Intelligent tea garden planting management system and method

Abstract

The invention discloses a smart tea garden planting management system and method. The smart tea garden planting management system comprises a multi-source sensor network, an edge computing gateway, an AI decision unit and a control execution unit. The soil fertility and pest and disease damage are monitored in real time through a high-precision sensor, and precise recognition of disease spots and pests is achieved in combination with a YOLOv5 algorithm; the AI decision-making unit dynamically generates variable fertilization and graded prevention and control instructions according to the growth stages of the tea trees, and controls the water and fertilizer integrated device and the biological prevention and control equipment to execute operation. The system supports adaptation of multiple tea types, compatibility of new and old facilities and stable operation in an extreme environment, the fertilizer utilization rate is remarkably increased, the pesticide usage amount is reduced, and the system is suitable for intelligent and standardized management of large-scale tea gardens.

Description

Technical Field

[0001] The present invention relates to the field of agricultural Internet of Things and artificial intelligence technology, and in particular to a smart tea garden planting management system and method. Background Art

[0002] Smart tea garden planting management refers to the use of modern information technology and intelligent equipment to conduct all-round and refined management and monitoring of tea gardens to improve the efficiency, quality and sustainability of tea production.

[0003] At present, there are still some problems in tea garden planting. For example, traditional tea garden management relies on manual experience, the manual sampling and testing cycle is long, and key indicators such as nitrogen, phosphorus and potassium cannot be monitored in real time, resulting in inaccurate fertilization, low fertilizer utilization rate and delayed pest and disease control, reliance on visual inspection, high missed detection rate, and excessive pesticide residues caused by the abuse of chemical pesticides. Existing smart devices have single functions (such as only supporting irrigation or monitoring), lack multi-dimensional data collaborative analysis capabilities, and the hardware is difficult to adapt to the complex terrain of tea gardens. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the shortcomings of the existing technology, the present invention adopts a smart tea garden planting management system and method based on multi-source data fusion, which is used to achieve precise regulation of soil fertility in the tea garden, intelligent early warning of pests and diseases, and integrated water and fertilizer management, so as to improve tea quality and planting efficiency.

[0006] (II) Technical solution

[0007] To achieve the above object, the present invention provides the following technical solutions: a smart tea garden planting management system, including a multi-source sensor network, composed of soil nitrogen, phosphorus and potassium sensors, pH sensors, organic matter detectors, multi-spectral imagers, insect monitoring lights and weather stations, which are deployed in a tea garden in a 20m×20m grid;

[0008] The AI ​​decision-making unit includes a soil fertility analysis module, which generates a partitioned variable fertilization plan based on a three-dimensional heat map and a tea tree growth cycle model. The amount of fertilizer is dynamically adjusted according to the budding period, growth period, and picking period of the tea tree;

[0009] The pest and disease early warning module uses the YOLOv5 algorithm to detect targets in multispectral images, identify tea green leafhoppers, tea geometrids, and anthracnose and brown star disease lesions, and trigger graded prevention and control instructions;

[0010] Control execution unit, including intelligent irrigation machine, water and fertilizer integrated device, sound and light insect repellent equipment and predatory mite release device;

[0011] The water-fertilizer integrated device supports mixing nitrogen, phosphorus, potassium and trace elements in proportion, and the fertilization error rate is ≤1%;

[0012] The sound and light pest repelling device can emit sound waves with specific frequencies (20 - 50 kHz) and pulsed light to drive phototactic pests away.

[0013] As a preferred solution, the detection accuracy of the nitrogen, phosphorus and potassium sensor is ±0.5%, supporting dynamic monitoring of soil fertility at the level of 0.1 mg / kg;

[0014] The multispectral imager covers the visible light to near-infrared band, and is used to capture images of the tea tree canopy and identify leaf lesions;

[0015] The insect situation forecasting lamp integrates an infrared counting module to count the types and densities of pests in real time.

