Crop yield prediction and synergism management system based on big data

Abstract

The invention relates to the technical field of crop production, in particular to a big data-based crop yield prediction and synergistic management system, which comprises a data acquisition module, a data processing module, a multi-scale crop growth prediction module and a plant stomata dynamic adjustment and transpiration monitoring module, the data acquisition module acquires multi-dimensional agricultural data; the data processing module receives and preprocesses the collected multi-dimensional agricultural data; the multi-scale crop growth prediction module predicts growth trends of crops in different time dimensions; and the plant stomata dynamic adjustment and transpiration monitoring module analyzes and predicts the transpiration rate of the crops, and adjusts the irrigation amount in real time in combination with the prediction result of the growth trend of the crops. According to the invention, the sustainability of water resource utilization is improved, the waste of agricultural resources is reduced, and finally the benefits of agricultural production and the sustainability of the environment are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of crop production, and in particular to a crop yield prediction and efficiency-enhancing management system based on big data. Background Art

[0002] With the growth of global population and the intensification of climate change, it is difficult to ensure the stable yield and quality of crops. In order to meet these challenges, modern agriculture is gradually developing in the direction of digitalization and intelligence. The crop management system based on big data has shown great potential in improving agricultural production efficiency and optimizing resource utilization. Through the real-time collection and processing of multi-dimensional agricultural data such as sensor networks, image analysis, and meteorological data, agricultural managers can understand the growth status of crops more comprehensively and accurately, and formulate precise management strategies, thereby improving crop yields and the benefits of agricultural production.

[0003] However, the existing agricultural management systems still have some shortcomings in practical applications. First, data collection often lacks systematicity and fails to fully integrate environmental data, soil data, plant physiological data, and crop growth image data, resulting in incomplete input data for the prediction model, affecting the prediction accuracy of the system. Secondly, when processing complex multimodal data, the existing technology has a relatively single analysis method and fails to achieve accurate identification and dynamic prediction of crop growth status, making it difficult to adjust management strategies in real time. In addition, irrigation management is often based on fixed preset rules and fails to combine the actual water demand and growth trend of crops, resulting in inefficient water resource utilization and an inability to effectively respond to dynamic demand changes during crop growth.

[0004] The present invention aims to overcome the deficiencies of the prior art and proposes a crop yield prediction and efficiency management system based on big data, which can greatly improve the accuracy of crop yield prediction, optimize the allocation of agricultural resources, and realize intelligent management and sustainable development of agricultural production. Summary of the invention

[0005] The present invention provides a crop yield prediction and efficiency-enhancing management system based on big data.

[0006] The crop yield prediction and efficiency management system based on big data includes a data acquisition module, a data processing module, a multi-scale crop growth prediction module, and a plant stomatal dynamic regulation and transpiration monitoring module, among which; The data acquisition module collects multi-dimensional agricultural data from the farmland environment in real time, including environmental data, soil data, plant physiological data, and image data of crop growth; The data processing module receives and pre-processes the collected multi-dimensional agricultural data, including data cleaning, noise reduction, and format conversion; The multi-scale crop growth prediction module predicts the growth trend of crops in different time dimensions (hours, days, months, seasons) and predicts the future growth status and yield of crops; The plant stomatal dynamic regulation and transpiration monitoring module uses stomatal monitoring sensors to obtain the real-time opening and closing status of crop stomata, combines environmental data, analyzes and predicts the transpiration rate of crops, and combines the predicted results of crop growth trends to adjust the irrigation volume in real time and optimize the utilization efficiency of water resources. Specifically, it includes: Stomata status monitoring: The stomata monitoring sensor can be used to obtain the opening and closing status of crop stomata in real time; Transpiration rate analysis and prediction: Combine the stomatal opening and closing status and environmental data to analyze the transpiration rate of crops and predict future transpiration rate changes; Irrigation regulation and water resource optimization: Based on the transpiration rate prediction results and combined with the prediction results of crop growth trends, the irrigation strategy is adjusted dynamically in real time.

