Intelligent and rapid pre-examination and diagnosis method and system for tree health

Abstract

The invention provides an intelligent and rapid pre-examination and diagnosis method and system for tree health, and relates to the technical field of tree management. The method comprises the following steps: constructing a liquid flow database of a standard tree of a target tree species, carrying out model training on an established liquid flow calculation model by utilizing the liquid flow database, then collecting liquid flow data of a to-be-detected target tree in a preset detection time and meteorological and soil information of a growth environment, inputting an optimal liquid flow calculation model, and calculating a theoretical liquid flow value of the target tree, comparing with an actually monitored liquid flow value, calculating a liquid flow deviation degree, grading according to a daily maximum liquid flow deviation degree, and comprehensively evaluating a tree health grade in combination with phenotype information to obtain a diagnosis result; the rapid and intelligent pre-examination diagnosis of the tree health is realized, and the intervention timeliness is improved; and multi-dimensional data fusion is adopted, so that the diagnosis accuracy is improved, and automatic monitoring and real-time early warning are realized.

Description

Technical Field

[0001] This invention relates to the technical field of tree management, and in particular to a method and system for intelligent and rapid pre-detection and diagnosis of tree health. Background Technology

[0002] With the acceleration of urbanization, urban green spaces play an important role in improving urban climate, protecting biodiversity, reducing noise and preventing soil erosion, and directly affect the aesthetics, ecological balance and quality of life of urban residents. Therefore, the importance of rapid and intelligent diagnosis and refined management of the health status of trees in green spaces is becoming increasingly prominent.

[0003] However, urban trees often suffer health damage due to transplantation, environmental stress, and pests and diseases. Traditional methods for diagnosing tree health mainly rely on manual observation based on tree morphology and appearance (such as leaf color, withered branches, and trunk damage) or physical detection (such as detecting the degree of hollow decay). These methods have the following limitations: 1. Lag: External phenotypic symptoms often only appear after internal physiological imbalances, leading to delayed diagnosis and missing the optimal intervention window. 2. Subjectivity: In urban landscaping, transplanted trees are subject to changes in the planting environment, making manual observation susceptible to experience-based biases, difficult to quantify, and unable to distinguish between "false death" and "false survival" trees (e.g., trees whose root systems have not recovered after transplantation but whose branches and leaves are temporarily normal). 3. Limitations of single indicators: Relying solely on phenotypic changes or detecting the degree of hollow decay makes it difficult to quickly and accurately determine tree health, and also fails to comprehensively reflect key physiological states such as water metabolism and environmental adaptability.

[0004] In recent years, sap flow monitoring technology has been used to study the patterns of water transport and transpiration in trees. Water is a major factor affecting plant life activities, and the characteristics of water transport and consumption in plants can effectively reflect their adaptability to the environment. Sap flow refers to the process by which the transpiration pull drives the tree roots to absorb soil water and transport it upwards along the xylem under the influence of plant transpiration. Monitoring sap flow parameters in tree growth is an important basis for estimating tree transpiration water use and diagnosing tree water status.

[0005] However, existing technologies are mostly used in eco-hydrological research and have not yet formed an intelligent tree health diagnosis method that combines environmental factors, sap flow dynamics, and phenotypic information. Therefore, there is an urgent need for a comprehensive diagnostic technology that can provide early warning and objectively quantify the assessment of tree health status. Summary of the Invention

[0006] The main objective of this invention is to provide a method and system for intelligent and rapid pre-detection and diagnosis of tree health, which solves the technical problems of lag, subjectivity and limitation of single indicators in the prior art.

