In-greenhouse environment remote monitoring method applied to greenhouse planting

Abstract

The invention discloses an in-greenhouse environment remote monitoring method applied to greenhouse planting, and relates to the technical field of big data, and the method comprises the following steps: building a sensor network for environment parameter collection; preprocessing the environmental data; performing multi-dimensional environment analysis according to the cleaning data; performing environment modeling according to an analysis result; performing temperature and humidity prediction and adjustment suggestion on the environment model; performing automatic control according to the adjustment suggestions; performing system optimization on an automatic control result; and an environment parameter acquisition result is fed back in real time. The invention provides an improved multi-layer sensor (MLP) network structure, and a multi-dimensional input and weighting mechanism is introduced, so that the model can more accurately capture a complex nonlinear relationship among greenhouse environment parameters. Especially in the face of rapid change of environmental factors such as temperature and humidity, the improved MLP network can flexibly adjust the weight of the model, and the accuracy of temperature and humidity change trend prediction is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of big data, and specifically to a method for remotely monitoring the environment inside a greenhouse applied to greenhouse cultivation. Background Art

[0002] With the continuous development of modern agricultural technology, especially in the field of greenhouse cultivation, the refinement and intelligence of environmental control technology have become the key to improving the growth efficiency and quality of crops. The monitoring and regulation system of the greenhouse cultivation environment usually optimizes environmental parameters such as temperature, humidity, air flow, light, and carbon dioxide inside the greenhouse by integrating technologies such as sensors, data acquisition, real-time monitoring, and environmental regulation.

[0003] In order to achieve the intelligence of environmental regulation, researchers have gradually applied machine learning algorithms such as multi-layer perceptron (MLP) and long short-term memory network (LSTM) for data modeling and prediction in recent years. However, the existing technology still faces some challenges, especially in the prediction of temperature and humidity change trends and the modeling of complex relationships between environmental factors, where there are obvious deficiencies.

[0004] In the existing technology, the deficiencies of the multi-layer perceptron (MLP) network structure: Currently, most applications of MLP-based models in greenhouse environment monitoring and regulation usually adopt a standard feedforward neural network structure to directly perform input-output mapping from sensor data. Although this structure can capture the basic relationship between environmental factors and crop growth, it has limitations in dealing with highly nonlinear, complex time series changes, and the interaction between environmental factors. The existing MLP models cannot fully explore the time series dependence relationship between multi-dimensional environmental parameters, resulting in insufficient prediction accuracy. Especially when the environmental parameters change violently, the robustness and adaptability of the model are poor. Summary of the Invention

[0005] Aiming at the deficiencies of the existing technology, the present invention provides a method for remotely monitoring the environment inside a greenhouse applied to greenhouse cultivation to solve the problems raised in the above background art.

[0006] To achieve the above object, the present invention provides the following technical solutions: In a first aspect, an embodiment of the present invention provides a method for remotely monitoring the environment inside a greenhouse applied to greenhouse cultivation, including the following steps: S1. Build a sensor network for collecting environmental parameters; Establish an Internet of Things-based sensor network for collecting environmental parameters inside the greenhouse to obtain environmental data; S2. Preprocess the environmental data; Correct the environmental data through data preprocessing technology to obtain cleaned data; S3. Conduct multi-dimensional environmental analysis based on the cleaned data; Use principal component analysis technology to identify the relationships between environmental data in the greenhouse and obtain the analysis results; S4. Conduct environmental modeling based on the analysis results; Build a prediction model between the environment in the greenhouse and crop growth based on the multi-layer perceptron MLP network structure model to obtain the environmental model; S5. Make temperature and humidity predictions and adjustment suggestions for the environmental model; Based on the long short-term memory network LSTM, make temperature and humidity predictions for the environmental model and generate adjustment suggestions to obtain the adjustment suggestions; S6. Conduct automatic control according to the adjustment suggestions; Use an automatic control system based on the industrial automation communication protocol Modbus protocol to remotely control the temperature and humidity adjustment equipment to obtain the automatic control results; S7. Collect real-time environmental parameters based on the automatic control results and provide real-time feedback on the collection results.