[0016] As a preferred solution, the multispectral imager is linked with the insect situation forecasting lamp. When the insect situation forecasting lamp detects that the pest density exceeds the standard, it automatically triggers the multispectral imager to take high-definition images of the surrounding area.

[0017] As a preferred solution, the soil fertility analysis module includes a tea tree variety database, supports the call of fertilizer requirement models for different tea tree varieties, and dynamically adjusts the ratio of organic fertilizer to inorganic fertilizer according to the soil pH value, controlling the soil acidification rate ≤ 0.1 pH / year.

[0018] As a preferred solution, the execution priority of the control and execution unit is: biological control > physical control > chemical control; when the pest and disease warning module determines that the insect population density ≥ 50 heads / ㎡, the predatory mite release device is preferentially activated. If the decline rate of the insect population density within 72 hours < 30%, the integrated water and fertilizer device is triggered to inject a low dose of biological pesticide.

[0019] A smart tea garden planting management method includes the following steps:

[0020] S1. Data collection and preprocessing, real-time collection of soil nitrogen, phosphorus and potassium content, pH value, organic matter content, canopy multispectral image and pest density data through a sensor network;

[0021] S2. AI dynamic decision-making, the soil fertility analysis module generates a variable fertilization plan for different zones according to the tea tree growth stages (germination stage, picking stage) and soil data, where the nitrogen fertilizer ratio is increased to 60% during the germination stage and the potassium fertilizer ratio is increased to 45% during the picking stage; the pest and disease warning module identifies the proportion of the diseased spot area through the multispectral image. If ≥ 5%, it is determined as severe infection and an emergency prevention and control instruction is triggered;

[0022] S3. Precise execution and feedback, the control and execution unit starts the integrated water and fertilizer device according to the instruction, applies fertilizers according to the set ratio, and at the same time the sound and light pest repelling device turns on the pest repelling mode at night;

[0023] S4. Multi-tea variety adaptation and expansion. For different tea tree varieties, call the preset models in the AI decision-making unit. Add soil water retention monitoring indicators in the Pu'er tea area and add bud and leaf growth rate tracking functions in the Longjing tea area.

[0024] As a preferred solution, in the step S2, the calculation formula for the variable fertilization plan is:

[0025] Q = α·S·(N target - N current ) + β·ln(P ratio ) + γ·T growth

[0026] where Q is the fertilization amount, S is the partition area, N target is the target nitrogen content, N current is the current detected value, T growth is the tea tree growth cycle coefficient, and α, β, and γ are weight parameters.

[0027] As a preferred solution, in the step S3, if the pest density exceeds the threshold for 3 consecutive days, the system automatically sends a warning notice to the administrator and generates a decision-making advice report including prevention and control plans, cost budgets, and expected effects.

[0028] (III) Beneficial effects

[0029] Compared with the prior art, the present invention provides a smart tea garden planting management system and method, which have the following beneficial effects:

[0030] First, the present invention realizes precise fertilization through the combination of a multi-source sensor network and an AI decision-making unit, further improves the fertilizer utilization rate, and effectively controls pests and diseases through the pest warning module;

[0031] Second, the present invention adopts a smart tea garden planting management system and method covering the entire link of "data collection - analysis - execution", and deeply integrates the modular sensor network with the AI algorithm to solve the problems of low tea garden management efficiency, resource waste, and environmental pollution. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a schematic diagram of the management system module of the present invention;

[0033] Figure 2 It is a schematic diagram of the management method flow of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0034] In order to better understand the purpose, structure, and function of the present invention, the present invention will be further described below in conjunction with the drawings and specific embodiments of a smart tea garden planting management system and method.

[0035] Embodiment 1

[0036] Reference Figure 1 , a smart tea garden planting management system of the present invention includes:

[0037] A multi-source sensor network, which consists of a soil nitrogen, phosphorus and potassium sensor, a pH sensor, an organic matter detector, a multispectral imager, an insect situation forecasting lamp and a weather station, and is deployed in the tea garden in a 20m×20m grid;

[0038] Among them, the detection accuracy of the nitrogen, phosphorus and potassium sensor is ±0.5%, and it supports dynamic monitoring of soil fertility at the 0.1mg / kg level;

[0039] The multispectral imager covers the visible light to near-infrared band, and is used to capture the tea tree canopy image and identify leaf surface disease spots;

[0040] The insect situation forecasting lamp integrates an infrared counting module to count the types and densities of pests in real time.