[0007] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a method for intelligent and rapid pre-detection and diagnosis of tree health, comprising the following steps: S1: Collect sap flow data of standard trees and meteorological and soil information of the growth environment during a preset time period to construct a sap flow database of standard trees of the target tree species; S2: Based on machine learning algorithms, establish a sap flow calculation model for the target tree species, train the model using a sap flow database, and save the optimal sap flow calculation model; wherein the sap flow calculation model uses meteorological and soil information as independent variables and sap flow magnitude as the dependent variable; S3: Collect phenotypic information, sap flow data, and meteorological and soil information of the growth environment of the target trees to be tested within a preset detection time; the phenotypic information includes the degree of crown deviation, the degree of trunk damage, the degree of branch dieback, the degree of main body tilt, and the threat of diseases and pests; S4: Input the sap flow data of the target tree to be detected and the meteorological and soil information of the growing environment into the optimal sap flow calculation model, calculate the theoretical sap flow value of the target tree, compare it with the actual monitored sap flow value, calculate the sap flow deviation, and classify it according to the maximum daily sap flow deviation. S5: Combining sap flow deviation and phenotypic information, comprehensively evaluate the health level of trees and obtain diagnostic results.

[0008] In the preferred embodiment, the standard timber must meet the following conditions: normal shape, no visible damage, no pests or diseases, and good growth condition, specifically: The criteria for judging normal morphology include trunk uprightness, bark integrity, canopy symmetry, and branch dieback rate all meeting their respective preset standards. The criteria for determining whether a plant is free of pests and diseases are visual screening or pest monitoring.

[0009] In the preferred embodiment, the expression for the fluid flow calculation model is: ; In the formula, Let the fluid density be... For meteorological and soil information.

[0010] In the preferred embodiment, the fluid density is calculated using Granier's empirical formula, which is as follows: ; In the formula, Liquid flow density, which is the flow rate of liquid passing through a unit area of ​​edge material per unit time. To adapt to instantaneous temperature differences, The maximum temperature difference between 0:00 and 24:00 is the diurnal temperature range, and 3600 is the time conversion coefficient for converting instantaneous flow rate to hourly flow rate.

[0011] In the preferred embodiment, step S3 further includes: monitoring meteorological and soil data of the target tree's growth environment, and issuing an early warning when an anomaly is detected, wherein the anomaly is specifically: Air temperature greater than 40.9℃, wind speed greater than 28.9m / s, soil pH less than 4 or greater than 10, soil temperature higher than 35℃, soil moisture less than 30% or greater than 90%, and soil electrical conductivity greater than 4dS / m.

[0012] In the preferred embodiment, the meteorological information includes air temperature, relative humidity, effective solar radiation, wind direction, wind speed, and saturated vapor pressure; The soil information includes soil temperature, moisture content, electrical conductivity, pH value, and salinity.

[0013] In the preferred embodiment, step S4 involves grading based on the daily maximum liquid flow deviation, specifically as follows: ; In the formula, D represents the deviation degree.

[0014] In the preferred embodiment, S3 uses a sap flow measuring device to collect phenotypic information of the target tree to be tested, and uses a meteorological and soil measuring device to collect meteorological and soil information of the target tree to be tested. The sap flow measuring device is installed 1.3m north of the target tree, and the meteorological and soil measuring device is installed within 5m of the target tree. The installation positions of both the fluid flow measuring device and the image and soil measuring device were verified through HIL testing to determine their optimal installation positions.

[0015] In a preferred embodiment, the machine learning algorithm in S2 includes, but is not limited to, random forest, support vector machine, or neural network algorithm.

[0016] Secondly, a tree health intelligent rapid pre-detection and diagnosis system is provided, including: The data acquisition module is used to collect sap flow data of standard trees over a preset time period, as well as meteorological and soil information of the growing environment, to construct a sap flow database of standard trees of the target tree species. The data processing module is used to establish a sap flow calculation model for the target tree species based on machine learning algorithms, train the model using a sap flow database, and save the optimal sap flow calculation model; wherein the sap flow calculation model uses meteorological and soil information as independent variables and sap flow magnitude as the dependent variable; The test data module is used to collect phenotypic information, sap flow data, and meteorological and soil information of the growth environment of the target trees within a preset detection time. The phenotypic information includes the degree of crown imbalance, the degree of trunk damage, the degree of branch dieback, the degree of main body tilt, and the threat of pests and diseases. The deviation module takes the sap flow data of the target tree to be detected and the meteorological and soil information of the growing environment as input into the optimal sap flow calculation model, calculates the theoretical sap flow value of the target tree, compares it with the actual monitored sap flow value, calculates the sap flow deviation, and classifies it according to the maximum daily sap flow deviation. The diagnostic module is used to comprehensively evaluate the health level of trees by combining sap flow deviation and phenotypic information, and obtain diagnostic results.

