intervention module includes a high temperature intervention unit and a low temperature intervention unit;

[0029] The high temperature intervention unit is used to start the sunshade means and adjust the ventilation system when the predicted temperature data is higher than the preset temperature threshold;

[0030] The low temperature intervention unit is used to start heating measures when the predicted temperature data is lower than a preset temperature threshold.

[0031] The present invention also provides an electronic device, comprising: a memory and a processor; the memory is used to store a program; the processor is used to execute the program to implement each step of the heat preservation control method for horticultural facilities for soilless cultivation.

[0032] The present invention also provides a readable storage medium on which a computer program is stored. When the computer program is executed by a processor, each step of the method for controlling the thermal insulation of a horticultural facility for soilless cultivation is implemented.

[0033] Compared with the prior art, the present invention has the following advantages and technical effects:

[0034] The present invention discloses a method for heat preservation control of horticultural facilities for soilless cultivation. The method obtains historical environmental parameter data inside the facility, performs preprocessing, and then establishes a temperature prediction model. Combined with real-time future temperature data, the temperature inside the facility is dynamically predicted and corrected. When the predicted temperature exceeds the crop growth threshold, an early warning signal is triggered and the temperature control strategy is automatically adjusted. At the same time, by continuously monitoring the error between the actual temperature and the predicted value, the adaptive optimization of the model is achieved, and the prediction accuracy is continuously improved. The uniqueness of the present invention lies in the organic combination of historical data analysis, real-time monitoring, future forecasting, and intelligent control, which realizes the accurate prediction and active regulation of the temperature inside the facility. This forward-looking temperature management method can effectively avoid the adverse effects of abnormal temperature on crop growth, improve the yield and quality of soilless cultivation, and has important practical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0036] Figure 1 is a flow chart of a method according to an embodiment of the present invention;

[0037] Figure 2 Schematic diagram of the system structure of an embodiment of the present invention. DETAILED DESCRIPTION

[0038] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0039] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0040] Embodiment 1

[0041] like Figure 1 As shown, this embodiment provides a method for controlling heat preservation of horticultural facilities for soilless cultivation, comprising the following steps:

[0042] Obtain historical environmental parameter data inside horticultural facilities and preprocess them to obtain a time series data set;

[0043] Build a temperature prediction model based on machine learning algorithms;

[0044] Measure the temperature data of the horticultural facility in the future time period, fuse the time series data set and the measured temperature data in the future time period, and input the data into the temperature prediction model for parameter optimization;

[0045] Based on the optimized temperature prediction model, the temperature inside the horticultural facilities is predicted in real time. When the predicted temperature data exceeds the preset temperature threshold, an early warning is issued.

[0046] Based on the predicted temperature data, an automatic adjustment control strategy is used to intervene in the temperature inside the horticultural facilities in advance.

[0047] It is feasible to obtain historical environmental parameter data inside soilless horticultural facilities, including temperature data, meteorological factors, humidity, light, etc., and perform data preprocessing to remove outliers and noise data to obtain a normalized time series data set. The process includes:

[0048] Historical environmental parameter data, including temperature, humidity, light intensity, etc., are obtained through the sensor network inside the soilless cultivation facility, and the data is stored in the database. The original environmental parameter data obtained is preprocessed, and the moving average filtering algorithm is used to remove high-frequency noise in the data to improve the data quality. The data is judged whether it is abnormal according to the preset threshold. If the data exceeds the normal range, it is marked as an outlier and removed from the data set. The preprocessed data is subjected to time series analysis to extract the trend characteristics of environmental parameter changes at different time scales. The Z-score standardization method is used to normalize the time series data, unifying data of different dimensions to the same scale for subsequent analysis and modeling.

[0049] As a specific embodiment, in a soilless cultivation facility, environmental parameter data on temperature, humidity and light intensity are continuously collected through an internal sensor network. For example, data is recorded once a minute and then stored in a database for subsequent analysis. In order to improve data quality, a moving average filtering algorithm is used to process raw data. The specific method is to select a suitable window size, such as 10 minutes, and calculate the average value of the data during this period to smooth short-term fluctuations and remove high-frequency noise. In the data preprocessing stage, outliers need to be identified and processed. By setting specific thresholds, such as the temperature should not be lower than 10°C or higher than 35°C, once the data exceeds these thresholds, it is marked as abnormal and excluded. The processed data will be used for time series analysis to extract the trend of environmental parameter changes under time scales such as daily changes and seasonal changes. In order to unify data of different dimensions to the same scale, the Z-score standardization method is used to normalize the time series data, that is, the mean is subtracted from each data point and divided by the standard deviation. Next, a time series prediction model is constructed, such as using an ARIMA model or an LSTM neural network to predict future environmental parameters. These models are trained based on historical data and can predict temperature, humidity, and light intensity over a period of time in the future. Ultimately, these preprocessed and standardized time series data sets will be used as input for machine learning algorithms to optimize environmental control and estimate yields in soilless cultivation facilities, thereby achieving more efficient crop growth and management.