[0007] Optionally, the data acquisition module includes: Environmental data collection: Collect environmental data through multiple environmental sensors, including temperature, humidity, wind speed, and light intensity; Soil data collection: Real-time monitoring of soil moisture, pH value, and nutrient content through soil sensors; Plant physiological data collection: Plant physiological data, including leaf temperature and chlorophyll content, are collected through plant physiological monitoring sensors; Image data collection of crop growth: Use multispectral cameras to obtain image data of crop growth and analyze the health and growth status of crops.

[0008] Optionally, the data processing module includes: Data cleaning: The mean filling method is used to clean up outliers and missing data in multidimensional agricultural data cleaning; Data denoising: Use sliding window filtering algorithm to denoise multidimensional agricultural data collected by sensors; Format conversion: Convert heterogeneous data collected by different sensors into a structured data table format.

[0009] Optionally, the multi-scale crop growth prediction module includes: Data integration: The pre-processed multi-dimensional agricultural data is integrated and standardized to form comprehensive data; Crop growth status recognition: Analyze the integrated data through the convolutional neural network (CNN) model to identify the current crop growth status; Crop growth status prediction: Based on the fused comprehensive data and the identified current growth status, predict the future growth status, demand and yield of crops.

[0010] Optionally, the data integration includes: Data fusion: A timestamp matching algorithm is used to align and fuse the environmental data, soil data, plant physiological data, and image data obtained from different sensors in time and space. For each data source, a unified timestamp is used. Collect data from different data sources Alignment is performed, where Indicates the type of data, including environmental, soil, plant physiological, or image data; Standardization: Standardize data of different dimensions to eliminate the impact of data dimensions.

[0011] Optionally, the convolutional neural network (CNN) model includes: Multimodal input processing of the convolution layer: Based on the fused comprehensive data, two-dimensional convolution is used to process image data, and one-dimensional convolution is used for environmental data, soil data, and plant physiological data to extract multimodal features; Multimodal feature fusion: The features extracted from image data and environmental data, soil data, and plant physiological data are fused through splicing operations to form a comprehensive feature tensor representation; Activation function and pooling operation: Apply the ReLU activation function and pooling operation to the fused feature tensor; Fully connected layer and output: The features after convolution, activation and pooling are input into the fully connected layer, and the growth status category or growth health score of the crop is output.

[0012] Optionally, the crop growth status prediction includes: Growth status and demand forecast: Based on the current crop growth status and comprehensive data, the exponential smoothing method is used to predict the future crop growth status and resource demand; Crop yield prediction: Combined with the predicted future growth status of crops, Bayesian regression is used to predict the future yield of crops.

[0013] Optionally, the pore status monitoring includes: Data detection: The stomatal monitoring sensor uses a laser interferometer to detect the opening and closing status of the stomata on the surface of crop leaves in real time and collect the physical parameters of the stomata, including stomatal conductance, stomatal size, and stomatal opening and closing frequency; Sensor signal conversion: The stomatal monitoring sensor converts the collected stomatal physical parameters into electrical signals, stomatal conductance It is calculated by detecting the transpiration rate and gas exchange amount of leaves under the state of stomatal opening and closing.

[0014] Optionally, the transpiration rate analysis and prediction includes: Data collection and fusion: Based on the real-time stomatal opening and closing state data (stomatal conductance ), combined with environmental data (air temperature , relative humidity , light intensity and wind speed ), forming a multi-dimensional input data set; Transpiration rate calculation: Based on the Penman-Monteith formula, calculate the current transpiration rate ; Transpiration rate prediction: Use exponential smoothing to predict future changes in transpiration rate.

[0015] Optionally, the irrigation regulation and water resource optimization includes: Water resource demand calculation: Calculate the water demand of crops in the future based on the predicted transpiration rate and growth trend forecast results; Dynamic adjustment of irrigation strategy: Based on the predicted water demand, combined with the current water resource availability of farmland and weather forecast, the irrigation strategy is dynamically adjusted through optimization algorithm to determine the optimal irrigation amount and prioritize water resources in different areas.