[0017] This invention provides a method for intelligent and rapid pre-detection and diagnosis of tree health. It constructs a sap flow database of standard trees of a target species and uses this database to train a sap flow calculation model. Then, it collects phenotypic information, sap flow data, and meteorological and soil information of the target trees within a preset detection time period. This data is input into the optimal sap flow calculation model to calculate the theoretical sap flow value of the target trees. This theoretical value is compared with the actual monitored sap flow value to calculate the sap flow deviation. Based on the deviation, the trees are graded, and the overall health level is comprehensively evaluated to obtain a diagnostic result. This method achieves rapid and intelligent pre-detection and diagnosis of tree health, improving the timeliness of intervention. The use of multi-dimensional data fusion improves diagnostic accuracy and enables automated monitoring and real-time early warning.

[0018] It has the following significant advantages: 1. Early warning is achieved, and the timeliness of intervention is improved: Through sap flow deviation analysis, abnormal water transport can be detected before the appearance of tree symptoms, and stress (such as drought, root damage, and latent period of pests and diseases) can be detected in advance, which greatly advances the maintenance and management window.

[0019] 2. The use of multi-dimensional data fusion improves diagnostic accuracy: By combining three types of information—fluid flow (physiological), environment (meteorology / soil), and phenotype (morphology)—a machine learning model is used to establish the relationship between theoretical fluid flow values ​​and environmental factors, reducing the risk of misjudgment based on a single indicator. This is especially suitable for identifying the "false life / false death" state of transplanted trees.

[0020] 3. It achieves intelligent and standardized assessment, improving applicability: It realizes objective quantitative evaluation through machine learning models, avoiding subjective bias of human experience, and provides a scientific basis for refined maintenance (such as irrigation and pest control); it can also be applied to the protection of ancient and famous trees, urban greening tree management and ecological restoration projects, and is particularly suitable for the adaptability monitoring of large-sized transplanted trees, reducing tree mortality.

[0021] 4. It realizes automated monitoring and real-time early warning, and improves response efficiency: Data is collected in real time through IoT sensors, and early warning is automatically triggered in combination with environmental anomaly thresholds (such as extreme temperature, soil salinization). Attached Figure Description

[0022] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1 This is a flowchart of the pre-examination and diagnosis method of the present invention; Figure 2 This is a flowchart of the standard wood requirement judgment process of the present invention. Detailed Implementation

[0023] Example 1 like Figure 1-2 As shown, a smart and rapid pre-detection and diagnosis method for tree health includes the following steps: S1: Collect sap flow data of standard trees over a preset time period, as well as meteorological and soil information of the growing environment, to construct a sap flow database of standard trees of the target tree species.

[0024] S2: Based on machine learning algorithms, establish a sap flow calculation model for the target tree species, train the model using a sap flow database, and save the optimal sap flow calculation model; wherein the sap flow calculation model uses meteorological and soil information as independent variables and sap flow magnitude as the dependent variable.

[0025] S3: Collect phenotypic information, sap flow data, and meteorological and soil information of the target trees to be tested within a preset detection time. Phenotypic information includes the degree of crown imbalance, trunk damage, branch dieback, main body tilt, and pest and disease threats.

[0026] S4: Input the sap flow data of the target tree to be detected and the meteorological and soil information of the growing environment into the optimal sap flow calculation model, calculate the theoretical sap flow value of the target tree, compare it with the actual monitored sap flow value, calculate the sap flow deviation, and classify it according to the maximum daily sap flow deviation. Among them, trees with abnormal phenotype, poor growth status or obvious diseases and pests are directly judged as unhealthy.