[0041] An AI decision-making unit, which includes a soil fertility analysis module. Based on a three-dimensional heat map and a tea tree growth cycle model, it generates a variable fertilization plan for different regions, and the fertilization amount is dynamically adjusted according to the budding period, growth period and picking period of the tea tree;

[0042] A pest and disease early warning module, which uses the YOLOv5 algorithm to perform target detection on multispectral images, identifies Empoasca vitis, Ectropis obliqua pests and anthracnose and red star disease spots, and triggers a hierarchical prevention and control instruction;

[0043] A control execution unit, which includes an intelligent irrigation machine, a water and fertilizer integration device, a sound and light insect repelling device and a predatory mite release device;

[0044] The present invention also includes a data middle platform, which integrates a MySQL database and an LSTM time series prediction model, is used to store the whole-link data of planting, optimizes the decision-making algorithm through historical operation logs, and supports remote OTA firmware upgrade;

[0045] An edge computing gateway, which is built with a LoRa communication module and a data encryption unit, is used to denoise, normalize and locally compress and transmit sensor data. The communication distance is ≥5km, and it supports operation in an environment of -20℃ to 60℃.

[0046] The grid deployment solution of the present invention adopts a 20m×20m grid layout to form a spatially continuous monitoring network. Each grid node integrates NPK sensors (detection accuracy of ±0.5%; monitoring resolution of 0.1mg / kg level) to implement the non-destructive capacitive coupling measurement principle; the pH sensor adopts potentiometric analysis technology and integrates a temperature compensation module; the organic matter detector is based on near-infrared absorption spectroscopy analysis, and the multispectral imager is equipped with a high-dynamic-range image sensor; the insect situation forecasting lamp is equipped with an ultraviolet insect-trapping light source (365nm band) and an infrared counting device. The sound and light insect repelling device can emit specific frequency sound waves (20 - 50kHz) and pulsed light to drive phototactic pests. The multispectral imager is linked with the insect situation forecasting lamp. When the insect situation forecasting lamp detects that the pest density exceeds the standard, it automatically triggers the multispectral imager to capture high-definition images of the surrounding area.

[0047] Its AI decision-making engine includes a soil optimization algorithm. The improvements to the original algorithm include: adding a lesion edge detection module, setting a grading threshold for insect body target detection: the detection rate of small targets above 10×10 pixels reaches 92%, and constructing a special dataset for tea tree pests and diseases.

[0048] Furthermore, the water and fertilizer integration device of the present invention supports the proportional mixing of nitrogen, phosphorus, potassium and trace elements, and the fertilization error rate ≤ 1%; the soil fertility analysis module includes a tea tree variety database, supports the call of the fertilizer requirement models of different tea tree varieties, and dynamically adjusts the ratio of organic fertilizer to inorganic fertilizer according to the soil pH value, controlling the soil acidification rate ≤ 0.1pH / year.

[0049] Even further, the execution priority of the control execution unit of the present invention is: biological control > physical control > chemical control; when the pest and disease warning module determines that the insect population density ≥ 50 heads / ㎡, the predatory mite release device is preferentially started. If the decline rate of the insect population density within 72 hours < 30%, the water and fertilizer integration device is triggered to inject a low-dose biological pesticide.

[0050] Its water and fertilizer integration device adopts a parallel peristaltic pump group (6-channel independent control), with a flow accuracy of 0.1mL / min, to achieve the precise proportioning of NPKMg compound fertilizers.

