Fruit tree illumination regulation and control cultivation method capable of improving staining degree and monitoring system

Abstract

The invention provides a fruit tree illumination regulation and control cultivation method and monitoring system capable of improving the coloring degree, and belongs to the technical field of fruit tree planting, the fruit tree illumination regulation and control cultivation method comprises an illumination monitoring unit used for collecting fruit tree illumination parameters and fruit growth state data; the data transmission unit is connected with the illumination monitoring unit and is used for transmitting the illumination parameters and the fruit growth state data to the intelligent control unit; the intelligent control unit is used for analyzing and processing the illumination parameters and the fruit growth state data, constructing an illumination-staining degree correlation model and generating an illumination regulation and control instruction based on the illumination-staining degree correlation model; and the illumination regulation and control execution unit is connected with the intelligent control unit and is used for receiving the illumination regulation and control instruction and executing corresponding illumination regulation and control operation. The problems that fruit coloring is uneven due to the fact that current fruit tree cultivation depends on natural illumination or traditional manual intervention, existing illumination regulation and control precision is low, a real-time monitoring and intelligent association model is lacked, and a regulation and control execution unit is single in function are solved, and the fruit coloring degree and quality are effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fruit tree cultivation, in particular to a fruit tree light regulation cultivation method for improving color degree and a monitoring system. BACKGROUND

[0002] In the field of fruit tree cultivation, fruit color degree is one of the main indicators affecting the commodity value of agricultural products, directly related to consumer willingness to purchase and market price. The uniform color of high-quality fruit depends on stable and suitable light conditions during the growth cycle. Light intensity, spectral distribution, and light duration not only affect the synthesis and metabolism of chlorophyll, carotenoids, and anthocyanins in fruit, but also indirectly regulate sugar accumulation and flavor substance formation in fruit. Currently, fruit tree cultivation mostly adopts open cultivation or simple facility cultivation mode, and light conditions mainly depend on the natural environment, which is easily affected by seasonal changes, weather fluctuations (such as rain, fog, and haze), and fruit tree planting density and tree structure, leading to uneven fruit color and insufficient color degree.

[0003] To meet the demand for fruit tree light regulation, existing technologies mainly use artificial intervention or simple equipment assistance, such as artificial pruning of branches to improve ventilation and light transmission, laying of reflective film to enhance local light, and using ordinary supplemental light lamps to supplement light. However, these methods have obvious limitations: first, the regulation precision is low, and light parameters cannot be dynamically adjusted according to the needs of different growth stages of fruit. For example, during the key period of fruit coloration, traditional supplemental light lamps mostly use fixed spectrum and intensity, which cannot match the specific spectrum (such as 600-700 nm red light) required for anthocyanin synthesis, resulting in poor coloration effect. Second, there is a lack of real-time monitoring and feedback mechanism. Artificial observation of fruit growth status has a lag, and cannot timely discover problems such as insufficient light and unbalanced spectrum, and cannot quantify the relationship between light parameters and color degree. Third, the degree of intelligence is low. Existing technologies mostly rely on artificial experience operation, and cannot realize automation and precision of light regulation. Especially in large-scale fruit trees, artificial regulation has high cost and low efficiency, which cannot meet the needs of large-scale cultivation.

[0004] With the development of agricultural intelligent technology, some researches attempt to apply sensors and control devices to fruit tree light management, but there are still technical bottlenecks. On the one hand, existing monitoring systems mostly only collect single parameter of light intensity, ignoring spectrum distribution, light duration, and other indicators that are also crucial for fruit coloration, and lack real-time recognition ability of fruit growth status (such as maturity, color change, and fruit surface defects), resulting in incomplete data collection and inability to provide sufficient basis for accurate regulation. On the other hand, existing systems do not build an effective light-color degree correlation model, and cannot dynamically output suitable light parameters according to fruit tree varieties and growth stages. Regulation instructions are mostly generated based on fixed thresholds, with poor flexibility and adaptability, and are difficult to meet complex light requirements in different fruit tree environments.

[0005] Therefore, there is an urgent need for a fruit tree light regulation cultivation method and monitoring system that integrates multi-parameter monitoring, intelligent model analysis, and accurate execution of regulation and control, which is the inevitable direction to break through the current technical bottleneck and solve the problem of fruit coloring. SUMMARY

[0006] In order to overcome the shortcomings of the prior art, the purpose of the present application is to provide a fruit tree light regulation cultivation method and monitoring system for improving coloration, which solves the problems of uneven fruit coloration caused by relying on natural light or traditional manual intervention, low accuracy of existing light regulation, lack of real-time monitoring and intelligent correlation model, and single function of regulation and control execution unit in current fruit tree cultivation by multi-parameter light and fruit growth state monitoring, light-coloration correlation model construction, and multi-module light regulation.

[0007] To achieve the above purpose, the present application provides the following scheme: A fruit tree light regulation cultivation monitoring system for improving coloration, comprising a light monitoring unit, a data transmission unit, an intelligent control unit, and a light regulation execution unit. The light monitoring unit is used to collect fruit tree light parameters and fruit growth state data. The data transmission unit is connected to the light monitoring unit and is used to transmit the light parameters and fruit growth state data to the intelligent control unit. The intelligent control unit is used to analyze and process the light parameters and fruit growth state data, construct a light-coloration correlation model, and generate light regulation instructions based on the light-coloration correlation model. The light regulation execution unit is connected to the intelligent control unit and is used to receive the light regulation instructions and perform corresponding light regulation operations.