[0027] S5: Combining sap flow deviation and phenotypic information, comprehensively evaluate the tree health level and obtain diagnostic results. Specifically, for trees that cannot be determined solely by tree phenotype, calculate the tree sap flow deviation. That is, input the sap flow data of the target tree to be tested and the meteorological and soil information of the growing environment into the optimal sap flow calculation model, calculate the theoretical sap flow value of the target tree, compare it with the actual monitored sap flow value, and classify the tree health according to the deviation.

[0028] This embodiment constructs a sap flow database of standard trees of the target species and uses it to train a sap flow calculation model. Then, it collects phenotypic information, sap flow data, and meteorological and soil information of the target trees within a preset detection time period. This data is input into the optimal sap flow calculation model to calculate the theoretical sap flow value of the target trees. This theoretical value is compared with the actual monitored sap flow value to calculate the sap flow deviation. Based on the deviation, the trees are graded, and the overall health level is comprehensively evaluated to obtain a diagnostic result. This achieves rapid and intelligent pre-detection and diagnosis of tree health, improving the timeliness of intervention. The use of multi-dimensional data fusion improves diagnostic accuracy and enables automated monitoring and real-time early warning.

[0029] The preset time period used in this embodiment is 1 year, and the preset detection time is 72 consecutive hours.

[0030] In the preferred embodiment, the machine learning algorithm in step S2 includes, but is not limited to, random forest, support vector machine, or neural network algorithm.

[0031] In the preferred solution, such as Figure 2 As shown, standard timber must meet the following conditions: normal shape, no visible damage, no pests or diseases, and good growth.

[0032] The specific content includes: 1) The conditions for judging normal morphology include the trunk uprightness, bark integrity, canopy symmetry, and branch dieback rate all meeting their respective preset standards.

[0033] The standards used in this embodiment are shown in Table 1.

[0034] Table 1. Data Table for Judging Normal Morphology

[0035] 2) Criteria for determining the absence of pests and diseases: Visual screening or pest monitoring, including checking leaves (insect holes, disease spots), trunks (boring holes, gum exudation), etc.

[0036] This embodiment combines IoT sensor data to replace manual experience-based judgment, thereby improving the reliability of standard wood data.

[0037] In the preferred scheme, the expression for the fluid flow calculation model is: ; In the formula, Let the fluid density be... For meteorological and soil information, For machine learning models.

[0038] In the preferred embodiment, the fluid density is calculated using Granier's empirical formula, which is as follows: ; In the formula, Liquid flow density ( This refers to the liquid flow rate per unit time through a unit area of ​​edge material. To adapt to instantaneous temperature differences, The maximum temperature difference between 0:00 and 24:00 is the diurnal temperature range, and 3600 is the time conversion coefficient for converting instantaneous flow rate to hourly flow rate.

[0039] In the preferred embodiment, step S3 further includes: monitoring meteorological and soil data of the target tree's growth environment; issuing an early warning when environmental parameters exceed the tolerance range of general trees; and using empirical values ​​to synthesize the results of multiple previous studies on abnormal parameters, specifically: Air temperature greater than 40℃, wind speed greater than 13.9m / s, soil pH less than 4 or greater than 10, soil temperature higher than 35℃, soil moisture less than 30% or greater than 90%, and soil electrical conductivity greater than 4dS / m.

[0040] In the preferred embodiment, the meteorological information includes air temperature, relative humidity, effective solar radiation, wind direction, wind speed, and saturated vapor pressure; Soil information includes soil temperature, moisture content, electrical conductivity, pH value, and salinity.