[0016] Beneficial effects of the present invention: The present invention obtains environment, soil, plant physiology and crop growth image data comprehensively and in real time through the data acquisition module, and combines the data cleaning, noise reduction and format conversion in the data processing module to ensure the quality and consistency of multi-dimensional agricultural data. By processing outliers and noise, the system can provide high-precision data support and accurately reflect the dynamic changes of farmland. This process effectively improves the accuracy of crop yield prediction, optimizes the utilization efficiency of agricultural resources, and significantly improves the level of intelligent agricultural production.

[0017] The present invention uses data integration and convolutional neural network models to accurately identify and predict the growth status of crops. Combined with the time series model and exponential smoothing method, the system can effectively predict future crop demand and yield, ensuring the scientific nature of the management strategy. At the same time, the plant stomatal dynamic regulation and transpiration monitoring module combines stomatal conductance, transpiration rate and environmental data to make real-time predictions on crop water demand, and dynamically adjusts the irrigation strategy through the irrigation regulation and water resource optimization module to ensure that crops obtain the best water supply at different growth stages, optimize the utilization efficiency of water resources, and increase crop yields.

[0018] The present invention, through stomatal status monitoring, transpiration rate analysis and prediction, the system can monitor the crop water demand in real time, and combined with the crop growth trend prediction, dynamically calculate the crop water requirement, adjust the irrigation strategy, and prevent excessive or insufficient irrigation. It not only ensures that the crops develop in the best growth environment and improves the crop yield and quality, but also greatly improves the sustainability of water resource utilization, reduces agricultural resource waste, and ultimately improves the efficiency of agricultural production and environmental sustainability. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 A schematic diagram of system function modules according to an embodiment of the present invention; Figure 2 Schematic diagram of a multi-scale crop growth prediction module according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0022] It should be noted that the references to "one embodiment", "embodiment", "exemplary embodiments", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0023] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0024] like Figure 1-Figure 2 As shown, the crop yield prediction and efficiency management system based on big data includes a data acquisition module, a data processing module, a multi-scale crop growth prediction module, and a plant stomatal dynamic regulation and transpiration monitoring module, among which; The data acquisition module collects multi-dimensional agricultural data from the farmland environment in real time, including environmental data, soil data, plant physiological data, and image data of crop growth; The data processing module receives and pre-processes the collected multi-dimensional agricultural data, including data cleaning, noise reduction, and format conversion to ensure data consistency and accuracy; The multi-scale crop growth prediction module predicts the growth trend of crops in different time dimensions (hours, days, months, seasons) and predicts the future growth status and yield of crops; The plant stomatal dynamic regulation and transpiration monitoring module uses stomatal monitoring sensors to obtain the real-time opening and closing status of crop stomata, and combines environmental data to analyze and predict the transpiration rate of crops. Combined with the predicted results of crop growth trends, the irrigation volume is adjusted in real time to optimize the utilization efficiency of water resources and ensure that the crop growth environment is in the best state. Specifically, it includes: Stomata status monitoring: The stomata monitoring sensor can be used to obtain the opening and closing status of crop stomata in real time; Transpiration rate analysis and prediction: Combine the stomatal opening and closing status and environmental data to analyze the transpiration rate of crops and predict future transpiration rate changes; Irrigation regulation and water resource optimization: Based on the transpiration rate prediction results and the prediction results of crop growth trends, the irrigation strategy is adjusted dynamically in real time to ensure that crops can obtain appropriate water at different growth stages; Through the above content, precise irrigation management and water resource optimization are achieved, and management strategies can be dynamically adjusted according to the growth status of crops and future needs, thereby improving the accuracy of crop yield prediction and resource utilization efficiency, ensuring that crops are always in the best growth environment, thereby improving the efficiency and sustainability of agricultural production.