[0007] Further optimize this technical solution. In step S1, the environmental parameters include: Temperature and humidity, air flow velocity, light intensity, and carbon dioxide concentration.

[0008] Further optimize this technical solution. In step S2, the data preprocessing technologies include: Data imputation: Fill in the missing values with the mean, median, or mode of the environmental data; Outlier detection: Calculate the standard score of the environmental data points. If the difference between an environmental data point and the mean is greater than two or three standard deviations, then this point is considered an outlier; Data smoothing: A filtering technology to reduce the impact of noise, which reduces noise by calculating the average value within the window around the environmental data points.

[0009] Further optimize this technical solution. In step S3, the principal component analysis technology includes: Find the direction with the largest variance in the cleaned data to extract the main features. The type of environmental parameters with the largest variance in the cleaned data has the greatest impact on crop growth.

[0010] Further optimize this technical solution. In step S4, the multi-layer perceptron MLP network structure model includes: Establish an environmental model of the environmental parameters in the greenhouse on crop growth, introduce a time series weighted mechanism, and conduct temperature and humidity prediction and analysis.

[0011] Further optimize this technical solution. The establishment of the environmental model of the environmental parameters in the greenhouse on crop growth includes: ; Among them, is the output of the crop growth state at time ; is the input environmental parameter, including temperature , humidity , air flow velocity , light intensity and carbon dioxide concentration at time ; is the preset value, representing the time-varying weight coefficient of each input parameter, and the relative influence of each environmental factor on crop growth at different time points; is the bias term, representing the basic output value of the network, with an initial value of 0; is the non-linear activation function, and its function expression is .

[0012] To further optimize this technical solution, the introduced time series weighting mechanism includes: Introduce a time window parameter, so that each input not only depends on the data at the current moment, but also depends on the environmental data within the historical time, which is expressed by the following formula: ; Among them, is the length of the time window, indicating that the current crop growth state not only depends on the environmental data at the current moment, but also depends on the environmental data in the past moments; represents the weighted sum of the environmental parameters within the past moments; is the output of the crop growth state within the past moments.

[0013] To further optimize this technical solution, the temperature and humidity prediction and analysis include: Data collection: Obtained by cleaning data and analyzing results, and recorded in time series; Model training: Input the collected data and train using the multi-layer perceptron MLP network structure model to learn the non-linear relationship between the input environmental factors and the temperature and humidity changes; Temperature and humidity prediction: After training, use the new real-time environmental data including the temperature and humidity, air flow velocity, and light intensity at the current moment to obtain the temperature and humidity prediction results for the future moment; Environmental adjustment suggestion: According to the predicted temperature and humidity prediction results, adjust the temperature and humidity control equipment in the greenhouse to maintain a suitable crop growth environment.

[0014] To further optimize this technical solution, in step S5, the long short-term memory network LSTM includes: Obtain the environmental prediction model in the greenhouse using the multi-layer perceptron MLP network structure model; Multi-dimensional joint processing is adopted. Joint modeling is performed on all environmental factors and the environmental prediction model is used. By using activation functions for each environmental factor to achieve this, and these functions are the memory units of the long short-term memory network LSTM.

[0015] A time window weighting mechanism is adopted, that is, the historical data within each time window will be weighted according to the influence degree of the environmental factor. The variance is calculated to determine the influence of this environmental factor on the crop growth environment. The larger the variance, the greater the influence.

[0016] To further optimize this technical solution, in step S7, the sensor network established through step S1 is used to collect the environmental parameters in the greenhouse in real time, compare the gap between the result of the equipment adjustment after automatic control and the actual environmental parameters, and provide a reference basis for the adjustment of the equipment.

[0017] In a second aspect, an embodiment of the present invention provides a computer device, including a memory and a processor. The memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of a method for remote monitoring of the greenhouse environment applied to greenhouse cultivation as described in the first aspect of the present invention are implemented.

[0018] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, wherein: when the computer program instructions are executed by the processor, the steps of a method for remote monitoring of the greenhouse environment applied to greenhouse cultivation as described in the first aspect of the present invention are implemented.