[0041] In the preferred embodiment, the formula for calculating the fluid flow deviation in step S4 is:

[0042] In the formula, J s1 J represents the theoretical value of sap flow from the tree to be tested. s2 The actual monitored sap flow value of the tree to be tested is represented by D, where D represents the deviation. The deviation is graded based on the maximum daily sap flow deviation, as follows: ; In the preferred embodiment, step S3 uses a sap flow measuring device to collect phenotypic information of the target tree to be tested, and uses a meteorological and soil measuring device to collect meteorological and soil information of the target tree to be tested. The sap flow measuring device is installed 1.3m north of the target tree, and the meteorological and soil measuring device is installed within 5m of the target tree. The installation positions of both the fluid flow measuring device and the soil measuring device were verified through HIL testing to determine their optimal installation positions.

[0043] In this embodiment, the standardized installation location of the sap flow measuring device is 1.3m, which is the internationally accepted diameter at breast height (DBH) measurement height, ensuring data comparability. The meteorological and soil measuring devices are installed within 5m of the target tree. The microenvironment is representative: a 5m radius covers the root distribution area and canopy projection range of the target tree, reflecting the true growth environment. To avoid interference, the distance from the trunk is ≥1m to prevent root heat radiation or mechanical damage from affecting sensor accuracy.

[0044] In this embodiment, the HIL test first performs simulation environment modeling: in the HIL test platform, a three-dimensional tree microenvironment model is constructed, including: tree geometric parameters (crown width, root system range), dynamic lighting simulation (sun trajectory, canopy shading) and soil heterogeneity (humidity / temperature gradient distribution).

[0045] Next, virtual sensors are deployed, and multiple sensor location combinations are set in the model, including liquid flow sensors, relevant sensors from weather stations, and soil probes, each set in its corresponding location.

[0046] Dynamic scenario testing was conducted, inputting extreme environmental conditions (high temperature, drought, heavy rain), comparing sensor data from different locations, and then artificially introducing interference (such as obstructing the liquid flow sensor or localized soil drying) to evaluate the sensitivity of the installation location to anomaly detection.

[0047] Finally, the optimal installation location was determined using HIL test data, as shown in Table 2.

[0048] Table 2. HIL Test Data for Determining the Optimal Installation Location

[0049] The installation location, verified through HIL testing, reduced the misdiagnosis rate of Kang Diagnostics, and improved data accuracy and system robustness.

[0050] This embodiment achieves rapid and accurate diagnosis of tree health through the fusion and intelligent analysis of multi-source data from physiology, environment, and morphology, providing an innovative solution for the sustainable management of urban trees.

[0051] For example, by conducting one year of environmental and sap flow monitoring on standard specimens of healthy Shanghai Magnolia denudata trees, a database was constructed, a model was established, and tree health was assessed. Specific applications are as follows: A method for intelligent and rapid pre-detection and diagnosis of tree health, characterized by comprising the following steps: S1: Collect sap flow data of standard Magnolia trees for one year (2022), with a sampling interval of 10 min / time, as well as meteorological and soil information of the growing environment, including air temperature, relative humidity, wind speed, solar radiation, soil temperature, soil volumetric water content, soil electrical conductivity, and soil water potential, and construct a database. S2: A random forest model of sap flow in Magnolia officinalis was established using 75% of the sample data and ten-fold cross-validation. The accuracy was validated using 25% of the independent sample data. The model R... 2 Up to 0.964; S3: Investigate the phenotypic information of the trees to be tested (3 white magnolias in Shanghai World Expo Culture Park). The trees to be tested have normal phenotypic morphology and no obvious diseases or pests. Therefore, it is not possible to directly determine the health status of the trees. S3: Collect sap flow data of the target tree to be tested and meteorological and soil information of its growing environment for at least 24 consecutive hours (including at least one day's data from 00:00 to 24:00); S4: Input the sap flow data of three white magnolia trees to be tested in a park from April 1st to April 2nd, 2024, along with meteorological and soil information about their growing environment, into the sap flow calculation model to calculate the theoretical sap flow value of the target trees. The results are shown in Table 3: Table 3. Data on Flow of Liquid Detected in Magnolia sylvestris

[0052] S5: Calculate the maximum daily sap flow deviation of the three Magnolia sylvestris plants and classify them according to the deviation. The results are shown in Table 4. Table 4. Maximum Deviation and Classification of Daily Liquid Flow in Magnolia Flower

[0053] S6: Combining the fluid flow deviation and phenotypic information, the overall evaluation is that trees 1 and 2 are in normal health, while tree 3 is in slightly abnormal health.