[0025] The data acquisition module includes: Environmental data collection: Collect environmental data through multiple environmental sensors, including temperature, humidity, wind speed, and light intensity; Soil data collection: Real-time monitoring of soil moisture, pH value, and nutrient content through soil sensors; Plant physiological data collection: Plant physiological data, including leaf temperature and chlorophyll content, are collected through plant physiological monitoring sensors; Image data collection of crop growth: Use multispectral cameras to obtain image data of crop growth and analyze the health and growth status of crops; Through the above content, comprehensive and real-time collection of environmental data, soil data, plant physiological data and crop growth image data is achieved, providing all-round monitoring of the crop growth environment and health status, which can more accurately reflect the dynamic changes of farmland, provide high-precision data support for subsequent intelligent decision-making and management, improve the accuracy of crop yield prediction and production efficiency, optimize resource utilization, and enhance the level of intelligence in agricultural production.

[0026] The data processing module includes: Data cleaning: The mean filling method is used to clean up outliers and missing data in multidimensional agricultural data cleaning, which is expressed as: ; in, Indicates the first data points, is the number of valid data points; Data denoising: Use the sliding window filtering algorithm to denoise the multidimensional agricultural data collected by the sensor to reduce the impact of noise in the sensor data, which is expressed as: ; in, is the window size, The data value after noise reduction ensures the smoothness and reliability of the data; Format conversion: Convert heterogeneous data collected by different sensors into a structured data table format to ensure that all data has the same format; Through the above content, the quality and consistency of multidimensional agricultural data are ensured, the impact of outliers and noise on data analysis can be eliminated, the accuracy and completeness of the data are ensured, the system's prediction accuracy and resource optimization capabilities are improved, thereby significantly improving the efficiency and benefits of agricultural production.

[0027] The multi-scale crop growth prediction module includes: Data integration: The pre-processed multi-dimensional agricultural data is integrated to ensure that different data sources are synchronized in time and space, and comprehensive data is formed through standardized processing to provide unified input for subsequent analysis; Crop growth status recognition: Analyze the integrated data through the convolutional neural network (CNN) model to identify the current crop growth status; Crop growth status prediction: Based on the integrated data and the identified current growth status, predict the future growth status, demand and yield of crops; Through the above content, the synchronization and consistency of different data are ensured. State recognition uses CNN to accurately judge the current growth status of crops, and state prediction predicts future demand and output based on the time series model, which improves the scientific nature of management and resource utilization efficiency, helps increase crop yields and optimize agricultural decisions.

[0028] Data integration includes: Data fusion: A timestamp matching algorithm is used to align and fuse the environmental data, soil data, plant physiological data, and image data obtained from different sensors in time and space. For each data source, a unified timestamp is used. Collect data from different data sources Alignment is performed, where Indicates the type of data, including environmental, soil, plant physiological or image data. The fused data set is represented as: ; in, It's in time On the fusion of multidimensional data sets; Standardization: Standardize data of different dimensions to eliminate the impact of data dimensions, expressed as: ; in, Represents the original value of a data dimension. is the mean value of the dimension data, is the standard deviation, is the standardized data; Through the above content, agricultural data from different sources are integrated to form a consistent and structured comprehensive data set. Spatiotemporal alignment ensures the synchronization of different data in the same time dimension. Standardization processing eliminates the impact of different data dimensions, making various types of data comparable. It not only improves the consistency and accuracy of the data, but also provides high-quality data support for subsequent crop growth status identification and prediction, thereby improving the overall performance of the system and prediction accuracy.

