Environmental information collection device, water-fertilizer integrated control system and control method

Abstract

The present invention relates to the field of agricultural intelligent technology, and in particular to an environmental information collection device, an integrated water and fertilizer control system, and a control method. The environmental information collection device provided by the present invention uses an improved YOLOv5 algorithm model to extract features from specific areas of pest and disease images; the integrated water and fertilizer control system adjusts humidity, fertilizer application amount, light intensity, and carbon dioxide concentration based on environmental information. The integrated water and fertilizer control system provided by the present invention is combined with the environmental information collection device to achieve precise irrigation and fertilization of crops while accurately collecting growth information; by applying the attention mechanism to the YOLOv5 feature fusion module, an improved YOLOv5 algorithm model is obtained, and it is deployed on the environmental information collection device to achieve real-time detection of pests and diseases; by introducing a fuzzy PID algorithm in the control system, precise control of humidity is achieved.

Description

Technical Field

[0001] The present invention relates to the field of agricultural intelligent technology, and in particular to an environmental information collection device, a water-fertilizer integrated control system and a control method. Background Art

[0002] As an efficient irrigation technology that saves water and fertilizer, integrated water and fertilizer technology relies on a pressure control system to add soluble fertilizers to irrigation water in appropriate proportions. After the water and fertilizer are fully mixed, the water and fertilizer solution is accurately delivered to the roots of crops through drip irrigation or micro-sprinkler irrigation networks for crop absorption, achieving the effect of temporal synchronization and spatial coupling of water and nutrients, allowing crops to grow in the most suitable soil environment.

[0003] The collection of plant information (e.g., crop pests and diseases, ambient temperature and humidity, light intensity, and CO2 concentration) is one of the most widely used information technologies in agriculture. It plays a crucial role in understanding crop-related traits, providing a quantitative assessment of plant characteristics and their behavior under various environmental conditions. Currently, limited research exists on the acquisition of information within the context of integrated water and fertilizer systems, particularly localized information. This information is primarily collected manually, by drone, and from fixed platforms.

[0004] Manual information acquisition requires professional technicians who rely on experience and expertise, and the working environment is complex and harsh, which is both time-consuming and labor-intensive. Although drones have many advantages in collecting data and information, their accuracy is generally not high and professional operators need to be trained. In addition, drones are suitable for field environments and are easily affected by objective factors (light, fog, wind, rain, etc.). Fixed platforms do not collect comprehensive information and are expensive.

[0005] In the process of collecting agricultural information, the detection of pests and diseases mainly focuses on leaves containing diseased areas, and there is little research on specific diseased areas. The existing agricultural information collection methods make it difficult to monitor the specific areas where pests and diseases occur, and it is difficult to provide data reference for subsequent research and prevention.

[0006] Furthermore, in traditional agricultural technology, humidity control is mostly based on traditional PID control. However, agricultural working environments are harsh and complex, nonlinear systems, making traditional PID less adaptable to such systems and resulting in poor humidity control accuracy. Summary of the Invention

[0007] The purpose of the present invention is to overcome the defects of the existing technology and propose an environmental information collection device, a water-fertilizer integrated control system and a control method, which can achieve precise, accurate and intelligent irrigation and fertilization as well as accurate collection of crop information.

[0008] To achieve the above objectives, the present invention adopts the following specific technical solutions:

[0009] The environmental information collection device provided by the present invention includes:

[0010] Road condition information collection module, used to collect road condition images in real time;

[0011] Image information acquisition module, used to collect pest and disease images and assist in collecting road condition images;

[0012] Information collection integration module, used to collect environmental information;

[0013] The data processing module is used to process the collected pest and disease images and transmit environmental information, road condition images, and processed pest and disease images. The improved YOLOv5 algorithm model is used to extract features of specific pest and disease areas.

[0014] Data sending module, used to send road condition images, environmental information and processed pest and disease images for users to view;

[0015] The remote control module is used to receive remote control instructions and control the movement of the environmental information collection device.