[0050] The process of measuring the temperature data of the facility in the future time period, fusing the time series data set and the measured temperature data in the future time period, and inputting the data into the temperature prediction model for parameter optimization includes:

[0051] According to the pre-processed historical environmental parameter data, a temperature prediction model is constructed, and the optimized model parameters are obtained through model training. This embodiment uses an LSTM neural network model for modeling. In the LSTM neural network model, the historical environmental parameter data is used as input, and the time series characteristics of temperature changes and the influence of environmental factors are captured through the learning and training of a multi-layer neural network to obtain optimized model parameters. Specifically, in the LSTM model, the number of network layers is set to 3, each layer contains 50 neurons, and the historical environmental parameters are used as input. The network weights are optimized through the back propagation algorithm, and training is performed for 100 cycles until the prediction error of the model stabilizes at a low level.

[0052] It is feasible to obtain weather forecast data for a period of time in the future, extract characteristic parameters related to the internal temperature of the facility, such as air temperature, humidity, light, etc., input them into the temperature prediction model together with the current environmental parameter data, dynamically correct the prediction results, and improve the prediction accuracy.

[0053] By acquiring future meteorological data, temperature parameters, humidity parameters and light parameters are extracted from them, which will be used to describe the environmental conditions in the future. Through feature extraction technology, key information such as temperature, humidity and light intensity in these meteorological data are extracted to prepare for subsequent data processing. Combined with the current environmental data, which includes but is not limited to indoor and outdoor temperature readings, humidity conditions and light intensity, these real-time data are integrated with future predicted meteorological data through data combination methods. The integrated data is used as model input and input into the temperature prediction model. This model is based on machine learning algorithms and can predict future indoor temperatures based on input environmental and meteorological parameters. The prediction results of the model are dynamically adjusted, and the results are corrected according to the deviation between real-time data and predicted data to ensure the accuracy of the prediction. Through this dynamic adjustment mechanism, the prediction of the model can be updated in real time to adapt to environmental changes, thereby improving the overall prediction accuracy.

[0054] As a specific embodiment, the weather forecast data for the next week is obtained through the meteorological data interface, including parameters such as the highest temperature, lowest temperature, average temperature, relative humidity and light duration of each day. Data cleaning and feature extraction technology are used to remove missing values ​​and outliers in the meteorological data, and three key parameters of temperature, humidity and light are extracted. The temperature parameter includes the highest temperature, the lowest temperature and the average temperature, the humidity parameter is the average relative humidity, and the light parameter is the light duration. At the same time, indoor and outdoor environmental data are collected in real time through environmental sensors, including indoor and outdoor temperature, humidity and light intensity. The meteorological forecast data and the real-time environmental data are aligned according to the timestamp, and the two types of data are fused using a data fusion algorithm such as Kalman filtering to obtain the temperature, humidity and light prediction values ​​at each moment in the next week. The fused data is input into a temperature prediction model based on a long short-term memory network (LSTM), which learns the nonlinear relationship between temperature and other environmental factors through training, and can predict the future temperature change trend based on historical data and current environmental conditions. The input of the model is the temperature, humidity, and light data of the past 24 hours and the prediction data of the next week, and the output is the temperature prediction value at each moment in the next week. The model parameters are updated in real time through a dynamic adjustment mechanism, and the parameters are fine-tuned according to the deviation between the predicted value and the actual value to continuously improve the prediction accuracy.

[0055] Furthermore, the integrated data is used as model input and fed into a temperature prediction model. This model is based on a machine learning algorithm and can predict the future indoor temperature based on the input environmental and meteorological parameters. The process includes:

[0056] Acquire multi-source environmental and meteorological data, including indoor and outdoor temperature and humidity, air quality, solar radiation intensity, etc., and perform data cleaning and standardization to obtain a unified format of data sets. According to the time series characteristics of the data, the sliding time window method is used to divide the data to generate training samples and test samples. Input the current environmental and meteorological parameters into the trained prediction model to obtain the temperature prediction value for a period of time in the future. Compare the temperature prediction result with the actual temperature value, calculate the prediction error, and dynamically update the model parameters according to the error size to achieve online learning and adaptive optimization of the model, and continuously improve the prediction accuracy.