[0008] Preferably, the light monitoring unit collects fruit tree light parameters including one or more of light intensity, spectral distribution, and light duration. The light monitoring unit includes a light intensity sensor for collecting light intensity, a spectral sensor for collecting spectral distribution, a light duration recorder for recording light duration, and an AI vision monitoring module for collecting fruit images and identifying fruit growth state, which is used to identify fruit maturity, color change, and fruit surface defects.

[0009] Preferably, the data transmission unit includes a wireless transmission module and an edge computing node. The wireless transmission module is used for transmitting data at a preset frequency, and triggering instant transmission when the illumination parameter is abnormal, the spectral distribution is abnormal, or the fruit growth state data changes significantly; The edge computing node is used for preliminary screening and abnormality judgment of the illumination parameter and fruit growth state data, and generating preliminary warning information.

[0010] Preferably, the intelligent control unit comprises a cloud server and a terminal module for user interaction; The cloud server is used for storing historical data, and constructing the illumination-chroma correlation model through a random forest algorithm, which can dynamically output suitable illumination parameters according to fruit tree varieties and growth stages; The terminal module supports illumination parameter display, abnormality alarm receiving, remote control of the illumination regulation and control execution unit, and entry of fruit tree farming operation records; The abnormality alarm comprises one or more of insufficient illumination alarm, spectral imbalance alarm, and fruit growth abnormality alarm.

[0011] Preferably, the illumination regulation and control execution unit comprises a light supplement module for supplementing illumination, a sunshade module for adjusting strong light, a light reflection module for enhancing local illumination, and a bionic illumination regulation module for adaptively adjusting illumination; The light supplement module is used for adjusting illumination intensity and spectral proportion; The sunshade module is used for adjusting the sunshade degree according to the illumination intensity; The light reflection module is used for adjusting the laying position and angle to change the light reflection direction, and focusing on the fruit tree canopy and fruit area; The bionic illumination regulation module is used for automatically adjusting the light transmittance or reflectivity according to the illumination condition, to adapt to the illumination requirements of different growth stages of the fruit tree.

[0012] The application also provides a fruit tree illumination regulation and control cultivation method for improving chroma, which is applied to the fruit tree illumination regulation and control cultivation monitoring system for improving chroma, and comprises the following steps: S1, completing deployment and debugging of the illumination monitoring unit, data transmission unit, intelligent control unit and illumination regulation and control execution unit, and entering fruit tree basic information; S2, collecting fruit tree illumination parameters and fruit growth state data through the illumination monitoring unit, and transmitting the data to the intelligent control unit through the data transmission unit; S3, the intelligent control unit analyzes the collected data, and constructs an illumination-chroma correlation model in combination with the fruit tree basic information; S4, based on the light-chroma correlation model, dynamically monitoring and judging the light parameters of the fruit trees, generating light regulation instructions, and executing corresponding light regulation operations by the light regulation execution unit; S5, during the key period of fruit coloring, optimizing the light regulation parameters based on the light-chroma correlation model, and strengthening the control of the light conditions of the fruit trees; S6, after the fruit is picked, collecting fruit-related data to evaluate the light regulation effect, and optimizing the light-chroma correlation model based on the evaluation results.

[0013] Preferably, in S3, the process of constructing the light-chroma correlation model includes: Preprocessing the collected fruit tree light parameters and fruit growth state data; Extracting feature parameters that affect fruit coloring as model inputs; Taking fruit chroma as the output target, training the light-chroma correlation model through a random forest algorithm; Verify the accuracy of the light-chroma correlation model. If the accuracy does not meet the preset requirements, adjust the model parameters and retrain.

[0014] Preferably, in S4, based on the light-chroma correlation model, the light parameters of the fruit trees are dynamically monitored and judged, light regulation instructions are generated, and corresponding light regulation operations are executed by the light regulation execution unit. Specifically, it includes: Continuously collecting fruit tree light parameters and fruit growth state data and transmitting them to the intelligent control unit; Based on the light-chroma correlation model, combining the appropriate light parameter range corresponding to the fruit tree variety and growth stage, judge whether the real-time light parameter meets the preset range; According to the judgment result, the corresponding light regulation instruction is generated, which includes one or more of the light supplement instruction, the sunshade instruction, and the spectrum adjustment instruction; The light regulation execution unit receives and executes the light regulation instruction.

[0015] Preferably, in S5, during the key period of fruit coloring, the light regulation parameters are optimized based on the light-chroma correlation model, and the control of the light conditions of the fruit trees is strengthened. Specifically, it includes: Based on the light-chroma correlation model, adjust the appropriate light parameter range of the fruit tree coloring key period; Control the reflection module to enhance the light intensity of the local area of the fruit trees; Continuously collect fruit coloring data and dynamically adjust the light regulation operation according to the coloring data.

[0016] Preferably, in S6, after the fruits are picked, fruit-related data is collected to evaluate the light regulation effect, and the light-chroma correlation model is optimized based on the evaluation results, specifically including: Collecting fruit coloring-related indicators and quality indicators as effect evaluation data; Comparing the effect evaluation data with preset standards or single-fruit-tree indicators under traditional management methods to determine the light regulation effect; According to the effect determination result, the light-chroma correlation model is adjusted in parameters or retrained to realize model iteration optimization.