[0019] Compared with the prior art, the present invention provides a method for remote monitoring of the greenhouse environment applied to greenhouse cultivation, having the following beneficial effects: The method for remote monitoring of the greenhouse environment applied to greenhouse cultivation proposes an improved multi-layer perceptron (MLP) network structure. By introducing multi-dimensional inputs and a weighting mechanism, the model can more accurately capture the complex non-linear relationships between greenhouse environmental parameters (such as temperature and humidity, air flow velocity, light intensity, carbon dioxide concentration, etc.). In this improved structure, by jointly modeling multi-dimensional environmental factors, the network can more comprehensively understand the interaction effects between different environmental factors. Especially when facing rapid changes in environmental factors such as temperature and humidity, the improved MLP network can flexibly adjust the weights of the model, improving the accuracy of predicting the trends of temperature and humidity changes. This not only enhances the prediction accuracy of the model but also improves the real-time performance and stability of greenhouse environment regulation, effectively optimizing the crop growth environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0021] Figure 1 It is a schematic flow chart of a method for remote monitoring of the greenhouse environment applied to greenhouse cultivation proposed by the present invention; Figure 2 It is a flow chart of temperature and humidity prediction and analysis of a method for remote monitoring of the greenhouse environment applied to greenhouse cultivation proposed by the present invention; Figure 3 It is a schematic diagram of the usage method and explanation of the multi-layer perceptron MLP network structure model formula model of a method for remote monitoring of the greenhouse environment applied to greenhouse cultivation proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will provide a detailed description of the specific embodiments of the present invention in conjunction with the drawings in the specification.

[0023] Many specific details are set forth in the following description in order to provide a thorough understanding of the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the spirit of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0024] Second, the "one embodiment" or "embodiment" referred to herein means a specific feature, structure, or characteristic that may be included in at least one implementation manner of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or selectively exclusive embodiments from other embodiments.

[0025] Embodiment 1: Referring to Figures 1 to 3 , which is the first embodiment of the present invention. This embodiment provides a method for remote monitoring of the greenhouse environment applied to greenhouse cultivation, including the following steps: S1. Build a sensor network for environmental parameter collection; Establish an Internet of Things-based sensor network for collecting environmental parameters in the greenhouse to obtain environmental data.

[0026] The environmental parameters include: Temperature and humidity, air flow velocity, light intensity, and carbon dioxide concentration.

[0027] In this embodiment, through the sensor data acquisition module, various environmental data in the greenhouse are transmitted to the central processing unit in real time. These data are crucial for subsequent environmental analysis and model construction. The LoRaWAN network technology is used, which can transmit data over a large range and has the advantages of low power consumption and long-distance transmission, especially suitable for the long-term operation of equipment in the greenhouse environment.

[0028] S2. Preprocess the environmental data; Correct the environmental data through data preprocessing technology to obtain cleaned data. The data preprocessing technology includes: Data imputation: Fill in the missing values with the mean, median, or mode of the environmental data; Outlier detection: Calculate the standard score of the environmental data points. If the difference between the environmental data points and the mean is greater than two or three standard deviations, then this point is considered an outlier; Data smoothing: A filtering technology to reduce the influence of noise, which reduces noise by calculating the average value within the window around the environmental data points.

[0029] In this embodiment, the environmental data collected by the sensors will face problems such as noise, missing values, and outliers. Therefore, it is necessary to preprocess and clean the data.

[0030] S3. Conduct multi-dimensional environmental analysis based on the cleaned data; Use principal component analysis technology to identify the mutual relationships between the environmental data in the greenhouse to obtain the analysis results.

[0031] The principal component analysis technology includes: Find the direction with the largest variance in the cleaned data to extract the main features. The environmental parameters with the largest variance in the cleaned data have the greatest impact on crop growth.

[0032] In this embodiment, by analyzing the correlation between temperature and humidity, air flow velocity and crop growth, it is possible to find out which environmental factors have the greatest impact on crop growth. Using the principal component analysis (PCA) technique, the multi-dimensional data is reduced in dimension, the main influencing factors are found, and input data is provided for the next intelligent model. Through PCA, the computational amount can be effectively reduced while ensuring the accuracy of the analysis results.

