Citrus fruit cracking prevention and control method based on intelligent monitoring and regulation

Abstract

The invention relates to the technical field of citrus fruit cracking prevention and control, and provides a citrus fruit cracking prevention and control method based on intelligent monitoring and regulation, and the method comprises the following steps: collecting soil humidity data, fruit growth data and meteorological condition data of monitored citrus; identifying soil humidity data through a soil humidity analysis model to obtain a soil humidity evaluation state; identifying fruit growth data through a fruit growth analysis model to obtain a fruit growth evaluation result; identifying meteorological condition data through a meteorological condition risk assessment model to obtain a meteorological condition influence degree; identifying a soil humidity evaluation state, a fruit growth evaluation result and a meteorological condition influence degree through a comprehensive risk judgment model to obtain a citrus fruit cracking risk degree of the monitored citrus; an irrigation system and a fertilization device of the monitored citrus are controlled based on the citrus fruit cracking risk degree, and the irrigation amount, the fertilization amount and ventilation equipment are adjusted; when it is monitored that the citrus fruit cracking risk degree exceeds a preset threshold value, early warning is sent to planters to take prevention and control measures. According to the method, the accuracy of citrus fruit cracking risk assessment is realized, so that the management efficiency of citrus planting and the yield and quality of fruits are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of citrus fruit cracking prevention and control, and in particular to a citrus fruit cracking prevention and control method based on intelligent monitoring and regulation. Background Art

[0002] Citrus fruit cracking is a common problem in citrus cultivation, seriously affecting the yield and quality of citrus. Currently, the prevention and control of citrus fruit cracking mainly relies on traditional planting and management experience, lacking scientific and effective precise prevention and control methods.

[0003] Existing technologies often lack accurate and timely monitoring and regulation of the citrus growing environment, failing to meet the actual needs of citrus growth. Traditional irrigation and fertilization methods are typically based on experience and fixed schedules, failing to adjust to the real-time growth conditions and environmental changes of the citrus. This results in poor real-time risk assessment of citrus fruit cracking, leading to unstable soil moisture and nutrient supply, increasing the risk of fruit cracking. Summary of the Invention

[0004] In view of this, the present invention proposes a method for preventing and controlling citrus fruit cracking based on intelligent monitoring and regulation, which solves the problem that the existing technology has poor real-time risk assessment effect on citrus fruit cracking.

[0005] The technical solution of the present invention is implemented as follows: The present invention provides a method for preventing and controlling citrus fruit cracking based on intelligent monitoring and regulation, comprising the following steps:

[0006] S1, collecting raw soil moisture data, raw fruit growth data, and raw meteorological condition data of monitored citrus fruits through a sensor network, and preprocessing the raw soil moisture data, raw fruit growth data, and raw meteorological condition data to obtain soil moisture data, fruit growth data, and meteorological condition data;

[0007] S2, constructing a soil moisture analysis model based on a logistic regression model, identifying the soil moisture data through the soil moisture analysis model, and obtaining a soil moisture assessment state;

[0008] S3, constructing a fruit growth analysis model based on a deep neural network, identifying the fruit growth data through the fruit growth analysis model, and obtaining a fruit growth evaluation result;

[0009] S4, constructing a meteorological condition risk assessment model using a decision tree model, identifying the meteorological condition data through the meteorological condition risk assessment model, and obtaining the degree of meteorological condition impact;

[0010] S5, constructing a comprehensive risk judgment model using a random forest model, identifying the soil moisture assessment status, the fruit growth assessment result, and the degree of influence of the meteorological conditions through the comprehensive risk judgment model, and obtaining the risk level of citrus fruit cracking of the monitored citrus;

[0011] S6, controlling the irrigation system and fertilization device of the monitored citrus fruits based on the risk level of citrus fruit cracking, and adjusting the irrigation amount, fertilization amount, and ventilation equipment;

[0012] S7: When it is detected that the risk level of citrus fruit cracking exceeds a preset threshold, an early warning is sent to the growers so that they can take prevention and control measures.

[0013] Based on the above technical solution, preferably, step S1 includes:

[0014] Raw soil moisture data is collected through a soil moisture sensor, and the raw soil moisture data is filtered and normalized to obtain soil moisture data. Raw fruit growth data is collected through a fruit growth monitor, and the raw fruit growth data is smoothed and normalized to obtain fruit growth data. Raw meteorological condition data is collected through a meteorological monitoring device, and missing value filling and outlier detection are performed on the raw meteorological condition data to obtain meteorological condition data.

