Application of temperature prediction method based on LSTM algorithm in plateau precision agriculture
By constructing the LSTM algorithm model, the nonlinear and timing problems of temperature prediction in plateau environments are solved, high-precision temperature prediction is achieved, and the scientificity of precision agriculture and resource utilization efficiency are improved.
Patent Information
- Application Number
- CN202510553618.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-29
- Publication Date
- 2025-07-18
AI Technical Summary
In the prior art In plateau environment, traditional temperature prediction methods are difficult to accurately process nonlinear and strong timing correlation meteorological data, and lack LSTM algorithm solutions for systematic application in precision agriculture.
The temperature prediction model is constructed using the LSTM algorithm, and the future temperature prediction results are generated through data collection and preprocessing, model construction and optimization, and backpropagation training, and applied to precision agricultural management.
It improves the accuracy and stability of temperature prediction, and improves the scientific production nature and resource utilization efficiency in precision agriculture.
Smart Images
Figure CN120336731A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of precision agriculture, and specifically to the application of a temperature prediction method based on the LSTM algorithm in plateau precision agriculture. Background Art
[0002] Precision agriculture is a method of fine management of agricultural production based on information technology, aiming to improve production efficiency, reduce resource waste and protect the ecological environment. Temperature, as a key environmental factor affecting the growth and development of crops, its accurate prediction is crucial for agricultural production decisions, such as irrigation scheduling, pest control and crop growth monitoring. Especially in the plateau environment, extreme weather occurs frequently, and the temperature prediction for the field meteorological environment and greenhouse microclimate is extremely important.
[0003] Traditional temperature prediction methods mainly rely on statistical models (such as regression analysis) and physical models (such as numerical weather prediction), but these methods show certain limitations when dealing with non-linear and strongly time-series correlated data. In recent years, deep learning technology, especially the long short-term memory (LSTM) network, has been widely used in prediction tasks in fields such as finance and healthcare due to its excellent performance in processing time-series data. However, in precision agriculture, the application of the LSTM algorithm for temperature prediction is not yet mature, lacking systematic research and specific implementation plans. Therefore, developing a temperature prediction method based on LSTM to improve the temperature management level in precision agriculture has important practical significance. Summary of the Invention
[0004] The purpose of the present invention is to provide the application of a temperature prediction method based on the LSTM algorithm in plateau precision agriculture, and by constructing an LSTM model, accurately predict the temperature change in the future for a period of time, providing a scientific decision-making basis for agricultural production.
[0005] To achieve the above purpose, the present invention provides the following technical solution: The application of a temperature prediction method based on the LSTM algorithm in plateau precision agriculture, which is characterized in that it includes the following steps: S1: Data collection and preprocessing: Collect historical temperature data, meteorological data and farmland environment data, and perform data cleaning and normalization processing; S2: LSTM model construction: Design and construct an LSTM network model, and determine the network structure and parameters; S3: Model training and optimization: Use the backpropagation algorithm to train the model, and adopt an optimizer to adjust the learning rate to prevent overfitting; S4: Temperature prediction and application: Use the trained model to predict the future temperature, and apply the prediction results to precision agriculture management.
[0006] Preferably, the data preprocessing includes removing outliers and processing the data using standardization or normalization methods, collecting historical temperature data, meteorological data (such as humidity, wind speed), and farmland environmental data (such as soil humidity). Missing values and outliers are removed through data cleaning, and the data is preprocessed using standardization or normalization methods to ensure that the data meets the input requirements of the LSTM model.
[0007] Preferably, the LSTM model uses a sliding window method to convert time series data into a supervised learning problem, designs and constructs an LSTM network model, and determines parameters such as the network structure (number of layers, number of hidden units) and activation function. The sliding window method is used to convert time series data into a supervised learning problem, generating a training set and a test set.
[0008] Preferably, the model training uses the Adam optimizer and prevents overfitting through early stopping and dropout techniques. The backpropagation algorithm is used to train the LSTM model, the Adam optimizer is used to adjust the learning rate, and early stopping and dropout techniques are used to prevent overfitting. The mean squared error (MSE) is used as the loss function to evaluate the model performance.
[0009] Preferably, the temperature prediction results are applied to irrigation management, pest control, and crop growth monitoring in precision agriculture. The trained LSTM model is used to predict future temperatures and generate prediction results. The results are applied to links such as irrigation management, pest control, and crop growth monitoring in precision agriculture to optimize production decisions.
[0010] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. The application of the temperature prediction method based on the LSTM algorithm in plateau precision agriculture uses the LSTM algorithm to capture long-term dependencies in the temperature time series, improving the prediction accuracy. 2. The application of the temperature prediction method based on the LSTM algorithm in plateau precision agriculture ensures the stability and generalization ability of the prediction model through data preprocessing and model optimization. 3. The application of the temperature prediction method based on the LSTM algorithm in plateau precision agriculture combines the temperature prediction results with precision agriculture practices, improving the scientific nature of agricultural production and the resource utilization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] Figure 1 This is the overall flowchart of the system of the present invention, showing the complete process from data collection to prediction application. Figure 2 This is the schematic diagram of the LSTM model structure of the present invention, illustrating the number of network layers and unit composition. Figure 3 This is the data preprocessing flowchart of the present invention, describing the data cleaning and normalization steps. Figure 4 This is the flowchart of the model training and optimization of the present invention, showing the training process and parameter adjustment; Figure 5 This is an example diagram of the temperature prediction result of the present invention, presenting the comparison between the predicted value and the actual value. Detailed implementation manners
[0012] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.