[0016] Furthermore, the information collection integrated module includes a temperature and humidity sensor, a soil moisture sensor, a light intensity sensor, and a carbon dioxide concentration sensor.

[0017] Furthermore, the data processing module includes an upper computer and a lower computer. The upper computer extracts features of pest and disease images based on the improved YOLOv5 algorithm model and sends them to the lower computer. The lower computer is used to transmit road condition images, environmental information and processed pest and disease images to the data sending module.

[0018] Furthermore, the improved YOLOv5 algorithm model is obtained by applying the attention mechanism to the YOLOv5 algorithm model; wherein, the YOLOv5 algorithm model includes a feature extraction module, a feature fusion module and a feature detection module, and the attention mechanism is introduced after the SPPF layer in the feature fusion module.

[0019] Furthermore, the improved YOLOv5 algorithm model is used to transform the feature map of the pest and disease image and output the feature map W', as follows:

[0020] ;

[0021] ;

[0022] Among them, C, W, and H are the number of channels, width, and height of the feature map respectively. Indicates the GAP operation on the feature map on the kth channel, v k (i, j) represents the value of coordinate (i, j) in the feature map on the kth channel, Conv1 represents a one-dimensional convolution operation, Represents the multiplication of corresponding positions of matrix tensor elements, σ ​​represents the Sigmoid activation function; the specific calculation of k is as follows:

[0023] ;

[0024] Among them, C represents the number of channels, ||odd means that k is an odd number, γ and b are used to change the ratio between the number of channels C and the convolution kernel size k, γ is 2, and b is 1.

[0025] Furthermore, the environmental information collection device also includes an ultrasonic module and a display module. The ultrasonic module is used to assist the environmental information collection device in avoiding obstacles, and the display module is used to display the pest and disease images processed by the data processing module in real time.

[0026] The water-fertilizer integrated control system provided by the present invention includes the above-mentioned environmental information collection device, a sprinkler device, greenhouse equipment, a main control platform, a network transmission module, a server and a user terminal;

[0027] The main control platform is used to receive environmental information sent by the environmental information collection device, and control the sprinkler device to adjust the humidity and fertilizer amount, and control the greenhouse equipment to adjust the light intensity and carbon dioxide concentration; the network transmission module is used to upload the environmental information and adjustment parameters received by the main control platform to the server for users to remotely access and view at the user end.

[0028] The control method of the water-fertilizer integrated control system provided by the present invention comprises the following steps:

[0029] S1, the environmental collection device moves between the ridges to collect environmental information and sends it to the main control platform;

[0030] S2. The main control platform controls the sprinkler system and greenhouse equipment based on the received environmental information to adjust humidity, fertilizer application, light intensity, and carbon dioxide concentration;

[0031] S3. The main control platform uploads the environmental information and adjustment parameters to the server through the network transmission module for users to view remotely through the user terminal.

[0032] Furthermore, the main control platform controls the sprinkler device based on the fuzzy PID algorithm to adjust the humidity. The specific steps are as follows:

[0033] S1. Write the control rules for the input quantity and the three parameter correction values ​​of the PID controller Kp, Ki, and Kd, and store the control rules in the fuzzy rule base;

[0034] S2. The humidity set according to the crop growth conditions is used as the system set value, and the humidity measured by the environmental information acquisition device is used as the output response. The difference between the system set value and the output response is calculated to obtain the deviation e and the deviation change rate ec;

[0035] S3, the deviation e and the deviation change rate ec are quantized by the factor k e and k ec After adjustment, the fuzzy quantity is obtained by matching with the control rule;

[0036] The actual domains of the deviation e and the deviation change rate ec are [-e, e] and [-ec, ec], which are respectively called the basic domain of the deviation and the basic domain of the deviation change rate. The domain of the fuzzy set is [-N, N], and the quantization factor k is e =N / e,k ec =N / ec;

[0037] S4, defuzzifying the fuzzy quantity by the area centroid method to obtain kp, ki, kd and passing them to the PID controller for parameter tuning;

[0038] S5. Apply the signal output by the PID controller to the sprinkler device to adjust the humidity using the fuzzy PID algorithm.