[0015] Based on the above technical solution, preferably, step S2 includes:

[0016] Acquire historical soil moisture data and citrus growth requirement data, and construct a soil moisture analysis model based on a logistic regression model, the historical soil moisture data, and the citrus growth requirement data;

[0017] Soil moisture content and soil temperature were selected as characteristic variables and regression coefficients were determined. The soil moisture analysis model was trained using historical soil moisture data and citrus growth requirement data, and the regression coefficients were adjusted to minimize the prediction error.

[0018] The trained soil moisture analysis model was validated using the cross-validation method to evaluate the performance of the model on different data sets and obtain the soil moisture analysis model;

[0019] The soil moisture data is input into the soil moisture analysis model to obtain a soil moisture assessment state.

[0020] Based on the above technical solution, preferably, step S3 includes:

[0021] An initial fruit growth analysis model is constructed based on a deep neural network, and historical fruit growth data and citrus growth patterns are obtained. The initial fruit growth analysis model is further trained using the historical fruit growth data and the citrus growth patterns to obtain a fruit growth analysis model.

[0022] On the basis of the above technical solution, preferably, the calculation formula of the fruit growth analysis model is:

[0023] G=σ(W3·σ(W2·σ(W1·X+b1)+b2)+b3);

[0024]

[0025] Among them, G is the fruit growth evaluation result, X is the fruit growth data, are the fruit size, fruit weight, and fruit growth rate, respectively; W1, W2, and W3 are the weight matrices of fruit size, fruit weight, and fruit growth rate, respectively; σ(·) is the activation function; b1, b2, and b3 are the bias constants of fruit size, fruit weight, and fruit growth rate, respectively;

[0026] The fruit growth data is input into the fruit growth analysis model to obtain a fruit growth evaluation result.

[0027] Based on the above technical solution, preferably, step S4 includes:

[0028] Obtaining citrus meteorological factors affecting citrus, extracting meteorological characteristic variables of the citrus meteorological factors, constructing a decision tree model based on the meteorological characteristic variables, obtaining historical meteorological data, iteratively training the decision tree model using the historical meteorological data until an iteration stopping condition is met, thereby obtaining a trained decision tree model, and validating and evaluating the trained decision tree model through cross-validation to obtain a meteorological condition risk assessment model;

[0029] The meteorological condition data is identified through the meteorological condition risk assessment model to obtain the degree of meteorological condition impact.

[0030] Based on the above technical solution, preferably, step S5 includes:

[0031] A random forest model was constructed with soil moisture, fruit growth, and meteorological conditions as characteristic variables. The random forest model was trained based on multi-source historical data, which included historical soil moisture feature vectors, historical fruit growth feature vectors, and historical meteorological condition feature vectors. The trained random forest model was verified and evaluated through cross-validation to obtain a comprehensive risk judgment model.

[0032] On the basis of the above technical solution, preferably, the calculation formula of the comprehensive risk judgment model is:

[0033]

[0034] Among them, Q is the risk level of citrus fruit cracking, U is the number of decision trees in the random forest model, and θ u is the fitting parameter of the u-th decision tree, P(y=k|x,θ u ) is the predicted probability of the u-th decision tree for sample x belonging to the k-th risk level;

[0035] The soil moisture assessment status, the fruit growth assessment results and the degree of influence of the meteorological conditions are identified through a comprehensive risk judgment model to obtain the risk level of citrus fruit cracking of the monitored citrus and output the corresponding risk level.

[0036] Based on the above technical solution, preferably, step S6 includes:

[0037] The irrigation volume of the irrigation system is set according to the risk level of citrus fruit cracking. When the risk level of citrus fruit cracking is higher than the preset high-risk threshold, the irrigation volume is increased; when the risk level of citrus fruit cracking is lower than the preset low-risk threshold, the irrigation volume is reduced.

[0038] Adjust the amount of fertilizer applied by the fertilizing device. Based on the risk assessment results, when the risk level of citrus fruit cracking is higher than the preset high-risk threshold, increase the amount of fertilizer applied; when the risk level of citrus fruit cracking is lower than the preset low-risk threshold, reduce the amount of fertilizer applied.

[0039] Control the opening and closing of ventilation equipment. When the risk level of citrus fruit cracking is higher than the preset high-risk threshold, turn on the ventilation equipment to reduce the humidity inside the fruit; in low-risk situations, turn off the ventilation equipment to maintain the growth environment.

[0040] Based on the above technical solution, preferably, step S7 includes:

[0041] When it is monitored that the risk level of citrus fruit cracking exceeds the preset threshold, an early warning message will be sent to the growers via SMS, email or APP push, reminding them to take corresponding prevention and control measures, such as increasing irrigation frequency, adjusting fertilization plans, and turning on ventilation equipment.