[0029] The Convolutional Neural Network (CNN) model consists of: Multimodal input processing of the convolution layer: Based on the fused comprehensive data, two-dimensional convolution is used to process the image data, and one-dimensional convolution is used for the environmental data, soil data, and plant physiological data to extract multimodal features, which are expressed as: ; in, After convolution, at position The image feature output value on (the output feature map of the convolutional layer), is the two-dimensional convolution kernel of the image data, which is used to extract image features in the local area and has a size of , , Is the index parameter of the convolution kernel, which indicates the range of movement of the convolution kernel on the image in the two-dimensional convolution operation. is the pixel value covered by the convolution kernel in the image, is the convolution bias term, which is applied to the image convolution calculation to adjust the output. is the size range of the convolution kernel, that is, the span of the convolution kernel in the local area; ; in, is the convolution at time The output feature value of the point (convolution feature for non-image data), is a one-dimensional convolution kernel for non-image data, with size , applied to time series data to extract features, For in time Input non-image data at any time (such as environmental data, soil data, plant physiological data), is the bias term for the convolution of non-image data, is the size range of the one-dimensional convolution kernel, defining the local convolution area in the time dimension; Multimodal feature fusion: The features extracted from image data and environmental data, soil data, and plant physiological data are fused through splicing operations to form a comprehensive feature tensor representation, which is expressed as: ; in, is the fused multimodal feature, which represents the combination of all features extracted from image and non-image data. is the image feature extracted after convolution, is the environmental data feature extracted after convolution, is the soil data feature extracted after convolution, is the plant physiological data feature extracted after convolution, Represents a concatenation operation that combines features from different data sources into an overall feature tensor; Activation function and pooling operation: Apply the ReLU activation function and pooling operation to the fused feature tensor to extract effective features, reduce the amount of calculation and prevent overfitting, expressed as: ; in, is the output after activation, For the location The fusion eigenvalues ​​on ; ; in, For the location The pooled output value on is the activation value within the pooling window; Fully connected layer and output: The features after convolution, activation and pooling are input into the fully connected layer, and the growth status category or growth health score of the crop is output, which is expressed as: ; in, is the weight of the fully connected layer, is the pooling result, For output (e.g. crop growth status classification or health score), To transform the pooled feature tensor Flatten into a one-dimensional vector for easy input into the fully connected layer; Through the above content, multi-dimensional data sources can be analyzed simultaneously to accurately identify the current growth status of crops. The convolution operation extracts the key features of image and non-image data. The activation function and pooling operation further improve the efficiency and robustness of the model. Finally, the growth status or health score of the crop is output through the fully connected layer. It can not only accurately identify the health status of crops, but also flexibly adapt to complex and changeable agricultural environmental data, thereby improving the prediction accuracy and resource utilization efficiency of the system.

[0030] Crop growth status forecast includes: Growth status and demand forecast: Based on the current crop growth status and comprehensive data, the exponential smoothing method is used to predict the future crop growth status and resource demand, expressed as: ; in, For time The smoothed value of , that is, the predicted growth state, For time The actual value (current growth status or comprehensive data), is the smoothing coefficient, and its value range is , For time The smoothing value of ; in, To predict future time points through recursive calculation The growth status and needs of crops; Crop yield prediction: Combined with the predicted future growth status of crops, Bayesian regression is used to predict the future yield of crops, expressed as: ; in, For time The crop yields To integrate data and predict growth status, is the weight coefficient of the Bayesian regression model (obtained through Bayesian inference), is the error term (subject to normal distribution); Through the above content, it is possible to quickly and accurately predict the future growth status and resource requirements of crops, effectively capture trends and seasonal changes in the data, and maintain a high level of prediction accuracy. It not only provides accurate yield forecasts, but also quantifies the uncertainty of the prediction results, providing agricultural managers with a more robust basis for decision-making.

[0031] Stomatal status monitoring includes: Data detection: The stomatal monitoring sensor uses a laser interferometer to detect the opening and closing status of the stomata on the surface of crop leaves in real time and collect the physical parameters of the stomata, including stomatal conductance, stomatal size, and stomatal opening and closing frequency; Sensor signal conversion: The stomatal monitoring sensor converts the collected stomatal physical parameters into electrical signals, stomatal conductance The transpiration rate and gas exchange rate of leaves are calculated by detecting the state of stomatal opening and closing, which is expressed as: ; in, is the stomatal conductance, is the transpiration rate, is the leaf water vapor pressure difference; Through the above content, using parameters such as stomatal conductance and transpiration rate, the water demand of crops can be accurately reflected, which can quickly respond to environmental changes, optimize irrigation strategies, prevent excessive or insufficient irrigation, thereby improving water resource utilization efficiency, ensuring that crops are in the best growth state, and ultimately improving the efficiency and sustainability of agricultural production.