[0017] According to the specific embodiments provided by the present application, the following technical effects are disclosed: (1) The present application integrates light intensity sensors, spectrum sensors, light duration recorders, and AI visual monitoring modules through the light monitoring unit to realize comprehensive collection of fruit tree light parameters (intensity, spectrum, duration) and fruit growth state (maturity, color change, fruit surface defects). The preset frequency transmission and instant trigger transmission functions of the data transmission unit are matched to solve the problem of traditional monitoring focusing on only a single light parameter and the lag of fruit state observation, providing comprehensive, real-time, and accurate data support for subsequent light regulation, effectively avoiding regulation deviation caused by data loss.

[0018] (2) The present application uses a random forest algorithm to construct a light-chroma correlation model through the intelligent control unit, which can dynamically output suitable light parameters according to fruit tree varieties and growth stages. Through the collaborative work of the light regulation execution unit's light supplement, shading, light reflection, and bionic light adjustment modules, targeted light supplement, shading, and spectrum adjustment operations can be performed to solve the defects of low regulation precision, reliance on human experience, and single-function execution unit in the prior art, realizing dynamic and precise regulation of fruit tree light, especially strengthening light control during the critical period of fruit coloring, and significantly improving fruit coloring uniformity and coloring degree.

[0019] (3) The present application evaluates the regulation effect by collecting coloring and quality indicators after fruit picking and optimizes the light-chroma correlation model based on the evaluation results to form an integrated closed-loop management system of monitoring, analysis, regulation, evaluation, and optimization. Not only can it continuously improve the adaptability and effectiveness of subsequent fruit tree light regulation, but also reduces the labor management cost of large-scale fruit trees, solves the problem of low efficiency of large-scale application due to the lack of iteration optimization capability of existing systems, provides a long-term reusable technical solution for fine and intelligent management of fruit trees, and helps fruit farmers reduce economic losses caused by light problems. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0021] Figure 1 A whole modular schematic diagram of a fruit tree light regulation cultivation monitoring system for improving coloring degree according to the present application; Figure 2 A light monitoring unit modular schematic diagram provided for Embodiment 1 of the present application; Figure 3 A data transmission unit modular schematic diagram provided for Embodiment 1 of the present application; Figure 4 An intelligent control unit modular schematic diagram provided for Embodiment 1 of the present application; Figure 5 A light regulation execution unit modular schematic diagram provided for Embodiment 1 of the present application; Figure 6 A flow chart of a fruit tree light regulation cultivation method for improving coloring degree according to the present application. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without any creative effort fall within the scope of the present application.

[0023] In order to make the above-mentioned purposes, features and advantages of the present application more apparent and easy to understand, the present application will be further described in detail with reference to the drawings and specific embodiments.

[0024] Embodiment 1 As Figure 1As shown in the figure, the fruit tree light regulation cultivation and monitoring system for improving color intensity provided in this embodiment mainly consists of a light monitoring unit, a data transmission unit, an intelligent control unit, and a light regulation execution unit. The light monitoring unit is used to collect fruit tree light parameters and fruit growth status data. The data transmission unit is connected to the light monitoring unit and is used to transmit the light parameters and fruit growth status data to the intelligent control unit. The intelligent control unit is used to analyze and process the light parameters and fruit growth status data, construct a light-color intensity correlation model, and generate light regulation commands based on the light-color intensity correlation model. The light regulation execution unit is connected to the intelligent control unit and is used to receive the light regulation commands and execute the corresponding light regulation operations.

[0025] Specifically, each unit establishes a communication connection via wired or wireless means. During deployment, the location of each component needs to be determined based on the fruit tree area, fruit tree variety (such as apples and grapes), and planting row spacing: the light monitoring unit is deployed in multiple locations on a single fruit tree (one set each for the top of the canopy, the central sunny side, the central shady side, and the bottom), with each monitoring unit installed at the same height as the corresponding fruit cluster (approximately 1.2-1.8m); in the light control execution unit, 2-3 supplementary lighting modules are installed for the weak light areas of the fruit tree canopy (such as the shady side), reflective modules are laid on the ground on both sides of the fruit tree roots, shading modules cover the top of the fruit tree canopy, and bionic light adjustment modules are installed on the sides of the fruit tree canopy; the edge computing nodes of the data transmission unit are set within a 10m radius around the fruit tree, and the cloud server of the intelligent control unit communicates with the edge computing nodes via the public network, while the terminal modules (such as tablets and PCs) interact with the cloud server through account login.

[0026] Among them, such as Figure 2 As shown, the light monitoring unit includes a light intensity sensor for collecting light intensity, a spectral sensor for collecting spectral distribution, a light duration recorder for recording light duration, and an AI visual monitoring module for collecting fruit images and identifying fruit growth status. The AI ​​visual monitoring module is used to identify fruit maturity, color changes, and fruit surface defects.