[0042] The method for preventing and controlling citrus fruit cracking based on intelligent monitoring and regulation of the present invention has the following beneficial effects compared with the prior art:

[0043] (1) By collecting soil moisture, fruit growth and meteorological condition data in real time, the soil moisture analysis model, fruit growth analysis model and meteorological condition risk assessment model are used to identify them respectively, and then the risk assessment is performed through the comprehensive risk judgment model, which improves the accuracy of citrus fruit cracking risk assessment. According to the assessment results, the irrigation, fertilization and ventilation equipment are automatically adjusted to optimize the growth environment. At the same time, when the risk exceeds the threshold, early warning information is sent to the planting personnel in time, thereby improving the management efficiency of citrus planting and the yield and quality of the fruit;

[0044] (2) By integrating the prediction results of the random forest model and multiple decision trees, combined with dynamic weight adjustment and multi-class risk prediction, a comprehensive and accurate assessment of the risk of citrus fruit cracking was achieved. This not only improved the accuracy of risk assessment, but also provided growers with more detailed risk analysis and timely decision support, thereby effectively optimizing citrus planting management strategies and reducing the risk of fruit cracking.

[0045] (3) By building a fruit growth analysis model based on a deep neural network and combining it with historical fruit growth data and training on citrus growth patterns, we can accurately assess the growth status of the fruit, improve the accuracy of fruit growth assessment, and identify growth anomalies in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0047] Figure 1 This is a flow chart of a method for preventing and controlling citrus fruit cracking based on intelligent monitoring and regulation according to the present invention. DETAILED DESCRIPTION

[0048] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0049] See also Figure 1 The present invention provides a method for preventing and controlling citrus fruit cracking based on intelligent monitoring and regulation, comprising the following steps:

[0050] S1, collecting raw soil moisture data, raw fruit growth data, and raw meteorological condition data of monitored citrus fruits through a sensor network, and preprocessing the raw soil moisture data, raw fruit growth data, and raw meteorological condition data to obtain soil moisture data, fruit growth data, and meteorological condition data;

[0051] S2, constructing a soil moisture analysis model based on a logistic regression model, identifying the soil moisture data through the soil moisture analysis model, and obtaining a soil moisture assessment state;

[0052] S3, constructing a fruit growth analysis model based on a deep neural network, identifying the fruit growth data through the fruit growth analysis model, and obtaining a fruit growth evaluation result;

[0053] S4, constructing a meteorological condition risk assessment model using a decision tree model, identifying the meteorological condition data through the meteorological condition risk assessment model, and obtaining the degree of meteorological condition impact;

[0054] S5, constructing a comprehensive risk judgment model using a random forest model, identifying the soil moisture assessment status, the fruit growth assessment result, and the degree of influence of the meteorological conditions through the comprehensive risk judgment model, and obtaining the risk level of citrus fruit cracking of the monitored citrus;

[0055] S6, controlling the irrigation system and fertilization device of the monitored citrus fruits based on the risk level of citrus fruit cracking, and adjusting the irrigation amount, fertilization amount, and ventilation equipment;

[0056] S7: When it is detected that the risk level of citrus fruit cracking exceeds a preset threshold, an early warning is sent to the growers so that they can take prevention and control measures.

[0057] Specifically, this embodiment collects soil moisture, fruit growth and meteorological condition data in real time through intelligent technology, and uses a soil moisture analysis model, a fruit growth analysis model and a meteorological condition risk assessment model to identify them respectively. Then, a risk assessment is performed through a comprehensive risk judgment model, thereby improving the accuracy of citrus fruit cracking risk assessment, and automatically adjusting irrigation, fertilization and ventilation equipment according to the assessment results to optimize the growth environment. At the same time, when the risk exceeds the threshold, early warning information is sent to the growers in a timely manner, thereby improving the management efficiency of citrus planting and the yield and quality of the fruit.

[0058] Step S1 includes:

[0059] Raw soil moisture data is collected through a soil moisture sensor, and the raw soil moisture data is filtered and normalized to obtain soil moisture data. Raw fruit growth data is collected through a fruit growth monitor, and the raw fruit growth data is smoothed and normalized to obtain fruit growth data. Raw meteorological condition data is collected through a meteorological monitoring device, and missing value filling and outlier detection are performed on the raw meteorological condition data to obtain meteorological condition data.

[0060] In a specific embodiment, raw soil moisture data is regularly collected by a soil moisture sensor, a filtering algorithm is used to remove noise in the data, and then normalization processing is performed to standardize the data to a range of 0 to 1.