[0032] Transpiration rate analysis and prediction include: Data collection and fusion: Based on the real-time stomatal opening and closing state data (stomatal conductance ), combined with environmental data (air temperature , relative humidity , light intensity and wind speed ), forming a multidimensional input dataset , used for analysis and prediction of transpiration rate; Transpiration rate calculation: Based on the Penman-Monteith formula, calculate the current transpiration rate , expressed as: ; in, is the transpiration rate , Net radiation , is the soil heat flux , is the air density , is the specific heat of air , is the saturated vapor pressure deficit (kPa), which is related to temperature and humidity. is the air resistance of the meteorological environment , Stomatal resistance of crops , and stomatal conductance Related, is the slope of the saturated vapor pressure curve , is the humidity constant; Transpiration rate prediction: Use exponential smoothing method to predict future transpiration rate changes, expressed as: ; in, For future time The predicted value of transpiration rate is is the transpiration rate at the current moment, is the smoothing coefficient, which controls the impact of historical data on the forecast. , For in time The predicted value of transpiration rate at the time; Through the above content, the water demand of crops can be reflected in real time and accurately, helping agricultural managers optimize irrigation strategies, avoid excessive or insufficient irrigation, and improve water resource utilization efficiency. At the same time, predicting future changes in transpiration rate provides a reliable basis for agricultural decision-making, ensuring that crops are always in the best growth conditions, and ultimately improving crop yield and quality.

[0033] Irrigation regulation and water resource optimization include: Water resource demand calculation: Based on the transpiration rate prediction value and growth trend prediction results, the water demand of crops in the future is calculated, expressed as: ; in, For future time The crop water requirement, is the crop coefficient of the crop (related to the crop growth stage), is the predicted transpiration rate, is the farmland area; Dynamic adjustment of irrigation strategy: Based on the predicted water demand, combined with the current water resource availability of farmland and weather forecast, the irrigation strategy is dynamically adjusted through the optimization algorithm to determine the optimal irrigation amount and prioritize water resources in different areas. The optimal irrigation amount is expressed as: ; in, For future time The actual irrigation volume, is the amount of water resources available; Through the above content, it is possible to dynamically calculate future water requirements and adjust irrigation strategies in real time to ensure that crops receive appropriate water supply. This not only optimizes the utilization efficiency of water resources, but also ensures that crops at different growth stages receive the best growth environment through intelligent irrigation control, thereby increasing crop yields and the sustainability of water resource utilization.

[0034] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0035] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A crop yield prediction and efficiency management system based on big data, characterized by: It includes data acquisition module, data processing module, multi-scale crop growth prediction module and plant stomatal dynamic regulation and transpiration monitoring module, among which; The data acquisition module collects multi-dimensional agricultural data from the farmland environment in real time, including environmental data, soil data, plant physiological data, and image data of crop growth; The data processing module receives and pre-processes the collected multi-dimensional agricultural data, including data cleaning, noise reduction, and format conversion; The multi-scale crop growth prediction module predicts the growth trend of crops in different time dimensions and predicts the future growth status and yield of crops; The plant stomatal dynamic regulation and transpiration monitoring module uses stomatal monitoring sensors to obtain the real-time opening and closing status of crop stomata, combines environmental data, analyzes and predicts the transpiration rate of crops, and combines the predicted results of crop growth trends to adjust the irrigation volume in real time and optimize the utilization efficiency of water resources. Specifically, it includes: Stomata status monitoring: The stomata monitoring sensor can be used to obtain the opening and closing status of crop stomata in real time; Transpiration rate analysis and prediction: Combine the stomatal opening and closing status and environmental data to analyze the transpiration rate of crops and predict future transpiration rate changes; Irrigation regulation and water resource optimization: Based on the transpiration rate prediction results and combined with the prediction results of crop growth trends, the irrigation strategy is adjusted dynamically in real time.

2. The crop yield prediction and efficiency management system based on big data according to claim 1 is characterized in that: The data acquisition module comprises: Environmental data collection: Collect environmental data through multiple environmental sensors, including temperature, humidity, wind speed, and light intensity; Soil data collection: Real-time monitoring of soil moisture, pH value, and nutrient content through soil sensors; Plant physiological data collection: Plant physiological data, including leaf temperature and chlorophyll content, are collected through plant physiological monitoring sensors; Image data collection of crop growth: Use multispectral cameras to obtain image data of crop growth and analyze the health and growth status of crops.