[0013] Please refer to Figures 1 - 5 , the present invention provides a technical solution: the application of a temperature prediction method based on the LSTM algorithm in high-altitude precision agriculture, which is characterized in that it includes the following steps: S1: Data collection and preprocessing: Collect historical temperature data, meteorological data, and farmland environment data, and perform data cleaning and normalization processing; S2: LSTM model construction: Design and construct an LSTM network model, and determine the network structure and parameters; S3: Model training and optimization: Use the backpropagation algorithm to train the model, and adopt an optimizer to adjust the learning rate to prevent overfitting; S4: Temperature prediction and application: Use the trained model to predict future temperatures, and apply the prediction results to precision agriculture management.
[0014] Furthermore, data preprocessing includes removing outliers and processing data using standardization or normalization methods. Collect historical temperature data, meteorological data (such as humidity, wind speed), and farmland environment data (such as soil humidity). Remove missing values and outliers through data cleaning, and preprocess the data using standardization or normalization methods to ensure that the data meets the input requirements of the LSTM model.
[0015] Furthermore, the LSTM model uses the sliding window method to convert time series data into a supervised learning problem, designs and constructs an LSTM network model, and determines parameters such as the network structure (number of layers, number of hidden units), activation function, etc. Use the sliding window method to convert time series data into a supervised learning problem, and generate a training set and a test set.
[0016] Furthermore, the model is trained using the Adam optimizer, and early stopping and dropout techniques are used to prevent overfitting. The LSTM model is trained using the backpropagation algorithm, and the Adam optimizer is used to adjust the learning rate. Early stopping and dropout techniques are used to prevent overfitting. The mean squared error (MSE) is used as the loss function to evaluate the model performance.
[0017] Furthermore, the temperature prediction results are applied to irrigation management, pest control, and crop growth monitoring in precision agriculture. The trained LSTM model is used to predict future temperatures and generate prediction results. The results are applied to aspects such as irrigation management, pest control, and crop growth monitoring in precision agriculture to optimize production decisions. Embodiment
[0018] In this embodiment, the daily average temperature data of a certain farmland in the past 5 years is selected as the research object. First, the data is cleaned to remove missing values and outliers; then, the Min-Max normalization method is used to scale the data to the [0, 1] interval. The sliding window method (window size of 7 days) is used to generate the training set and the test set. A two-layer LSTM network is constructed, with 50 hidden units in each layer, and the activation function is tanh. During training, the Adam optimizer (learning rate 0.001) is used, the batch size is set to 32, and the number of training epochs is 100. Early stopping is used to prevent overfitting. After training, the model is used to predict the temperature for the next 7 days, compared with the actual values, and the MSE and R² metrics are calculated. The results show that the prediction error is low and the model performance is good. Embodiment
[0019] In this embodiment, in addition to temperature data, meteorological data such as humidity and wind speed are introduced as auxiliary features. A three-layer LSTM model is designed, with 100 hidden units in each layer, and a dropout layer (dropout rate = 0.2) is added to enhance the generalization ability. The data preprocessing and the sliding window method are the same as in Embodiment 1. The Adam optimizer is used in the training process, and finally the temperature prediction for the next 14 days is achieved. The prediction results are applied to the automatic adjustment of the farmland irrigation system, significantly improving the irrigation efficiency.
[0020] Finally, it should be noted that the above content is only used to illustrate the technical solution of the present invention, rather than to limit the protection scope of the present invention. Any simple modification or equivalent replacement of the technical solution of the present invention by those of ordinary skill in the art does not depart from the essence and scope of the technical solution of the present invention.
Claims
1. Application of the temperature prediction method based on the LSTM algorithm in high-altitude precision agriculture, characterized in that: Including the following steps: S1: Data collection and preprocessing: Collect historical temperature data, meteorological data, and farmland environment data, and perform data cleaning and normalization processing; S2: LSTM model construction: Design and construct an LSTM network model, and determine the network structure and parameters; S3: Model training and optimization: Use the backpropagation algorithm to train the model, and adopt an optimizer to adjust the learning rate to prevent overfitting; S4: Temperature prediction and application: Use the trained model to predict future temperatures, and apply the prediction results to precision agriculture management.
2. Application of the temperature prediction method based on the LSTM algorithm according to claim 1 in high-altitude precision agriculture, characterized in that: The data preprocessing includes removing outliers and processing the data using standardization or normalization methods.
3. Application of the temperature prediction method based on the LSTM algorithm according to claim 2 in high-altitude precision agriculture, characterized in that: The LSTM model uses a sliding window method to convert time series data into a supervised learning problem.
4. Application of the temperature prediction method based on the LSTM algorithm according to claim 3 in high-altitude precision agriculture, characterized in that: The model training uses the Adam optimizer, and prevents overfitting through early stopping and dropout techniques.
5. Application of the temperature prediction method based on the LSTM algorithm according to claim 4 in high-altitude precision agriculture, characterized in that: The temperature prediction results are applied to irrigation management, pest control, and crop growth monitoring in precision agriculture.