[0027] Specifically, the light monitoring unit integrates a light intensity sensor, a spectrum sensor, a light duration recorder, and an AI visual monitoring module within a waterproof and dustproof housing. Each component is connected to the core controller (model STM32F407) via an onboard bus. Specific parameters and functions are as follows: Light intensity sensor: High-precision silicon-based photoelectric sensor (model S1226-88CQ) is selected, with a measurement range of 0-200000 lux, a measurement accuracy of ±2% FS, and a sampling frequency of 1 time / minute. By converting the light signal into a voltage signal (0-5V), the real-time light intensity value is obtained after AD conversion by the core controller, which can identify three states: "insufficient light (<30000 lux)", "suitable light (30000-80000 lux)", and "excessive light (>80000 lux)".

[0028] Spectral sensor: A miniature fiber optic spectrometer (model USB2000+) is used, with a spectral detection range of 350-1050 nm, a wavelength accuracy of ±0.5 nm, and a sampling interval of 1 nm. The real-time spectral power distribution of red light (600-700 nm), blue light (400-500 nm), and green light (500-600 nm) can be collected. The core controller calculates the proportion of each waveband (such as red light proportion >30% for anthocyanin synthesis) to determine whether the spectrum is unbalanced.

[0029] Light duration recorder: The core controller has a built-in clock module (accuracy ±1 s / day) to record the effective light duration. When the light intensity is >10000 lux, the timer starts, and when it is <5000 lux, the timer stops. The effective light duration of each day can be automatically accumulated, and the threshold value of different growth stages can be set (such as ≥10 h / day during the key coloration period).

[0030] AI vision monitoring module: It includes a 200 million pixel industrial camera (model MV-CA020-10GM, supporting 1080P@30fps), a fill light (wavelength 650nm red light), and an edge AI processor (model RK3588). The camera is vertically downward to the fruit cluster (shooting distance 0.5-0.8m), and takes a fruit image every 15 minutes. The AI processor has a fruit recognition model based on YOLOv8 training, which outputs growth state data through the following steps: Image preprocessing: Gaussian filter is used to remove noise, and histogram equalization is used to enhance color contrast; Fruit detection: Identify the number of fruits and the outline of each fruit in the image, and exclude interference such as leaves and branches; State recognition: Extract the RGB color features of the fruit (such as coloration degree = red area area / total fruit area x 100%), and divide the maturity level (unripe: coloration degree <30%; semi-ripe: 30%-70%; ripe: >70%), and detect defects on the fruit surface (such as spots and cracks, with an identification accuracy of ≥95%); Data output: The coloration degree, maturity level, and defect rate are packaged into JSON format and transmitted to the core controller.

[0031] For example, Figure 3As shown, the data transmission unit includes a wireless transmission module and an edge computing node; the wireless transmission module is used for transmitting data at a preset frequency, and triggering immediate transmission when the illumination parameter is abnormal, the spectral distribution is abnormal, or the fruit growth state data changes significantly; the edge computing node is used for preliminary screening, abnormality judgment of the illumination parameter and fruit growth state data, and generating preliminary warning information; hierarchical transmission of monitoring data, edge nodes and cloud servers is realized, and the specific implementation is as follows: Wireless transmission module: LoRa module (model SX1278) is selected, working frequency band is 433MHz, transmission distance is 0.5-1.5km (line of sight), communication rate is 9.6kbps, and the following transmission strategy is adopted: Regular transmission: the illumination parameters (intensity, spectrum, duration) of each part of the fruit tree and the fruit growth state data are packaged and transmitted to the edge computing node at a preset frequency (1 time / 10 minutes); Immediate transmission: when the illumination intensity of any part of the fruit tree is <20000 lux or >100000 lux (triggering illumination abnormality), the red light ratio is <20% (triggering spectral imbalance), and the fruit color degree changes by >15% in a single day (triggering significant growth state change), immediate transmission is triggered to ensure that abnormal data is reported in real time.

[0032] Edge computing node: an industrial gateway (model EC200S, equipped with Linux system) is adopted, and a data preprocessing algorithm is built-in, and the specific functions include: Data screening: eliminate obvious abnormal values (such as illumination intensity >300000 lux is regarded as sensor fault data), and keep valid data; Abnormality judgment: compare the real-time illumination parameters of each part of the fruit tree with the preset threshold value (such as the illumination intensity threshold value of the middle part of the crown layer in the coloring period is 30000-80000 lux), and generate preliminary warning information (such as “fruit tree back shadow light deficiency warning”); Data forwarding: the screened valid data and warning information are uploaded to the cloud server through the 4G module (model ME909s-821), and the data of nearly 7 days is stored locally (to prevent loss due to network interruption).

[0033] For example Figure 4As shown, the intelligent control unit includes a cloud server and a terminal module for user interaction; the cloud server is used to store historical data and construct the light-chrominance correlation model through a random forest algorithm, which can dynamically output suitable light parameters according to the variety and growth stage of the fruit trees; the terminal module supports light parameter display, abnormal alarm receiving, remote control of the light regulation execution unit, and entry of fruit tree farming operation records; the abnormal alarm includes one or more of insufficient light alarm, spectral imbalance alarm, and fruit growth abnormality alarm, thereby realizing an integrated closed loop of data analysis, instruction generation, and user interaction.