[0033] S4. Perform environmental modeling according to the analysis results; Based on the multi-layer perceptron MLP network structure model, a prediction model between the environment in the greenhouse and crop growth is constructed to obtain the environmental model.

[0034] The multi-layer perceptron MLP network structure model includes: Establish an environmental model of the environmental parameters in the greenhouse on crop growth: ; Among them, is the output of the crop growth state at time ; is the input environmental parameter, including temperature , humidity , air flow velocity , light intensity and carbon dioxide concentration at time ; is a preset value, representing the time-varying weight coefficient of each input parameter, the relative impact of each environmental factor on crop growth at different time points; is the bias term, representing the basic output value of the network, with an initial value of 0; is a non-linear activation function, and its function expression is .

[0035] Introduce a time series weighting mechanism, so that each input depends not only on the data at the current moment but also on the environmental data within the historical time.

[0036] It is expressed by the following formula: ; Among them, is the length of the time window, indicating that the crop growth state at the current moment depends not only on the environmental data at the current moment but also on the environmental data at past moments; represents the weighted sum of environmental parameters within the past moments; is the output of the crop growth state within the past moments.

[0037] Temperature and humidity prediction and analysis.

[0038] Data collection: Obtained by cleaning data and analysis results and recorded in time series; Model training: Input the collected data and use the multi-layer perceptron MLP network structure model for training to learn the non-linear relationship between input environmental factors and temperature and humidity changes; Temperature and humidity prediction: After the model training is completed, predict the temperature and humidity at future moments based on new real-time environmental data. The new real-time environment includes the temperature and humidity, air flow velocity, and light intensity at the current moment; Environmental regulation suggestions: Adjust the temperature and humidity control equipment in the greenhouse according to the predicted temperature and humidity change trend to maintain a suitable crop growth environment.

[0039] In this embodiment, the MLP network structure model can establish a non-linear relationship by combining environmental parameters (temperature and humidity, air flow, light intensity, carbon dioxide concentration) and crop growth states (such as growth stage, leaf area, etc.). Different from the traditional linear regression model, the MLP network structure model can capture complex non-linear relationships, so it can more accurately reflect the impact of environmental changes on crop growth. This model provides data support for the subsequent decision-making support system and can predict the growth changes of crops when environmental parameters change.

[0040] The usage method of the multi-layer perceptron MLP network structure model is as follows: Input end setting: The input end of this model includes four main environmental factors: temperature and humidity, air flow velocity, light intensity, and carbon dioxide concentration. Each factor is collected at each time point to form a time series data set. These data are input into the model after being standardized.

[0041] Dynamic weighting mechanism: In the traditional MLP network structure model, the input weights are fixed, while in this model, the weights are time-varying, which means that the influence degree of each environmental factor on crop growth will be different at different moments. Through learning historical data, these weights can be dynamically adjusted to better reflect the real-time impact of the environment on crop growth.

[0042] Time window mechanism: Considering the long-term impact of environmental changes on crop growth, a time window is introduced into the model so that the environmental data at each moment is not only related to the current crop growth state, but also interacts with the environmental data over a period of time in the past to capture the influence of long-term trends and short-term fluctuations.

[0043] Output and decision support: The values output by the network (crop growth state) can provide data support for the environmental control system in the greenhouse, and adjust environmental parameters according to the predicted crop growth state. For example, if the model predicts poor crop growth (i.e., low Yt), the system can automatically adjust environmental factors such as temperature, humidity, and light to optimize crop growth conditions.

[0044] If you want to use this model for predicting the trend of temperature and humidity changes, it can be extended and applied in the following ways: Only need to use environmental factor inputs (e.g., temperature, humidity, air flow velocity, etc.) and modify the output target to the future predicted values of temperature and humidity. Two strategies can be adopted: Predict temperature and humidity changes separately: Predict the future trend of temperature and humidity changes by inputting environmental factors (such as air flow velocity, light intensity, carbon dioxide concentration) into the model. The output of the model can be directly the predicted temperature and humidity values.