[0057] It is feasible to determine whether the temperature inside the facility will exceed the normal growth range of crops in the future based on the temperature prediction results. If the predicted temperature exceeds the threshold, an early warning signal will be triggered to notify the temperature control equipment to start the adjustment mechanism in advance:

[0058] The temperature monitoring system collects real-time temperature data inside the facility, which will be used to analyze and predict future temperature trends. Through data analysis within a time period, it is determined whether there is a risk of exceeding the normal growth range of crops. If the analysis results show that the temperature will exceed the growth range, the system will automatically trigger an early warning signal. The signal notifies the temperature control device, which starts the necessary adjustment mechanism to maintain the temperature within the appropriate range for crop growth. In this process, the data on predicted temperature and growth range will be used to set the adjustment parameters of the temperature control device to ensure the accuracy and timeliness of temperature adjustment. Through this series of automated controls, it is ensured that crops grow under the best environmental conditions, thereby improving the growth quality and yield of crops.

[0059] As a specific embodiment, in the intelligent temperature control system, indoor temperature data is first collected in real time through temperature sensors installed inside the facility, the data is recorded once a minute, and transmitted to the central processing system through a wireless network. The system uses a sliding window algorithm to average the temperature data in the past hour to obtain a more stable temperature reading. Next, the system uses a long short-term memory network (LSTM) model to analyze historical temperature data and weather forecast data to predict temperature trends in the next 24 hours. This model has been trained on a large amount of historical data and can identify patterns and trends in temperature changes. If the prediction results show that the future temperature is at risk of exceeding the normal growth range of crops, the system will automatically calculate the temperature difference that needs to be adjusted, and calculate the optimal start-up time and operating intensity of the air conditioner or heating equipment through the adjustment algorithm. For example, if the temperature is predicted to reach 32°C in the next 12 hours, the system will start the air conditioner 2 hours in advance and set the cooling intensity to 70% to gradually reduce the indoor temperature to 28°C, keeping it within the appropriate range for crop growth. In addition, the system will also monitor the humidity and light intensity indoors and outdoors in real time to ensure that these environmental factors are also within the appropriate range. In this way, the intelligent temperature control system can ensure that crops grow under optimal environmental conditions, thereby effectively improving the growth quality and yield of crops, while reducing energy consumption and improving energy efficiency.

[0060] It is feasible that the temperature control equipment automatically adjusts the control strategy according to the early warning signals and predicted temperature trends, such as adjusting ventilation, shading, heating, etc. to intervene in the temperature inside the facility in advance to ensure that the temperature is maintained within the normal growth range of crops:

[0061] In this embodiment, the temperature data inside the facility is continuously monitored by a temperature sensor, and the data is transmitted to the central processing unit, which performs real-time temperature analysis based on the received data. The central processing unit compares the analyzed temperature data with the preset suitable temperature for crop growth. If the current temperature deviates from the preset range, an early warning signal is generated. The early warning signal triggers the automatic adjustment mechanism of the temperature control device, and predicts the temperature change in the future period of time according to the temperature trend prediction model, so as to determine the control strategy to be adopted. According to the prediction results, if the predicted temperature is too high, the shading means and the ventilation system are activated; if the predicted temperature is too low, the heating measures are activated. The ventilation system controls the air circulation in the facility by adjusting the fan speed and the size of the open window to reduce the internal temperature or increase the air circulation to adapt to the changes in the external environment. The shading means automatically adjusts the coverage of the shading system to reduce the heat brought by direct sunlight, thereby controlling the temperature rise inside the facility. The heating measures increase the temperature inside the facility by activating the heating equipment to ensure that the crops can grow normally under lower external temperature conditions.

[0062] As a specific embodiment, the temperature sensor collects the temperature data inside the facility every 5 minutes and sends the data to the central processing unit through the wireless transmission module. The central processing unit uses the Kalman filter algorithm to filter the noise and fuse the received temperature data to obtain a more accurate temperature value. By comparing the current temperature with the preset crop growth temperature range (such as 23℃-28℃), if the temperature exceeds the range, an early warning signal is generated. The early warning signal triggers the temperature trend prediction model, which uses the ARIMA time series model to predict the temperature change trend within the next 4 hours based on the temperature data of the past 24 hours. If the predicted temperature will exceed 30℃, the shading system is started to reduce solar radiation by adjusting the shading area (such as 50%-80%), and the ventilation system is adjusted to increase the fan speed by 20%, and the skylight opening area is increased by 15% to accelerate heat loss. If the predicted temperature is lower than 20℃, the heating device is started to maintain the internal temperature of the facility by adjusting the heating power (such as 1000W-2000W). Through the above-mentioned adjustment means, the temperature inside the facility can be accurately controlled to create the best growth environment for crops.