[0051] Example 2

[0052] A smart tea garden planting management method includes the following steps:

[0053] S1. Data collection and preprocessing, real-time collection of soil nitrogen, phosphorus, potassium content, pH value, organic matter content, canopy multispectral image and pest density data through the sensor network;

[0054] S2. AI Dynamic Decision Making. The soil fertility analysis module generates a variable fertilization plan for different zones based on the growth stages of tea plants (germination stage, picking stage) and soil data. Specifically, during the germination stage, the proportion of nitrogen fertilizer is increased to 60%, and during the picking stage, the proportion of potassium fertilizer is increased to 45%. The pest and disease early warning module identifies the proportion of diseased area through multi-spectral images. If it is ≥5%, it is determined as severe infection and an emergency prevention and control instruction is triggered.

[0055] S3. Precise Execution and Feedback. The control execution unit starts the integrated water and fertilizer device according to the instruction, applies fertilizers according to the set ratio. At the same time, the sound and light insect repellent device turns on the insect repellent mode at night.

[0056] S4. Adaptation and Expansion for Different Tea Varieties. For different tea tree varieties, the preset models in the AI decision-making unit are called. The soil water retention monitoring index is added in the Pu-erh tea area, and the bud and leaf growth rate tracking function is added in the Longjing tea area.

[0057] Specifically, in step S2 of the present invention, the calculation formula for the variable fertilization plan is:

[0058] Q = α·S·(N target -N current ) + β·ln(P ratio ) + γ·T growth

[0059] where Q is the fertilization amount, S is the area of the zone, N target is the target nitrogen content, N current is the current detected value, T growth is the tea tree growth cycle coefficient, and α, β, γ are weight parameters.

[0060] In step S3, if the insect population density exceeds the threshold for 3 consecutive days, the system automatically sends a warning notice to the administrator and generates a decision-making recommendation report including prevention and control plans, cost budgets, and expected effects.

[0061] Furthermore, the pest and disease prevention and control logic of the present invention includes a four-level response mechanism, specifically: Level I (lesion <2%): only use sound and light insect repellent; Level II (2 - 5% lesions): release 5000 predatory mites per mu; Level III (insect population density ≥30 heads per square meter): use biological pesticides (200 mL of BT preparation per mu); Level IV (insect population density ≥50 heads per square meter): start the composite prevention and control protocol.

[0062] The system and method of the present invention are composed of a multi-source sensor network, an edge computing gateway, an AI decision-making unit, and a control execution unit. The soil fertility and pests and diseases are monitored in real time through high-precision sensors, and the YOLOv5 algorithm is combined to achieve accurate identification of disease spots and pests. The AI decision-making unit dynamically generates variable fertilization and hierarchical prevention and control instructions according to the growth stage of tea trees, and controls the water and fertilizer integration device and the biological control equipment to execute operations. The system supports multi-tea type adaptation, compatibility of new and old facilities, and stable operation in extreme environments, significantly improving the fertilizer utilization rate and reducing the amount of pesticide used, and is applicable to the intelligent and standardized management of large-scale tea gardens.

[0063] It can be understood that the present invention is described by some embodiments. Those skilled in the art know that various changes or equivalent substitutions can be made to these features and embodiments without departing from the spirit and scope of the present invention. In addition, under the teaching of the present invention, these features and embodiments can be modified to adapt to specific situations and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application belong to the scope protected by the present invention.

Claims

1. An intelligent tea garden planting management system, characterized in that, Including: A multi-source sensor network, which consists of soil nitrogen, phosphorus, and potassium sensors, pH sensors, organic matter detectors, multispectral imagers, insect situation forecasting lamps, and weather stations, and is deployed in a 20m×20m grid pattern in the tea garden; An AI decision-making unit, which includes a soil fertility analysis module. Based on a three-dimensional heat map and a tea tree growth cycle model, it generates a variable fertilization plan for different zones, where the fertilization amount is dynamically adjusted according to the budding period, growth period, and picking period of the tea tree; A pest and disease early warning module, which uses the YOLOv5 algorithm to perform target detection on multispectral images, identify pests such as Empoasca vitis and Ectropis oblique, as well as anthracnose and red star disease spots, and trigger hierarchical prevention and control instructions; A control and execution unit, which includes an intelligent irrigation machine, a water and fertilizer integration device, a sound and light pest repelling device, and a predatory mite release device; Among them, the water and fertilizer integration device supports the proportional mixing of nitrogen, phosphorus, potassium, and trace elements, and the fertilization error rate ≤ 1%; The sound and light pest repelling device can emit specific frequency sound waves (20 - 50kHz) and pulsed light to drive phototactic pests away.