[0034] Wherein, the light-chrominance correlation model is constructed, including: Data preprocessing: after the cloud server receives the data uploaded by the edge node, the light parameters of each part of the fruit tree are standardized by Z-Score (to eliminate the dimension effect), and the missing data (such as 1-2 data points missing due to network interruption) are supplemented by interpolation method; Feature extraction: select "daily average light intensity of each part of the fruit tree, red light proportion, daily effective light duration, fruit growth stage (judged by accumulated temperature: early / middle / late coloring period), and light demand coefficient corresponding to tree age" as the model input features, and "fruit chrominance" as the output target; Model training: using the random forest algorithm, set the number of decision trees to 100, the maximum depth to 10 layers, and the minimum sample split number to 5, with the fruit tree basic information (such as variety: red Fuji apple requires red light proportion of 35%-45%; tree age: 8-year-old mature tree requires slightly higher light intensity than 3-year-old young tree) as the constraint condition, divide 80% of the data as the training set and 20% as the test set, and train the model; Precision verification: use the coefficient of determination R 2 Evaluate the precision of the model, requiring R 2 ≥0.85 (if not up to standard, adjust the number of decision trees to 120, the maximum depth to 12 layers, and retrain), and the final model can output the suitable light parameter range of each part of the fruit tree (such as light intensity 40000-70000 lux, red light proportion 35%-45%, and effective duration ≥10h / day in the middle of the crown layer in the middle coloring period) according to the input real-time feature parameters.

[0035] In addition, the regulation instruction generation includes: Real-time judgment: the cloud server receives the real-time light parameters of each part of the fruit tree once every 5 minutes, compares them with the suitable range output by the model, generates light supplement instructions if the real-time light intensity of the crown back is <40000 lux, generates sunshade instructions if the top light intensity is >70000 lux, and generates spectral adjustment instructions if the red light proportion of any part of the fruit tree is <35%; Instruction encapsulation: format the instruction as module ID + control parameters + execution duration (such as light supplement module 01 (shady side of fruit trees) + intensity 50000 lux + red light ratio 40% + execution 2h), and issue it to the edge computing node through the 4G network, and then forward it to the light control execution unit by the edge node.

[0036] Furthermore, the functions of the terminal module are: Parameter display: real-time display of light parameters (intensity, spectrum, duration) of each part of the fruit tree through the Web or APP, and the data update frequency is 1 / 10 minutes; Abnormal alarm: when the edge node receives a preliminary warning or model judges that the parameters are abnormal, the terminal pops up a pop-up window and sends a message (such as insufficient light on the shady side of Fuji apple trees, current intensity 18000 lux), and the alarm types include insufficient light, spectrum imbalance, and abnormal fruit growth (such as no change in color for 3 days); Remote control: users can manually send control instructions (such as turning on the light reflection module 02 (east side of fruit trees), angle 30°), and the priority of the instruction is higher than that of the automatic instruction; Farming record: support for entering pruning time, fertilization type and other farming operations, and data association to corresponding growth stages for subsequent model optimization.

[0037] As shown in Figure 5 The light control execution unit includes a light supplement module for supplementing light, a sunshade module for adjusting strong light, a light reflection module for enhancing local light, and a bionic light adjustment module for adaptive adjustment of light. The light supplement module is used to adjust the light intensity and spectrum ratio; the sunshade module is used to adjust the sunshade degree according to the light intensity; the light reflection module is used to adjust the laying position and angle to change the light reflection direction; the bionic light adjustment module is used to automatically adjust the light transmittance or reflectivity according to the light condition. Each module receives the instruction issued by the edge node through the RS485 bus, and the specific structure and execution logic are as follows: Light supplement module: LED light supplement panel (size 30cmx20cm) is adopted, which contains red light LED (660nm, power 1W / pea) and blue light LED (450nm, power 1W / pea), and the number ratio can be adjusted (3:1 to 5:1), and the total power is 10-15W; the light intensity adjustment (20000-80000lux) is realized through PWM dimming circuit, after receiving the light supplement instruction, the intensity + spectrum ratio set by the instruction is started, such as intensity 50000 lux + red light ratio 40%, corresponding to 30 red light LEDs and 10 blue light LEDs, and the light supplement duration is automatically turned off after the end.

[0038] Sun-shading module: composed of an electric sun-shading net (material: polyethylene, light-shading rate: 50%-80% adjustable, size: suitable for the diameter of the fruit tree canopy) and a stepper motor (model: 28BYJ-48), the motor drives the sun-shading net reel through a speed reducer, and after receiving the sun-shading instruction, the light-shading rate is adjusted according to the light intensity: when the intensity is 70,000-80,000 lux, the light-shading rate is 50%; when the intensity is 80,000-100,000 lux, the light-shading rate is 70%; when the intensity is >100,000 lux, the light-shading rate is 80%, and the adjustment accuracy is ±5%.

[0039] Reflective module: an aluminized film reflective board (size: 50cmx30cm, reflectivity: ≥85%) is used, and an electric push rod (model: XTL100, stroke: 100mm) is installed at the bottom to adjust the angle of the reflective board (0-60°); after receiving the reflective instruction, the angle is adjusted according to the position of the fruit canopy, for example, the angle is set to 30° in the middle of the coloring period to focus the reflected light on the back of the fruit (traditional reflective film cannot adjust the angle, which easily leads to insufficient coloring on the back), and the local light intensity is increased (by 20%-30%).