[0039] The present invention can achieve the following technical effects:

[0040] The integrated water and fertilizer control system provided by the present invention is combined with an environmental information collection device to achieve precise irrigation and fertilization of crops while accurately collecting growth information. By introducing an attention mechanism, the target detection algorithm based on YOLOv5 is improved, and by being deployed on the environmental information collection device, real-time detection of pests and diseases is achieved. The fuzzy PID algorithm is introduced into the integrated water and fertilizer control system. By using fuzzy logic and fuzzy reasoning, the parameters can be adaptively adjusted, which can better adapt to these complex systems. Compared with traditional PID, it has better robustness and response speed, and can achieve precise control of humidity. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 2 is a schematic diagram of the structure of an environmental information collection device provided according to an embodiment of the present invention.

[0042] Figure 2 This is an information collection flow chart of an environmental information collection device according to an embodiment of the present invention.

[0043] Figure 3 This is a logic control diagram of an environmental information collection device provided according to an embodiment of the present invention.

[0044] Figure 4Schematic diagram of the feature map of a pest and disease image converted by the improved YOLOv5 algorithm model provided in an embodiment of the present invention.

[0045] Figure 5 4 is a calculation flow chart of the improved YOLOv5 algorithm model provided according to an embodiment of the present invention.

[0046] Figure 6 This is a control flow chart of a water-fertilizer integrated control system provided according to an embodiment of the present invention.

[0047] Figure 7 4 is a flow chart of a fuzzy PID algorithm provided according to an embodiment of the present invention.

[0048] Figure 8 It is a schematic diagram of the overall operation of the environmental information collection device and the water-fertilizer integrated control system provided according to an embodiment of the present invention.

[0049] Reference numerals include:

[0050] Road condition information acquisition module 1, image information acquisition module 2, data sending module 3, information acquisition integration module 4, remote control module 5, host computer 6, slave computer 7, ultrasonic module 8, display module 9, environmental information acquisition device 10, field ridge 11, water pipe 12, field ridge 13, sprinkler 14. DETAILED DESCRIPTION

[0051] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the following description, identical modules are denoted by identical reference numerals. In the case of identical reference numerals, their names and functions are also identical. Therefore, their detailed description will not be repeated.

[0052] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and do not constitute a limitation of the present invention.

[0053] Figure 1 The structure of the environmental information collection device provided by an embodiment of the present invention is shown.

[0054] like Figure 1 As shown, the environmental information collection device provided by the embodiment of the present invention includes a road condition information collection module 1, an image information collection module 2, a data sending module 3, an information collection integration module 4, a remote control module 5, a host computer 6, and a slave computer 7.

[0055] The road condition information acquisition module 1 collects road condition information in real time and uploads it to the data sending module 3; the image information acquisition module 2 collects pest and disease images and uploads them to the host computer 6 in the data processing module, and at the same time performs auxiliary collection of road condition information and uploads it to the data sending module 3; the information acquisition integration module 4 includes a temperature and humidity sensor, a soil moisture sensor, a light intensity sensor and a CO2 concentration sensor, which are used to collect corresponding environmental information and upload it to the data sending module 3.

[0056] Figure 2 The information collection process of the environmental information collection device provided by the embodiment of the present invention is shown.

[0057] like Figure 2 As shown, the data processing module includes a host computer 6 and a slave computer 7. The host computer 6 processes the pest and disease images uploaded by the image information acquisition module 2 and transmits them to the slave computer 7 via the serial port. The slave computer 7 processes the environmental information collected by the information acquisition and integration module 4 and uploads the processed environmental information and pest and disease images to the data transmission module 3.