[0045] Joint prediction of crop growth and temperature and humidity: If the impact of both crop growth and temperature and humidity changes on environmental control needs to be considered simultaneously, a multi-task learning model can be designed. Based on the model of the impact of environmental parameters in the greenhouse on crop growth, add the prediction target of temperature and humidity changes so that the model can output both the crop growth state and the environmental change trend.

[0046] Specific methods and implementation: Input end adjustment: Environmental parameter input: For predicting the trend of temperature and humidity changes, the input end will mainly include historical data of temperature and humidity (observed values of temperature and humidity at several past moments) and other relevant environmental parameters (such as air flow velocity, light intensity, carbon dioxide concentration, etc.) to form a time series data set. Crop growth state: During the prediction process, the crop growth state can be used as an auxiliary input to help understand the impact of the environment on the crop and thus reflect the long-term trend of environmental changes.

[0047] When using the environmental model for predicting the trend of temperature and humidity changes, the model formula needs to be adjusted and applied as follows: According to the content in step S4 above, the environmental model of the environmental parameters in the greenhouse on crop growth is ; On this basis, Adjustment of the target output: The output of the model Directly set as the predicted values of temperature and humidity, that is, predict the temperature and humidity at future moments. For example, if predicting temperature, then: ; Among them, is the predicted temperature at the next moment. Similarly, the humidity prediction can be expressed similarly as: ; Among them, is the predicted humidity at the next moment.

[0048] Introduce time series characteristics: Since temperature and humidity have temporal properties, a historical data window (such as data at the past K moments) can be introduced as input to better capture the trend of temperature and humidity changes. For example, the prediction formula for temperature can be extended to: ; Among them, represents data such as temperature and humidity, air flow velocity, etc. at the past moments.

[0049] Model output and temperature and humidity change trends: After training, the model will be able to predict the temperature and humidity values at future moments based on current and past environmental factors (temperature and humidity, air flow velocity, light intensity, carbon dioxide concentration, etc.). This output can provide prediction information for the greenhouse environment control system, guiding the system on how to adjust environmental conditions at a future moment to maintain the best crop growth state.

[0050] Traditional temperature and humidity prediction methods usually rely on linear regression or statistical models and are difficult to handle complex environmental changes and non - linear relationships. By adopting the MLP network structure model in the present invention, the non - linear relationship between environmental factors and temperature and humidity can be better modeled, thereby improving the accuracy of prediction. Introducing environmental data over a past period into the model and considering the temporal changes of the environment enable the model to capture the trend and periodic characteristics of temperature and humidity changes, thus improving the reliability of prediction. By comprehensively considering multiple environmental factors (such as temperature and humidity, air flow velocity, etc.) and using a weighting mechanism to handle the importance of each factor, the model can be made more flexible, accurate, and adaptable to the complexity of the greenhouse environment.

[0051] S5. Perform temperature and humidity prediction and adjustment suggestions on the environmental model; Based on the long short - term memory network LSTM, perform temperature and humidity prediction on the environmental model and generate adjustment suggestions to obtain the adjustment suggestions.

[0052] The long short - term memory network LSTM includes: Use the multi - layer perceptron MLP network structure model to obtain the environmental prediction model inside the greenhouse.

[0053] Multi-dimensional joint processing is adopted to jointly model all environmental factors. By using activation functions for each environmental factor This is achieved, and these functions are the memory units of the long short-term memory network (LSTM). It can model the temporal dependence relationship of multi-dimensional data. This joint modeling method can capture the interactions between different environmental factors, such as the joint effect between temperature and humidity and light intensity, the correlation between air flow velocity and carbon dioxide concentration, etc., thereby improving the prediction accuracy.

[0054] A time window weighting mechanism is adopted, that is, the historical data within each time window will be weighted according to the influence degree of the environmental factor. By calculating the variance to determine the influence of this environmental factor on the crop growth environment, the larger the variance, the greater the influence. For example, when the temperature and humidity fluctuate greatly, the model will automatically increase the weight of the temperature and humidity input, making its influence on the prediction result more significant; conversely, if the change in air flow velocity is small, its influence on the prediction result will be relatively weak. This mechanism enables the model to adaptively adjust the input weight at each moment, enhancing the robustness of the model in a changing environment.