[0040] Bionic light adjustment module: imitating the light transmission characteristics of leaves, an electrochromic glass (size: 0.5mx0.5m, light transmission rate: 20%-80% adjustable) is used, and the light transmission rate is changed by applying different voltages (0-5V); when the light intensity fluctuates (such as sudden sunny weather after cloudy weather), the module receives the light intensity data in real time and automatically adjusts the light transmission rate: when the intensity is <30,000 lux, the light transmission rate is 80%; when the intensity is 30,000-80,000 lux, the light transmission rate is 50%; when the intensity is >80,000 lux, the light transmission rate is 20%, and the response time is <1s, avoiding the influence of sudden changes in light on fruit coloring.

[0041] In addition, with reference to Figure 6 The embodiment also provides a fruit tree light regulation and control cultivation method for improving coloring degree, which is applied to the fruit tree light regulation and control cultivation monitoring system and comprises the following steps: S1, deploying and debugging the light monitoring unit, the data transmission unit, the intelligent control unit and the light regulation and control execution unit, and inputting fruit tree basic information; The fruit tree basic information comprises fruit tree variety, tree age, canopy structure, planting position and the like.

[0042] S2, collecting fruit tree light parameters and fruit growth state data through the light monitoring unit, and transmitting the data to the intelligent control unit through the data transmission unit; The fruit tree light parameters cover light data of multiple positions of the fruit tree canopy, such as the top, the middle and the shady side.

[0043] S3, the intelligent control unit analyzes the collected data, and constructs a light-chroma correlation model in combination with the fruit tree basic information; S4, based on the light-chroma correlation model, the fruit tree light parameters are dynamically monitored and judged, and a light regulation instruction is generated, and the corresponding light regulation operation is executed by the light regulation execution unit; S5, in the key period of fruit coloring, the light regulation parameters are optimized based on the light-chroma correlation model, and the control of the light condition of the fruit tree is strengthened; S6, after the fruit is picked, the light regulation effect is evaluated by collecting the fruit related data, and the light-chroma correlation model is optimized based on the evaluation result.

[0044] Specifically, in S3, the process of constructing the light-chroma correlation model includes: The collected light parameters and fruit growth state data are preprocessed, including eliminating sensor failure data, supplementing missing data by interpolation method, and eliminating dimension influence by Z-Score standardization; The characteristic parameters affecting fruit coloring (including daily average light intensity of each part of the fruit tree, red light proportion, daily effective light duration, growth stage (judged by accumulated temperature, coloring early / middle / late), light demand coefficient corresponding to tree age, etc.) are extracted as model input; The fruit chroma is taken as the output target, and the light-chroma correlation model is generated by random forest algorithm training; The accuracy of the light-chroma correlation model is verified, if the accuracy does not reach the preset requirement (such as the determination coefficient R 2 <0.85), the model parameters (such as the number of decision trees, the maximum depth, and the minimum sample division number) are adjusted and retrained.

[0045] Specifically, in S4, based on the light-chroma correlation model, the fruit tree light parameters are dynamically monitored and judged, and a light regulation instruction is generated, and the corresponding light regulation operation is executed by the light regulation execution unit, which specifically includes: The fruit tree light parameters (light data of different parts of the crown are collected every 5 minutes) and fruit growth state data are continuously collected and transmitted to the intelligent control unit; Based on the light-chroma correlation model, it is judged whether the real-time light parameters meet the preset range; According to the judgment result, the corresponding light regulation instruction is generated, the light regulation instruction includes one or more of the light compensation instruction (for the shaded or insufficient light area of the fruit tree), the sunshade instruction (for the strong light area of the top of the fruit tree), and the spectrum adjustment instruction (adapted to the specific spectrum required for fruit coloring); The light regulation execution unit receives and executes the light regulation instruction, and feeds back the actual light parameter every 30 minutes during execution to ensure the regulation accuracy.

[0046] Specifically, in S5, during the fruit coloring critical period, the light regulation parameter is optimized based on the light-coloring degree correlation model, and the control of the light condition of the fruit tree is strengthened, specifically including: Based on the light-coloring degree correlation model, the suitable light parameter range of the coloring critical period is adjusted (such as the suitable light intensity of 45,000-75,000 lux and the red light ratio of 40%-50% for adult red Fuji apple during the coloring critical period); The angle of the light reflection module is controlled to adjust (such as 30-35°), and the reflected light is focused on the back area of the fruit of the fruit tree to enhance the local light amount; The fruit coloring data is continuously collected (the coloring degree data is obtained by the AI vision module every 15 minutes), and the light regulation operation is dynamically adjusted according to the coloring data (such as appropriately increasing the light supplement intensity and prolonging the light supplement time length when the coloring progress is slow).

[0047] Specifically, in S6, after the fruit is picked, the light regulation effect is evaluated by collecting fruit-related data, and the light-coloring degree correlation model is optimized based on the evaluation result, specifically including: The fruit coloring-related indicators (such as single fruit coloring degree and coloring uniformity rate) and quality indicators (such as sugar content, hardness, and anthocyanin content) are collected as effect evaluation data; The effect evaluation data is compared with the preset standard (such as the coloring degree of high-quality fruit ≥85% and the coloring uniformity rate ≥90%) or the indicators of single tree under the traditional management mode to judge the light regulation effect; According to the effect judgment result, the light-coloring degree correlation model is adjusted in parameters (such as correcting the light demand coefficient of different tree ages) or retrained (supplementing the fruit data of the picking season to the training set) to realize the iterative optimization of the model.