[0058] In some specific embodiments, the host computer 6 is a Raspberry Pi, and the slave computer 7 is an STM3 2F4.

[0059] The data transmission module 3 transmits the road condition information uploaded by the road condition information collection module 1 and the image information collection module 2, as well as the pest and disease images and environmental information processed by the data processing module to the server and the Internet of Things cloud platform for users to view;

[0060] In some specific embodiments, the data sending module 3 sends relevant data through a 4G module of the MQTT protocol.

[0061] Figure 3 The control flow of the environmental information collection device provided by an embodiment of the present invention is shown.

[0062] like Figure 3 As shown, the remote control module 5 includes an enhanced WIFI module, which transmits control information to the lower computer 7 through the serial port. The lower computer 7 controls the motor of the environmental information acquisition device according to the control instructions to realize the switching between the four motion states of the environmental information acquisition device: forward, backward, turning, and parking.

[0063] In some preferred embodiments, the environmental information collection device also includes an ultrasonic module 8 and a display module 9. The ultrasonic module 8 uploads the ultrasonic information to the lower computer 7 to control the servo and motor of the environmental information collection device, thereby assisting the environmental information collection device to avoid obstacles; the display module 9 displays the pest and disease images processed by the data processing module in real time.

[0064] In the data processing module, the upper computer 6 performs real-time analysis and processing of pest and disease images based on the improved YOLOv5 algorithm model. The improved YOLOv5 algorithm model introduces the attention mechanism (AM) into the YOLOv5 algorithm model. AM selects the most significant features of the target object from a large number of extracted global features and filters out useless redundant features, so that the improved YOLOv5 algorithm model can advance the features of the pest and disease images, which is conducive to providing data reference for subsequent research and prevention and control.

[0065] Figure 4 The figure shows the feature map of the pest and disease image transformed by the improved YOLOv5 algorithm model provided by the embodiment of the present invention. Figure 5 The figure shows the calculation process of the improved YOLOv5 algorithm model provided by the embodiment of the present invention.

[0066] like Figure 4-5 As shown, the improved YOLOv5 algorithm model in the embodiment of the present invention is obtained by applying the attention mechanism to the YOLOv5 algorithm model; wherein the YOLOv5 algorithm model includes a feature extraction module, a feature fusion module and a feature detection module, and the attention mechanism is introduced after the SPPF layer in the feature fusion module.

[0067] The improved YOLOv5 algorithm model enables the network to obtain more information about the target area, as follows:

[0068] Assume that the feature map of the pest and disease image is X, and its size is C×W×H, where C, W, and H are the number of channels, width, and height of the feature map respectively. The feature map output is W´. The processing process is as follows:

[0069] ;

[0070] ;

[0071] in, Indicates the k Perform GAP operation on the feature map of the channels, v k (i, j) represents the value of coordinate (i, j) in the feature map on the kth channel, Conv1 represents a one-dimensional convolution operation, Represents the multiplication of corresponding positions of matrix tensor elements, σ ​​represents the Sigmoid activation function; the specific calculation of k is as follows:

[0072] ;

[0073] Among them, C represents the number of channels, ||odd means that k can only be an odd number, γ and b are used to change the ratio between the number of channels C and the convolution kernel size k, γ takes a value of 2, and b takes a value of 1.

[0074] The improved YOLOv5 algorithm model introduces an attention mechanism after the SPPF layer on the basic structure of the YOLOv5 algorithm, making the model training more inclined to the area of ​​interest.

[0075] Figure 6 The control flow of the water-fertilizer integrated control system provided by an embodiment of the present invention is shown.

[0076] like Figure 6 As shown, the water-fertilizer integrated control system provided by the embodiment of the present invention includes the above-mentioned environmental information collection device, sprinkler irrigation device, greenhouse equipment, main control platform, network transmission module, server and user terminal.