[0055] In this embodiment, calculations are performed according to the model formula through the following steps: Input end setting: Collect and preprocess the historical data of temperature and humidity, air flow velocity, light intensity, and carbon dioxide concentration in the greenhouse. Assume that the time series length of each environmental factor is K, then the input of each environmental factor at time t is a vector of length K, containing the data of the past K moments. For example, the input of temperature and humidity is The input of air flow velocity is , and so on.

[0056] LSTM network training: Use the historical data of environmental factors in the greenhouse to train the LSTM model, and update the weights and bias terms in the network through the backpropagation algorithm (by calculating the gradient of the loss function with respect to the network parameters to update these parameters, aiming to minimize the loss function). The network will automatically learn the non-linear relationships between different environmental factors and dynamically adjust the weights of the input data through the time window weighting mechanism, thereby accurately predicting the changes in future environmental variables such as temperature and humidity.

[0057] Prediction and adjustment suggestions: After training, input the real-time environmental data into the trained LSTM network, and the model will output the prediction results of temperature and humidity, air flow velocity, or light intensity in the future for a period of time. According to the prediction results, the model can generate corresponding adjustment suggestions, such as automatically adjusting the temperature and humidity control equipment, increasing or decreasing the light, etc., to maintain the optimal crop growth environment.

[0058] S6. Perform automated control according to the adjustment suggestions; An automated control system based on the Modbus protocol, an industrial automation communication protocol, is used to remotely control the temperature and humidity adjustment equipment to obtain an automated control result.

[0059] In this embodiment, the temperature and humidity adjustment equipment comes with an automated control system when it leaves the factory. The Modbus protocol in this system is a common industrial automation communication protocol that can achieve interconnection and interoperability between devices. The temperature and humidity adjustment equipment (such as humidifiers, dehumidifiers, fans, etc.) exchanges real-time data with the control system through the Modbus protocol and automatically executes adjustment operations according to the adjustment suggestions. When the temperature and humidity adjustment equipment is a humidifier, this automated control system is used to increase humidity; when the temperature and humidity adjustment equipment is a dehumidifier, this automated control system is used to reduce humidity; when the temperature and humidity adjustment equipment is a fan, this automated control system is used to adjust the air flow rate. This step ensures the real-time adjustment of the greenhouse environment, reduces manual intervention, and improves the efficiency of environmental management.

[0060] S7. Perform real-time feedback based on the environmental parameter acquisition results; In step S7, the environmental adjustment equipment in the greenhouse is remotely controlled through the Modbus protocol. However, control errors may occur in real-time control. In step S7, the sensor network established in step S1 is used to collect the environmental information in the greenhouse again in real-time, compare the gap between the expected result of equipment adjustment and the actual environmental information, and provide a reference basis for the precise adjustment of the equipment; In the process of environmental parameter adjustment and control algorithm correction, the following steps are included: Real-time feedback data acquisition: Continuously monitor the environmental parameters (such as temperature, humidity, air flow rate, etc.) in the greenhouse through the sensor network and compare them with the preset target parameters. When the system detects a deviation between the actual environmental parameters and the target value, the feedback mechanism is automatically triggered.

[0061] Control error analysis and adjustment: According to the feedback data, analyze the deviation between the actual output of the temperature and humidity control equipment and the expected effect. By comparing the real-time monitoring data with the preset target, calculate the error value and generate the control quantity that needs to be corrected. This error value will be fed back to the adjustment equipment to indicate it to make fine adjustments.