[0054] In practice, this embodiment uses sap flow deviation analysis, taking tree sap flow as a key entry point: the optimal sap flow calculation model trained based on a standard tree database can compare the theoretical sap flow value of the tree under test with the actual monitoring value. Before the tree shows any apparent symptoms, it can accurately identify abnormal water transport, advance the maintenance and management window, achieve early warning, significantly improve the timeliness of intervention, and reduce the risk of tree death due to delayed intervention. The use of multi-dimensional data fusion improves diagnostic accuracy, achieves a combination of standardization and intelligence, and enhances the applicability and objectivity of the assessment. Furthermore, it realizes automated monitoring and real-time early warning, which is suitable for large-scale and wide-ranging tree management scenarios, improving the overall maintenance response efficiency.

[0055] Example 2 To further illustrate with reference to Embodiment 1, a tree health intelligent rapid pre-detection and diagnosis system includes: The data acquisition module is used to collect sap flow data of standard trees over a preset time period, as well as meteorological and soil information of the growing environment, to construct a sap flow database of standard trees of the target species.

[0056] The data processing module is used to establish a sap flow calculation model for the target tree species based on machine learning algorithms, train the model using a sap flow database, and save the optimal sap flow calculation model; wherein the sap flow calculation model uses meteorological and soil information as independent variables and sap flow magnitude as dependent variable.

[0057] The test data module is used to collect phenotypic information, sap flow data, and meteorological and soil information of the growth environment of the target trees within a preset detection time. The phenotypic information includes the degree of crown imbalance, the degree of trunk damage, the degree of branch dieback, the degree of main body tilt, and the threat of pests and diseases.

[0058] The deviation module is used to input the sap flow data of the target tree to be detected, as well as the meteorological and soil information of the growing environment, into the optimal sap flow calculation model, calculate the theoretical sap flow value of the target tree, compare it with the actual monitored sap flow value, calculate the sap flow deviation, and classify it according to the maximum daily sap flow deviation.

[0059] The diagnostic module is used to comprehensively evaluate the health level of trees by combining sap flow deviation and phenotypic information, and obtain diagnostic results.

[0060] The above embodiments are merely preferred technical solutions of the present invention and should not be considered as limitations on the present invention. The scope of protection of the present invention should be limited to the technical solutions described in the claims, including equivalent substitutions of the technical features described in the claims. That is, equivalent substitutions and improvements within this scope are also within the scope of protection of the present invention.

Claims

1. A tree health intelligent rapid pre-diagnosis method, characterized in that, The method comprises the following steps: S1: Collecting sap flow data of standard trees and meteorological and soil information of the growth environment in a preset time period, and constructing a sap flow database of the target tree species; S2: Based on a machine learning algorithm, a sap flow calculation model of the target tree species is established, the model is trained using the sap flow database, and the optimal sap flow calculation model is saved; wherein the sap flow calculation model takes meteorological and soil information as independent variables and sap flow size as dependent variables; S3: Collecting phenotype information, sap flow data and meteorological and soil information of the growth environment of the target tree to be detected within a preset detection time; the phenotype information includes crown inclination, trunk damage degree, branch damage degree, main body inclination degree and pest threat situation; S4: Inputting the sap flow data and meteorological and soil information of the growth environment of the target tree to be detected into the optimal sap flow calculation model, calculating the theoretical sap flow value of the target tree, comparing it with the actual monitoring sap flow value, calculating the sap flow deviation, and grading according to the daily maximum sap flow deviation; S5: Combining the sap flow deviation and the phenotype information, comprehensively evaluating the health grade of the tree, and obtaining the diagnosis result.