2. The intelligent tea garden planting management system according to claim 1, wherein, The detection accuracy of the nitrogen, phosphorus, and potassium sensors is ±0.5%, and it supports dynamic monitoring of soil fertility at the 0.1mg / kg level; The multispectral imager covers the visible light to near-infrared band, and is used to capture images of the tea tree canopy and identify leaf surface disease spots; The insect situation forecasting lamp is integrated with an infrared counting module to continuously count the types and densities of pests in real time.

3. The intelligent tea garden planting management system according to claim 2, wherein The multispectral imager is linked with the insect situation forecasting lamp. When the insect situation forecasting lamp detects that the pest density exceeds the standard, it automatically triggers the multispectral imager to take high-definition images of the surrounding area.

4. The intelligent tea garden planting management system according to claim 1, wherein, The soil fertility analysis module includes a tea tree variety database, supports the call of the fertilizer requirement models for different tea tree varieties, and dynamically adjusts the ratio of organic fertilizer to inorganic fertilizer according to the soil pH value, controlling the soil acidification rate ≤ 0.1pH / year.

5. A smart tea garden planting management system according to claim 1, characterized in that, The execution priority of the control and execution unit is: biological control > physical control > chemical control; when the pest and disease early warning module determines that the insect population density ≥ 50 heads / ㎡, the predatory mite release device is preferentially activated. If the decline rate of the insect population density within 72 hours < 30%, then the water and fertilizer integration device is triggered to inject a low dose of biological pesticide.

6. A method for planting and management in an intelligent tea garden, comprising the intelligent tea garden planting and management system according to any one of claims 1-5, characterized in that, Including the following steps: S1. Data collection and preprocessing, real-time collection of soil nitrogen, phosphorus, and potassium content, pH value, organic matter content, canopy multispectral image, and pest density data through the sensor network; S2. AI dynamic decision-making, the soil fertility analysis module generates a variable fertilization plan for different zones according to the tea tree growth stage (budding period, picking period) and soil data, where the proportion of nitrogen fertilizer in the budding period is increased to 60%, and the proportion of potassium fertilizer in the picking period is increased to 45%; the pest and disease early warning module identifies the proportion of the diseased area through multispectral images. If ≥ 5%, it is determined as severe infection and an emergency prevention and control instruction is triggered; S3. Precise execution and feedback, the control and execution unit starts the water and fertilizer integration device according to the instruction, discharges fertilizers according to the set ratio, and at the same time the sound and light pest repelling device turns on the pest repelling mode at night; S4. Adaptation and expansion for multiple tea types, for different tea tree varieties, call the preset models in the AI decision-making unit, increase the soil water retention monitoring index in the Pu'er tea area, and add the bud and leaf growth rate tracking function in the Longjing tea area.

7. The intelligent tea garden planting management method according to claim 6, characterized in that In the step S2, the calculation formula of the variable fertilization plan is as follows: Q = α·S·(N target - N current ) + β·ln(P ratio ) + γ·T growth Among them, Q is the fertilization amount, S is the area of the partition, N target is the target nitrogen content, N current is the current detected value, T growth is the tea tree growth cycle coefficient, and α, β, γ are weight parameters.

8. A method for planting and management of an intelligent tea garden according to claim 6, characterized in that, In the step S3, if the population density exceeds the threshold for three consecutive days, the system automatically sends a warning notice to the administrator and generates a decision-making recommendation report including the control plan, cost budget and expected effect.

Images