[0048] The feasibility of the above system and method is demonstrated by a specific experiment, specifically: First, each component is installed in the way of deploying the monitoring unit in multiple parts of single fruit tree (1 group on the top, middle sunny side, middle shady side, and bottom of the crown layer), and installing 2 miniature light supplement modules on the shady side of the crown layer, and connecting the power supply and communication line; The terminal module inputs the basic information of the fruit tree: fruit tree variety (red Fuji apple), tree age (8 years old), crown layer structure (sparse type, crown diameter 3m), coloring critical period (every year from September to October), and historical coloring degree standard (the coloring degree of high-quality fruit ≥85%); Debugging each unit: Start the light monitoring unit, calibrate the sensors on each part of the fruit tree with a standard light source (50000 lux, 40% red light), the error should be <3%; Start the control execution unit, send light supplement instructions to test the light supplement intensity and spectral ratio, ensure consistency with the instructions; Test the data transmission link to ensure that the communication success rate between the edge node and the cloud and the terminal is ≥99%.

[0049] Subsequently, the light monitoring unit transmits data in real time once every 10 minutes under normal circumstances and immediately under abnormal circumstances, and the edge node uploads the data to the cloud after preprocessing. For example, on September 5, the data for the Red Fuji fruit tree was as follows: the light intensity at the top of the canopy was 55000 lux, the light intensity on the sunny side of the middle part was 48000 lux, the light intensity on the shady side of the middle part was 32000 lux, the light intensity at the bottom was 20000 lux, the red light ratio was 38%, the effective duration was 10.5h, and the fruit coloration degree was 65% (semi-mature); based on the data collected in the past month (including data under different weather conditions and in different growth stages), the cloud server constructed a model according to the preprocessing, feature extraction, training, and verification process. The final output of the model for the 8-year-old Red Fuji apple in the coloration period was as follows: the suitable light intensity in the middle part of the canopy was 40000-70000 lux, the red light ratio was 35%-45%, and the effective duration was ≥10h per day.

[0050] Then, on September 10, the real-time light intensity on the shady side of the canopy of the fruit tree was 28000 lux (lower than the lower limit of the suitable range), and the cloud model generated a light supplement instruction after judgment: light supplement module 01-02 (located on the shady side of the fruit tree), intensity 50000 lux, red light ratio 40%, execution time 3h, the edge node received the instruction and sent it to the light supplement module, and the actual light intensity was fed back every 30 minutes during the execution period to ensure the control accuracy; On September 20, the fruit entered the late coloration period (the coloration degree should be ≥80%), and the model adjusted the suitable range as follows: the light intensity in the middle part of the canopy was 50000-75000 lux, the red light ratio was 40%-50%, and a light reflection instruction was generated: light reflection module 01-02 (located on both sides of the root of the fruit tree), angle 35°, to enhance the light on the back of the fruit, and after 3 days, the coloration degree was increased to 82%; Finally, on October 15, after the fruit was picked, the data for the fruit of the fruit tree was collected as follows: the average coloration degree was 88% (under traditional management, the average coloration degree was 72%), the coloration uneven rate was 5% (under traditional technology, the coloration uneven rate was 18%), and the sugar content was 15.2 Brix (under traditional technology, the sugar content was 13.5 Brix); by comparing the data with the traditional management, it was determined that the control effect was "excellent", and then the model was optimized by retraining the model with the data collected during the picking season.

[0051] Therefore, by using the fruit tree light regulation cultivation method for improving color depth and the monitoring system, the problems of uneven fruit color, low light regulation precision, lack of real-time monitoring and intelligent correlation model, and single function of the regulation execution unit in the current fruit tree cultivation which relies on natural light or traditional manual intervention are solved through multi-parameter light and fruit growth state monitoring, light-color depth correlation model construction, and multi-module light regulation.

[0052] The principles and implementation modes of the present application are described herein by using specific examples, and the above examples are only used to help understand the method of the present application and its core idea; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation modes and application ranges will be changed. In summary, the content of the present application should not be understood as a limitation of the present application.

Claims

1. A fruit tree light regulation cultivation monitoring system for improving coloring degree, characterized by, The application relates to a fruit tree light regulation system, which comprises a light monitoring unit, a data transmission unit, an intelligent control unit and a light regulation execution unit. The light monitoring unit is used for collecting fruit tree light parameters and fruit growth state data. The data transmission unit is connected with the light monitoring unit and is used for transmitting the light parameters and fruit growth state data to the intelligent control unit. The intelligent control unit is used for analyzing and processing the light parameters and fruit growth state data, constructing a light-chroma correlation model and generating light regulation instructions based on the light-chroma correlation model. The light regulation execution unit is connected with the intelligent control unit and is used for receiving the light regulation instructions and executing corresponding light regulation operations.

2. The fruit tree light regulation cultivation monitoring system for improving coloring degree according to claim 1, characterized in that, The fruit tree light parameters collected by the light monitoring unit include one or more of light intensity, spectral distribution and light duration. The light monitoring unit comprises a light intensity sensor for collecting light intensity, a spectral sensor for collecting spectral distribution, a light duration recorder for recording light duration and an AI vision monitoring module for collecting fruit images and identifying fruit growth states, wherein the AI vision monitoring module is used for identifying fruit maturity, color change and fruit surface defects.