[0062] Dynamically adjust the operating parameters of the equipment: Based on the environmental change prediction results generated by the multi-layer perceptron (MLP) and long short-term memory network (LSTM) in steps S4 and S5, combined with the real-time feedback data, dynamically adjust the operating mode of the control equipment. For example, if the response speed of the temperature and humidity adjustment equipment is slow, the system can adjust the opening duration and frequency of the equipment, or change the operating mode of the heating / humidifying equipment to more precisely control the environmental parameters; Collect real-time feedback data through the sensor network to obtain the actual environmental parameters and the preset target value . The error is defined as the difference between the two: ; where, represents the error value at time , is the real-time monitored environmental parameter, which can be temperature , humidity , air flow velocity , light intensity and carbon dioxide concentration any one of them, is the target parameter preset by the system; To correct the error, the control system adjusts the working parameters of the device according to the calculated error. The control quantity and the error The relationship can be modeled by the following correction formula: ; where, is the feedback gain, is the weighting factor of the control quantity at the previous moment, is the control output at the previous moment. This model reflects that when the system adjusts the control quantity, it combines the current error and the historical control result to adjust the device more smoothly and effectively.

[0063] Based on the environmental change prediction results generated by the multi-layer perceptron (MLP) and long short-term memory network (LSTM) models in steps S4 and S5, predict the future temperature and humidity change trends. Assume that the predicted future environmental state obtained by LSTM is , then the corrected control output can be further adjusted according to the prediction result: ; where, is a non-linear function that fuses the error and the prediction result, which can dynamically adjust the operating parameters of the device according to the environmental changes to ensure the stability and accuracy of the system.

[0064] Model usage and application: Real-time error calculation: By calculating , the system can detect environmental changes in real time and generate feedback.

[0065] Control correction: According to the formula , adjust the working intensity of the control device (such as heater, humidity regulator, etc.) according to the feedback error.

[0066] Prediction and combination: Combine the environmental trends predicted by the LSTM model to dynamically correct the control strategy and avoid inaccurate control problems when the environment changes violently.

[0067] Embodiment 2: This embodiment also provides a computer device, which is applicable to a situation of a method for remote monitoring of the environment inside a greenhouse for greenhouse cultivation, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement a method for remote monitoring of the environment inside a greenhouse for greenhouse cultivation as proposed in the above embodiment.

[0068] This embodiment also provides a storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements a method for remote monitoring of the environment inside a greenhouse for greenhouse cultivation as proposed in the above embodiment.

[0069] The computer device can be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0070] If a function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of the present invention. And the foregoing storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs, etc., which can store program codes.

[0071] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a definable sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in combination with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device.

[0072] More specific examples (non-exhaustive list) of computer-readable media include the following: electrical connection parts with one or more wirings (electronic devices), portable computer disk cartridges (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), fiber optic devices, and portable compact disc read-only memory (CDROM). Additionally, a computer-readable medium can even be paper or other suitable media on which a program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or other suitable processing when necessary, and then stored in a computer memory.

[0073] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or combinations thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits having suitable combinational logic gate circuits, programmable gate arrays (PGA), field programmable gate arrays (FPGA), etc.

[0074] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A remote monitoring method for greenhouse environment used in greenhouse planting, characterized in that: The following steps are involved: S1, build a sensor network to collect environmental parameters; Establish a sensor network based on the Internet of Things to collect environmental parameters in the greenhouse and obtain environmental data; S2, preprocessing environmental data; Correct environmental data through data preprocessing technology to obtain clean data; S3, conduct multi-dimensional environmental analysis based on cleaning data; Use principal component analysis technology to identify the relationship between the environmental data in the greenhouse and obtain the analysis results; S4. Conduct environmental modeling based on the analysis results; Based on the multi-layer perceptron MLP network structure model, a prediction model between the greenhouse environment and crop growth was constructed to obtain the environmental model; S5. Predict temperature and humidity and make adjustment suggestions for the environmental model; Based on the long short-term memory network LSTM, the temperature and humidity of the environmental model are predicted and adjustment suggestions are generated to obtain adjustment suggestions; S6. Perform automatic control according to the adjustment suggestions; Use the automation control system based on the industrial automation communication protocol Modbus protocol to remotely control the greenhouse environment adjustment equipment and obtain the automation control results; S7. Collect environmental parameters in real time based on the automated control results, and provide real-time feedback on the collected results.