[0063] It is feasible to continuously monitor the real-time temperature data inside the facility, compare it with the predicted value, calculate the prediction error, and when the prediction error exceeds the preset threshold, trigger the model adaptive optimization mechanism, use the latest temperature data to update the prediction model online and fine-tune the parameters to continuously improve the prediction accuracy:

[0064] Through real-time temperature data acquisition, the internal temperature of the facility is monitored in real time, and these data are transmitted to the prediction model for real-time prediction. By comparing the data with the model, the prediction error is calculated. If the prediction error calculation result shows that the error exceeds the preset threshold comparison result, the adaptive optimization mechanism will be triggered. At this time, the system will use the latest data to adjust the prediction model through online model update and parameter fine-tuning process. The adjusted model is used for prediction again to verify the effectiveness of the adjustment, thereby continuously improving the prediction accuracy. Through this circular feedback mechanism, the model is continuously optimized to adapt to environmental changes.

[0065] As a specific embodiment, the temperature control device collects the temperature data inside the facility every 5 minutes through the temperature sensor and transmits the data to the central processing unit. The central processing unit uses the LSTM neural network model to predict the temperature within the next hour, compares the prediction result with the actual temperature data, and calculates the mean absolute percentage error (MAPE). If the MAPE exceeds the preset threshold of 5%, the adaptive optimization mechanism is triggered. The system uses the latest 100 data points to fine-tune the weights of the LSTM model through the stochastic gradient descent algorithm, sets the learning rate to 0.1, and iterates 50 times. The adjusted model predicts again and calculates the MAPE. If the MAPE is reduced to less than 3%, the optimization is considered effective and the adjusted model is applied to subsequent predictions. If the MAPE does not meet the standard, the iterative optimization continues until the accuracy requirements are met. Through this real-time data-driven model optimization method, the temperature control system can adaptively adjust the prediction model according to the dynamic changes in the internal environment of the facility, continuously improve the accuracy of temperature prediction, and provide a reliable basis for intelligent temperature control decision-making.

[0066] like Figure 2 As shown, this embodiment also provides a heat preservation control system for a horticultural facility for soilless cultivation, which is used to implement a heat preservation control method for a horticultural facility for soilless cultivation, including: a data acquisition module, a model construction module, a model optimization module, a temperature prediction module and a temperature intervention module;

[0067] The data acquisition module is used to acquire historical environmental parameter data inside the horticultural facilities and perform preprocessing to obtain a time series data set;

[0068] The model building module is used to build a temperature prediction model based on a machine learning algorithm;

[0069] The model optimization module is used to measure the temperature data within the horticultural facility in the future time period, fuse the time series data set and the measured temperature data in the future time period, and input them into the temperature prediction model for parameter optimization;

[0070] The temperature prediction module is used to make real-time predictions of the temperature inside the horticultural facilities based on the optimized temperature prediction model, and issue an early warning when the predicted temperature data exceeds the preset temperature threshold;

[0071] The temperature intervention module is used to intervene in the temperature inside the horticultural facilities in advance by automatically adjusting the control strategy based on the predicted temperature data.

[0072] Furthermore, the model optimization module includes a temperature prediction unit and a deviation adjustment unit; the temperature prediction unit is used to predict temperature data in a future time period based on an input time series data set; the deviation adjustment unit is used to compare the predicted temperature data in the future time period with the input measured temperature data in the future time period to obtain a temperature deviation, and dynamically adjust the parameters of the temperature prediction model based on the temperature deviation to obtain an optimized temperature prediction model.

[0073] Furthermore, the temperature intervention module includes a high temperature intervention unit and a low temperature intervention unit; the high temperature intervention unit is used to start shading measures and adjust the ventilation system when the predicted temperature data is higher than a preset temperature threshold; the low temperature intervention unit is used to start heating measures when the predicted temperature data is lower than a preset temperature threshold.

[0074] Embodiment 2

[0075] The present invention also provides an electronic device, comprising: a memory and a processor; the memory is used to store a program; the processor is used to execute the program to implement each step of the heat preservation control method for horticultural facilities for soilless cultivation.

[0076] Embodiment 3

[0077] The present invention also provides a readable storage medium on which a computer program is stored. When the computer program is executed by a processor, each step of the method for controlling the thermal insulation of a horticultural facility for soilless cultivation is implemented.