[0077] The main control platform is used to receive environmental information sent by the environmental information collection device, and control the sprinkler device to adjust the humidity and fertilizer amount, and control the greenhouse equipment to adjust the light intensity and carbon dioxide concentration; the network transmission module is used to upload the environmental information and adjustment parameters received by the main control platform to the server for users to remotely access and view at the user end.

[0078] The control method of the water-fertilizer integrated control system includes the following steps:

[0079] S1, the environmental collection device moves between the ridges to collect environmental information and sends it to the main control platform;

[0080] S2. The main control platform controls the sprinkler system and greenhouse equipment based on the received environmental information to adjust humidity, fertilizer application, light intensity, and carbon dioxide concentration;

[0081] S3. The main control platform uploads the environmental information and adjustment parameters to the server through the network transmission module for users to view remotely through the user terminal.

[0082] Figure 7 The flowchart of the fuzzy PID algorithm provided by the embodiment of the present invention is shown.

[0083] like Figure 7 As shown, in the water-fertilizer integrated control system provided by the embodiment of the present invention, humidity is controlled using a fuzzy PID algorithm, and the specific steps are as follows:

[0084] S1. Write the control rules for the input quantity and the three parameter correction values ​​of the PID controller Kp, Ki, and Kd, and store the control rules in the fuzzy rule library.

[0085] Based on the knowledge and experience accumulated over the years by relevant experts or operators, the control rules of the input quantity and the three parameter correction values ​​of the PID controller are summarized and the corresponding fuzzy control rules are compiled into the fuzzy rule library.

[0086] S2. The humidity set according to the crop growth conditions is used as the system set value, and the humidity measured by the environmental information acquisition device is used as the output response. The difference between the system set value and the output response is used to obtain the deviation e and the deviation change rate ec.

[0087] When the controller is running, the system set value (to achieve the set humidity according to the crop growth conditions) and the output response (humidity measured by the sensor) are subtracted to obtain the deviation e and the deviation change rate ec.

[0088] S3, the deviation e and the deviation change rate ec are quantized by the factor k e and k ec After adjustment, the fuzzy quantity is obtained by matching it with the control rule.

[0089] The actual domain of the deviation e and the rate of change of the deviation ec is [-e,e] and [-ec,ec], which are called the basic domain of the deviation and the rate of change of the deviation. The domain of the fuzzy set is [-N,N]. Therefore, the quantization factor k e =N / e,k ec =N / ec.

[0090] S4. Defuzzify the fuzzy quantity using the area centroid method to obtain kp, ki, and kd and pass them to the PID controller for parameter tuning.

[0091] S5. Apply the signal output by the PID controller to the sprinkler device to adjust the humidity using the fuzzy PID algorithm.

[0092] Figure 8 The overall operation of the environmental information collection device and the water-fertilizer integrated control system provided by the embodiment of the present invention is shown.

[0093] like Figure 8 As shown, the integrated water and fertilizer control system is arranged in the farmland, wherein the water pipe 12 and the sprinkler 14 are arranged on the ridge 13 , and the environmental information collection device 10 moves in the ridge 11 and collects information about the crops on the ridge 13 .

[0094] In the description of this specification, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0095] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

[0096] The above specific embodiments of the present invention do not constitute a limitation on the scope of protection of the present invention. Any other corresponding changes and modifications made based on the technical concept of the present invention should be included in the scope of protection of the claims of the present invention.