2. A remote monitoring method for greenhouse environment applied to greenhouse planting according to claim 1, characterized in that: In step S1, the environmental parameters include: Temperature and humidity, air flow rate, light intensity and carbon dioxide concentration.

3. The method for remote monitoring of greenhouse environment for greenhouse planting according to claim 1 is characterized in that: In step S2, the data preprocessing technology includes: Data interpolation: fill in missing values ​​with the mean, median or mode of environmental data; Outlier detection: Calculate the standard score of the environmental data point. If the difference between the environmental data point and the mean is greater than two or three times the standard deviation, the point is considered an outlier. Data smoothing: Reduce noise by calculating the average value within a window around the ambient data points.

4. The method for remote monitoring of greenhouse environment for greenhouse planting according to claim 1, characterized in that: In step S3, the principal component analysis technique includes: The main features are extracted by finding the direction with the largest variance in the cleaned data. The environmental parameters with the largest variance in the cleaned data have the greatest impact on crop growth.

5. The method for remote monitoring of greenhouse environment for greenhouse planting according to claim 1, characterized in that: In step S4, the multi-layer perceptron MLP network structure model includes: Establish an environmental model for the effects of greenhouse environmental parameters on crop growth, introduce a time series weighting mechanism, and predict and analyze temperature and humidity.

6. A remote monitoring method for greenhouse environment used in greenhouse planting according to claim 5, characterized in that: The environmental model of the greenhouse environmental parameters for crop growth is established as follows: ； in, For the moment Output of crop growth status; To input environmental parameters, including temperature ,humidity , air flow velocity , light intensity and carbon dioxide concentration At the moment The value of is a preset value, which represents the time-varying weight coefficient of each input parameter and the relative impact of each environmental factor on crop growth at different time points; is the bias term, representing the basic output value of the network, and its initial value is 0; is a nonlinear activation function, and its function expression is .

7. The method for remote monitoring of greenhouse environment for greenhouse planting according to claim 5, characterized in that: The introduction of the timing weighting mechanism includes: The time window parameter is introduced so that each input depends not only on the data at the current moment, but also on the environmental data in the historical time, which can be expressed by the following formula: ； in, is the length of the time window, indicating that the current crop growth status depends not only on the current environmental data, but also on the past Environmental data at all times; Indicates in the past Environmental parameters at the time The weighted sum of For the past Output of crop growth status within a certain time period.

8. The method for remote monitoring of greenhouse environment for greenhouse planting according to claim 5, characterized in that: The temperature and humidity prediction and analysis include: Data collection: obtained by cleaning data and analyzing results, and recorded in time series; Model training: Input the collected data and use the multi-layer perceptron MLP network structure model training to learn the nonlinear relationship between the input environmental factors and the changes in temperature and humidity; Temperature and humidity prediction: After training is completed, new real-time environmental data including current temperature and humidity, air flow speed, and light intensity are used to obtain temperature and humidity prediction results for the future. Environmental adjustment suggestions: According to the predicted temperature and humidity forecast results, adjust the temperature and humidity control equipment in the greenhouse to maintain a suitable crop growth environment.

9. The method for remote monitoring of greenhouse environment for greenhouse planting according to claim 1, characterized in that: In step S5, the long short-term memory network LSTM includes: The multi-layer perceptron MLP network structure model is used to obtain the greenhouse environment prediction model; Adopt multi-dimensional joint processing, jointly model all environmental factors and use environmental prediction models, by using activation functions for each environmental factor To achieve this, the activation function is the memory unit of the long short-term memory network LSTM; A time window weighting mechanism is adopted, that is, the historical data in each time window will be weighted according to the degree of influence of the environmental factor, and the impact of the environmental factor on the crop growth environment will be determined by calculating the variance. The larger the variance, the greater the impact.

10. The method for remote monitoring of greenhouse environment for greenhouse planting according to claim 1, characterized in that: In step S7, the environmental parameters in the greenhouse are collected in real time through the sensor network established in step S1, and the gap between the results of equipment adjustment after automatic control and the actual environmental parameters is compared to provide a reference for equipment adjustment.

Images