3. The fruit tree light regulation cultivation monitoring system for improving coloring degree according to claim 1, characterized in that, The data transmission unit comprises a wireless transmission module and an edge computing node. The wireless transmission module is used for transmitting data at a preset frequency and triggering instant transmission when the light parameters are abnormal, the spectral distribution is abnormal or the fruit growth state data changes significantly. The edge computing node is used for preliminarily screening and abnormally judging the light parameters and fruit growth state data and generating preliminary warning information.

4. The fruit tree light regulation cultivation monitoring system for improving coloring degree according to claim 1, characterized in that, The intelligent control unit comprises a cloud server and a terminal module for user interaction. The cloud server is used for storing historical data and constructing the light-chroma correlation model through a random forest algorithm, wherein the light-chroma correlation model can dynamically output suitable light parameters according to fruit tree varieties and growth stages. The terminal module supports light parameter display, abnormal alarm receiving, remote control of the light regulation execution unit and entry of fruit tree farming operation records. The abnormal alarm comprises one or more of light deficiency alarm, spectral imbalance alarm and fruit growth abnormality alarm.

5. The fruit tree light regulation cultivation monitoring system for improving coloring degree according to claim 1, characterized in that, The light regulation execution unit comprises a light supplement module for supplementing light, a sunshade module for adjusting strong light, a light reflection module for enhancing local light and a bionic light regulation module for adaptively adjusting light. The light supplement module is used for adjusting light intensity and spectral proportion. The sunshade module is used for adjusting the sunshade degree according to light intensity. The light reflection module is used for adjusting the laying position and angle to change the light reflection direction and focus on the fruit tree canopy and fruit area. The bionic light regulation module is used for automatically adjusting the light transmittance or reflectivity according to light conditions to adapt to the light requirements of different growth stages of the fruit tree.

6. The method for improving the color depth of fruit trees by light regulation, applied to the monitoring system for improving the color depth of fruit trees by light regulation according to any one of claims 1-5, characterized in that, The application further relates to a fruit tree light regulation method, which comprises the following steps: S1, completing the deployment and debugging of the light monitoring unit, the data transmission unit, the intelligent control unit and the light regulation execution unit and entering fruit tree basic information; S2, collecting fruit tree light parameters and fruit growth state data by the light monitoring unit, transmitting the data to the intelligent control unit through the data transmission unit; S3, the intelligent control unit analyzes the collected data, and constructs a light-chroma correlation model combined with the basic information of the fruit tree; S4, based on the light-chroma correlation model, the fruit tree light parameters are dynamically monitored and judged, and the light control instruction is generated, and the corresponding light control operation is executed by the light control execution unit; S5, during the key period of fruit coloring, the light control parameters are optimized based on the light-chroma correlation model, and the control of the light conditions of the fruit tree is strengthened; S6, after the fruit is picked, the light control effect is evaluated by collecting fruit related data, and the light-chroma correlation model is optimized based on the evaluation result.

7. The method for improving the color intensity of fruit trees by light regulation according to claim 6, characterized in that, In S3, the process of constructing the light-chroma correlation model includes: preprocessing the collected fruit tree light parameters and fruit growth state data; extracting the characteristic parameters affecting fruit coloring as model input; training the light-chroma correlation model by random forest algorithm with fruit chroma as output target; verify the accuracy of the light-chroma correlation model, if the accuracy does not meet the preset requirement, adjust the model parameters and retrain.

8. The method for improving the color intensity of fruit trees by light regulation according to claim 6, characterized in that, In S4, based on the light-chroma correlation model, the fruit tree light parameters are dynamically monitored and judged, and the light control instruction is generated, and the corresponding light control operation is executed by the light control execution unit, specifically including: continuously collecting fruit tree light parameters and fruit growth state data and transmitting to the intelligent control unit; based on the light-chroma correlation model, combined with the suitable light parameter range corresponding to the fruit tree variety and growth stage, judge whether the real-time light parameter meets the preset range; generate corresponding light control instruction according to the judgment result, the light control instruction includes one or more of light supplement instruction, shading instruction, spectrum adjustment instruction; the light control execution unit receives and executes the light control instruction.

9. The method for improving the color intensity of fruit trees by light regulation according to claim 6, characterized in that, In S5, during the key period of fruit coloring, the light control parameters are optimized based on the light-chroma correlation model, and the control of the light conditions of the fruit tree is strengthened, specifically including: based on the light-chroma correlation model, adjust the suitable light parameter range of the fruit tree coloring key period; control the reflection module to enhance the light intensity of the local area of the fruit tree; continuously collect fruit coloring data and dynamically adjust the light control operation according to the coloring data.

10. The method for improving the color intensity of fruit trees by light regulation according to claim 6, characterized in that, In S6, after the fruit is picked, the light control effect is evaluated by collecting fruit related data, and the light-chroma correlation model is optimized based on the evaluation result, specifically including: collect fruit coloring related indicators and quality indicators as effect evaluation data; compare the effect evaluation data with the preset standard or the indicators of single fruit tree under traditional management mode to judge the light control effect; according to the effect judgment result, adjust the parameters of the light-chroma correlation model or retrain to realize the iteration optimization of the model.

Images