2. The intelligent rapid pre-detection diagnosis method for tree health according to claim 1, wherein the standard tree meets the conditions of normal shape, no visible damage, no pests and diseases, and good growth condition, specifically: The normal shape judgment conditions include that the trunk straightness, bark integrity, crown symmetry and branch damage rate all meet their respective preset standards; The pest-free judgment condition is determined by visual screening or pest monitoring.

3. The intelligent rapid pre-detection diagnosis method for tree health according to claim 1, wherein the expression of the sap flow calculation model is:

4. The intelligent rapid pre-detection diagnosis method for tree health according to claim 3, wherein the sap flow density is calculated by the Granier empirical formula, and the calculation formula is:

5. The intelligent rapid pre-detection diagnosis method for tree health according to claim 1, wherein S3 further comprises: monitoring the meteorological and soil data of the growth environment of the target tree, and giving a warning when an abnormality is detected, wherein the abnormality is specifically: ； wherein is the liquid flow density, is weather and soil information. Air temperature greater than 40.9℃, wind speed greater than 28.9m / s, soil pH less than 4 or greater than 10, soil temperature higher than 35℃, soil humidity less than 30% or greater than 90%, and soil conductivity greater than 4dS / m.

6. The intelligent rapid pre-detection diagnosis method for tree health according to claim 1, wherein the meteorological information includes air temperature, relative humidity, solar effective radiation, wind direction, wind speed and saturated water vapor pressure; ； wherein is the liquid flow density, i.e. the liquid flow per unit time through a unit of edge material area, is the instantaneous temperature difference, is 0:00-24:00 diurnal maximum temperature difference, 3600 is a time conversion factor for converting the instantaneous liquid flow rate into a liquid flow rate per hour. The soil information includes soil temperature, moisture content, conductivity, pH value and salinity.

7. The intelligent rapid pre-detection diagnosis method for tree health according to claim 1, wherein in S4, the daily maximum sap flow deviation is graded, specifically: Wherein, D is the deviation.

8. The intelligent rapid pre-detection diagnosis method for tree health according to claim 1, wherein ​ ​ ​ ​ ； ​ ​ The S3 adopts a liquid flow measuring device to collect the phenotype information of the target tree to be detected, and adopts a meteorological and soil measuring device to collect the meteorological and soil information of the target tree to be detected, the liquid flow measuring device is installed at 1.3 m north of the target tree, and the meteorological and soil measuring device is installed within 5 m of the target tree. And through the HIL test, the installation positions of the liquid flow measuring device and the meteorological and soil measuring device are verified to determine the optimal installation positions of each.

9. The intelligent rapid pre-diagnosis method for tree health according to claim 1, characterized in that, The machine learning algorithm in the S2 includes but is not limited to random forest, support vector machine or neural network algorithm.

10. A tree health intelligent rapid pre-diagnosis system, characterized in that, Including: A data acquisition module is configured to collect liquid flow data of standard trees and meteorological and soil information of growth environments in a preset time period, and to construct a liquid flow database of standard trees of a target tree species; A data processing module is configured to establish a liquid flow calculation model of the target tree species based on a machine learning algorithm, to train the model using the liquid flow database, and to save an optimal liquid flow calculation model; wherein the liquid flow calculation model takes meteorological and soil information as independent variables and liquid flow size as dependent variables; A to-be-detected data module is configured to collect phenotype information, liquid flow data and meteorological and soil information of a growth environment of the target tree to be detected within a preset detection time; the phenotype information includes crown deviation degree, trunk damage degree, branch damage degree, main body inclination degree and pest threat situation; A deviation module is configured to input the liquid flow data of the target tree to be detected and the meteorological and soil information of the growth environment into the optimal liquid flow calculation model, to calculate a theoretical liquid flow value of the target tree, to compare the theoretical liquid flow value with an actual monitored liquid flow value, to calculate a liquid flow deviation, and to classify according to the daily maximum liquid flow deviation; A diagnosis module is configured to combine the liquid flow deviation and the phenotype information, to comprehensively evaluate a tree health grade, and to obtain a diagnosis result.

Images