[0078] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for controlling heat preservation of horticultural facilities for soilless cultivation, characterized in that: The following steps are involved: Obtain historical environmental parameter data inside horticultural facilities and preprocess them to obtain a time series data set; Build a temperature prediction model based on machine learning algorithms; Measure the temperature data of the horticultural facility in the future time period, fuse the time series data set and the measured temperature data in the future time period, and input the data into the temperature prediction model for parameter optimization; Based on the optimized temperature prediction model, the temperature inside the horticultural facilities is predicted in real time. When the predicted temperature data exceeds the preset temperature threshold, an early warning is issued. Based on the predicted temperature data, an automatic adjustment control strategy is used to intervene in the temperature inside the horticultural facilities in advance.

2. The method for controlling heat preservation of horticultural facilities for soilless cultivation according to claim 1, characterized in that: The process of obtaining historical environmental parameter data inside soilless horticultural facilities and preprocessing them to obtain a time series data set includes: A sensor network is deployed inside a soilless horticultural facility to obtain historical environmental parameter data; the historical environmental parameter data is filtered, denoised and outliers are removed, and then the processed historical environmental parameter data is subjected to time series analysis and normalization to obtain a time series data set.

3. The method for controlling heat preservation of horticultural facilities for soilless cultivation according to claim 1, characterized in that: The process of measuring the temperature data within the facility in the future time period, fusing the time series data set and the measured temperature data in the future time period, and inputting the data into the temperature prediction model for parameter optimization includes: The temperature prediction model predicts temperature data in a future time period based on an input time series data set, compares the predicted temperature data in the future time period with the input measured temperature data in the future time period to obtain a temperature deviation, and dynamically adjusts the parameters of the temperature prediction model based on the temperature deviation to obtain an optimized temperature prediction model.

4. The method for controlling heat preservation of horticultural facilities for soilless cultivation according to claim 1, characterized in that: Based on the predicted temperature data, the process of using automatic adjustment control strategy to intervene in the temperature inside the horticultural facility in advance includes: If the predicted temperature data is higher than the preset temperature threshold, the shading measures and ventilation system adjustment are activated; if the predicted temperature data is lower than the preset temperature threshold, the heating measures are activated.

5. The method for controlling heat preservation of horticultural facilities for soilless cultivation according to claim 1, characterized in that: It also includes real-time monitoring of temperature data inside horticultural facilities and comparing it with the corresponding temperature prediction data. When the comparison result exceeds the preset threshold, the adaptive optimization mechanism is triggered to update the parameters of the temperature prediction model in real time.

6. A thermal insulation control system for horticultural facilities for soilless cultivation, characterized in that: Used to implement the heat preservation control method for horticultural facilities for soilless cultivation according to any one of claims 1 to 5, comprising: a data acquisition module, a model building module, a model optimization module, a temperature prediction module and a temperature intervention module; The data acquisition module is used to acquire historical environmental parameter data inside the horticultural facility and perform preprocessing to obtain a time series data set; The model building module is used to build a temperature prediction model based on a machine learning algorithm; The model optimization module is used to measure the temperature data within the horticultural facility in the future time period, and after data fusion of the time series data set and the measured temperature data in the future time period, input them into the temperature prediction model for parameter optimization; The temperature prediction module is used to predict the temperature inside the horticultural facility in real time based on the optimized temperature prediction model, and issue an early warning when the predicted temperature data exceeds a preset temperature threshold; The temperature intervention module is used to intervene in advance the temperature inside the gardening facility by adopting an automatic adjustment control strategy based on the predicted temperature data.

7. The heat preservation control system for horticultural facilities for soilless cultivation according to claim 6, characterized in that: The model optimization module includes a temperature prediction unit and a deviation adjustment unit; The temperature prediction unit is used to predict temperature data in a future time period based on an input time series data set; The deviation adjustment unit is used to compare the predicted temperature data in the future time period with the input measured temperature data in the future time period to obtain a temperature deviation, and dynamically adjust the parameters of the temperature prediction model based on the temperature deviation to obtain an optimized temperature prediction model.

8. The heat preservation control system for horticultural facilities for soilless cultivation according to claim 6, characterized in that: The temperature intervention module includes a high temperature intervention unit and a low temperature intervention unit; The high temperature intervention unit is used to start the sunshade means and adjust the ventilation system when the predicted temperature data is higher than the preset temperature threshold; The low temperature intervention unit is used to start heating measures when the predicted temperature data is lower than a preset temperature threshold.

9. An electronic device, characterized in that: include: Memory and processor; The memory is used to store programs; The processor is used to execute the program to implement the various steps of the heat preservation control method for horticultural facilities for soilless cultivation as described in any one of claims 1-5.

10. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, each step of the method for controlling heat preservation of a horticultural facility for soilless cultivation as claimed in any one of claims 1 to 5 is implemented.

Images