Claims

1. An environmental information collection device, characterized in that: include: Road condition information collection module, used to collect road condition images in real time; Image information acquisition module, used to collect pest and disease images and assist in collecting road condition images; Information collection integration module, used to collect environmental information; The data processing module is used to process the collected pest and disease images and transmit environmental information, road condition images, and processed pest and disease images. The improved YOLOv5 algorithm model is used to extract features of specific pest and disease areas. Data sending module, used to send road condition images, environmental information and processed pest and disease images for users to view; A remote control module, used to receive remote control instructions and control the movement of the environmental information collection device; It also includes an ultrasonic module and a display module, wherein the ultrasonic module is used to assist the environmental information collection device in avoiding obstacles, and the display module is used to display the pest and disease images processed by the data processing module in real time; The improved YOLOv5 algorithm model is obtained by applying the attention mechanism to the YOLOv5 algorithm model; wherein the YOLOv5 algorithm model includes a feature extraction module, a feature fusion module and a feature detection module, and the attention mechanism is introduced after the SPPF layer in the feature fusion module; The improved YOLOv5 algorithm model is used to transform the feature map of the pest and disease image and output the feature map W', as follows: ； ； Among them, C, W, and H are the number of channels, width, and height of the feature map respectively. Indicates the GAP operation on the feature map on the kth channel, v k (i, j) represents the value of coordinate (i, j) in the feature map on the kth channel, Conv1 represents a one-dimensional convolution operation, Represents the multiplication of corresponding positions of matrix tensor elements, σ ​​represents the Sigmoid activation function; the specific calculation of k is as follows: ； Where C represents the number of channels, ||odd indicates that k is an odd number, γ and b are used to change the ratio between the number of channels C and the convolution kernel size k, γ is 2, and b is 1; The data processing module includes an upper computer and a lower computer. The upper computer extracts features of pest and disease images based on the improved YOLOv5 algorithm model and sends them to the lower computer. The lower computer is used to transmit road condition images, environmental information and processed pest and disease images to the data sending module.

2. The environmental information collection device according to claim 1, characterized in that: The information acquisition integrated module includes a temperature and humidity sensor, a soil moisture sensor, a light intensity sensor, and a carbon dioxide concentration sensor.

3. A water-fertilizer integrated control system, characterized in that: The environmental information collection device according to any one of claims 1 to 2 further comprises a sprinkler device, greenhouse equipment, a main control platform, a network transmission module, a server and a user terminal; The main control platform is used to receive the environmental information sent by the environmental information collection device, and control the sprinkler device to adjust the humidity and fertilizer amount, and control the greenhouse equipment to adjust the light intensity and carbon dioxide concentration; the network transmission module is used to upload the environmental information and adjustment parameters received by the main control platform to the server for users to remotely access and view at the user end.

4. The control method of the water-fertilizer integrated control system according to claim 3, characterized in that: The steps include: S1, the environmental collection device moves between the ridges to collect environmental information and sends it to the main control platform; S2. The main control platform controls the sprinkler system and greenhouse equipment based on the received environmental information to adjust humidity, fertilizer application, light intensity, and carbon dioxide concentration; S3. The main control platform uploads the environmental information and adjustment parameters to the server through the network transmission module for users to view remotely through the user terminal.

5. The control method according to claim 4, characterized in that: The main control platform controls the sprinkler device based on the fuzzy PID algorithm to adjust the humidity. The specific steps are as follows: S1. Write the control rules for the input quantity and the three parameter correction values ​​of the PID controller Kp, Ki, and Kd, and store the control rules in the fuzzy rule base; S2. The humidity set according to the crop growth conditions is used as the system set value, and the humidity measured by the environmental information acquisition device is used as the output response. The difference between the system set value and the output response is calculated to obtain the deviation e and the deviation change rate ec; S3, the deviation e and the deviation change rate ec are quantized by the factor k e and k ec After adjustment, the fuzzy quantity is obtained by matching with the control rule; The actual domains of the deviation e and the deviation change rate ec are [-e, e] and [-ec, ec], which are respectively called the basic domain of the deviation and the basic domain of the deviation change rate. The domain of the fuzzy set is [-N, N], and the quantization factor k is e =N / e,k ec =N / ec; S4, defuzzifying the fuzzy quantity by the area centroid method to obtain kp, ki, kd and passing them to the PID controller for parameter tuning; S5. Apply the signal output by the PID controller to the sprinkler device to adjust the humidity using the fuzzy PID algorithm.

Images