3. The crop yield prediction and efficiency management system based on big data according to claim 1 is characterized in that: The data processing module comprises: Data cleaning: The mean filling method is used to clean up outliers and missing data in multidimensional agricultural data cleaning; Data denoising: Use sliding window filtering algorithm to denoise multidimensional agricultural data collected by sensors; Format conversion: Convert heterogeneous data collected by different sensors into a structured data table format.

4. The crop yield prediction and efficiency management system based on big data according to claim 1, characterized in that: The multi-scale crop growth prediction module includes: Data integration: The pre-processed multi-dimensional agricultural data is integrated and standardized to form comprehensive data; Crop growth status recognition: Analyze the integrated data through the convolutional neural network model to identify the current crop growth status; Crop growth status prediction: Based on the fused comprehensive data and the identified current growth status, predict the future growth status, demand and yield of crops.

5. The crop yield prediction and efficiency management system based on big data according to claim 4 is characterized in that: The data integration includes: Data fusion: A timestamp matching algorithm is used to align and fuse the environmental data, soil data, plant physiological data, and image data obtained from different sensors in time and space. For each data source, a unified timestamp is used. Collect data from different data sources Alignment is performed, where Indicates the type of data, including environmental, soil, plant physiological, or image data; Standardization: Standardize data of different dimensions to eliminate the impact of data dimensions.

6. The crop yield prediction and efficiency management system based on big data according to claim 5 is characterized in that: The convolutional neural network model includes: Multimodal input processing of the convolution layer: Based on the fused comprehensive data, two-dimensional convolution is used to process image data, and one-dimensional convolution is used for environmental data, soil data, and plant physiological data to extract multimodal features; Multimodal feature fusion: The features extracted from image data and environmental data, soil data, and plant physiological data are fused through splicing operations to form a comprehensive feature tensor representation; Activation function and pooling operation: Apply the ReLU activation function and pooling operation to the fused feature tensor; Fully connected layer and output: The features after convolution, activation and pooling are input into the fully connected layer, and the growth status category or growth health score of the crop is output.

7. The crop yield prediction and efficiency management system based on big data according to claim 6 is characterized in that: The crop growth status prediction includes: Growth status and demand forecast: Based on the current crop growth status and comprehensive data, the exponential smoothing method is used to predict the future crop growth status and resource demand; Crop yield prediction: Combined with the predicted future growth status of crops, Bayesian regression is used to predict the future yield of crops.

8. The crop yield prediction and efficiency management system based on big data according to claim 1, characterized in that: The pore status monitoring includes: Data detection: The stomatal monitoring sensor uses a laser interferometer to detect the opening and closing status of the stomata on the surface of crop leaves in real time, and collects the physical parameters of the stomata, including stomatal conductance, stomatal size, and stomatal opening and closing frequency. Sensor signal conversion: The stomatal monitoring sensor converts the collected stomatal physical parameters into electrical signals, stomatal conductance, and stomatal It is calculated by detecting the transpiration rate and gas exchange amount of leaves under the state of stomatal opening and closing.

9. The crop yield prediction and efficiency management system based on big data according to claim 8, characterized in that: The transpiration rate analysis and prediction includes: Data collection and fusion: Based on the acquired real-time stomatal opening and closing status data, combined with environmental data, a multi-dimensional input data set is formed; Transpiration rate calculation: Based on the Penman-Monteith formula, calculate the current transpiration rate ; Transpiration rate prediction: Use exponential smoothing to predict future changes in transpiration rate.

10. The crop yield prediction and efficiency management system based on big data according to claim 9, characterized in that: The irrigation regulation and water resource optimization include: Water resource demand calculation: Calculate the water demand of crops in the future based on the predicted transpiration rate and growth trend forecast results; Dynamic adjustment of irrigation strategy: Based on the predicted water demand, combined with the current water resource availability of farmland and weather forecast, the irrigation strategy is dynamically adjusted through optimization algorithm to determine the optimal irrigation amount and prioritize water resources in different areas.

Images