[0061] The original fruit growth data were collected regularly through the fruit growth monitor, and the smoothing method was used to reduce the data fluctuations and perform standardization to make the data at different time points comparable.

[0062] The original meteorological condition data is collected through meteorological monitoring equipment, and missing value filling (such as using the mean filling method) and outlier detection (such as using the Z-score method) are performed to ensure the integrity and accuracy of the meteorological data, and ultimately obtain reliable meteorological condition data.

[0063] Specifically, this embodiment improves the quality of soil moisture, fruit growth and meteorological condition data by filtering, normalizing, smoothing and standardizing the original data, integrating multi-source data on soil moisture, fruit growth and meteorological conditions to form a comprehensive monitoring system that can better reflect the actual growth conditions. By regularly collecting and processing data in real time, the ability to respond to changes in the citrus growth environment is enhanced, allowing growers to adjust management measures in a timely manner and reduce the risk of fruit cracking.

[0064] Step S2 includes:

[0065] Acquire historical soil moisture data and citrus growth requirement data, and construct a soil moisture analysis model based on a logistic regression model, the historical soil moisture data, and the citrus growth requirement data;

[0066] Soil moisture content and soil temperature were selected as characteristic variables and regression coefficients were determined. The soil moisture analysis model was trained using historical soil moisture data and citrus growth requirement data, and the regression coefficients were adjusted to minimize the prediction error.

[0067] The trained soil moisture analysis model was validated using the cross-validation method to evaluate the performance of the model on different data sets and obtain the soil moisture analysis model;

[0068] The calculation formula of the soil moisture analysis model is:

[0069]

[0070] Where H is the soil moisture assessment state, W is the soil moisture content, T is the soil temperature, a, b, and c are the regression coefficient of soil moisture content, the regression coefficient of soil temperature, and the bias constant, respectively;

[0071] The soil moisture data is input into the soil moisture analysis model to obtain a soil moisture assessment state.

[0072] In a specific embodiment, first, historical soil moisture data and citrus growth requirement data are collected. These data can be obtained through long-term monitoring and recording, covering soil moisture and citrus growth requirements at different growth stages.

[0073] Based on the collected historical soil moisture data and citrus growth requirement data, a soil moisture analysis model was constructed using a logistic regression model. Soil moisture content and soil temperature were selected as characteristic variables, and the corresponding regression coefficients were determined.

[0074] The soil moisture analysis model was trained using historical soil moisture data and citrus growth requirement data, and the regression coefficients were adjusted to minimize prediction error. The trained model was validated using cross-validation to evaluate its performance on different datasets.

[0075] The real-time collected soil moisture data is input into the trained soil moisture analysis model to obtain the soil moisture assessment state H, which reflects the impact of current soil moisture on citrus growth.

[0076] Specifically, this embodiment uses historical data for training and verification, and the logistic regression model can more accurately reflect the relationship between soil moisture and citrus growth requirements. Soil moisture content and soil temperature are selected as characteristic variables, so that the model can more effectively capture the key factors affecting citrus growth. By inputting real-time soil moisture data into the trained model, the soil moisture assessment status can be quickly obtained.

[0077] The soil moisture analysis model is built based on a logistic regression model, which is a machine learning algorithm commonly used for binary classification problems. In this embodiment, the soil moisture assessment state H is regarded as the output of the binary classification (high / low moisture).

[0078] Soil moisture content W and soil temperature T are selected as characteristic variables. These two factors are key environmental parameters affecting soil moisture and can better reflect the soil moisture status.

[0079] By training a logistic regression model using historical soil moisture data and citrus growth requirement data, we determined the three regression coefficients a, b, and c. These coefficients reflect the degree to which each characteristic variable affects the soil moisture assessment status.

[0080] The cross-validation method is used to verify and evaluate the trained soil moisture analysis model. This ensures that the model performs stably and reliably on different datasets.

[0081] Step S3 includes:

[0082] An initial fruit growth analysis model is constructed based on a deep neural network, and historical fruit growth data and citrus growth patterns are obtained. The initial fruit growth analysis model is further trained using the historical fruit growth data and the citrus growth patterns to obtain a fruit growth analysis model.

[0083] The calculation formula of the fruit growth analysis model is:

[0084] G=σ(W3·σ(W2·σ(W1·X+b1)+b2)+b3);

[0085]

[0086] Among them, G is the fruit growth evaluation result, X is the fruit growth data, are fruit size, fruit weight, and fruit growth rate, respectively; W1, W2, and W3 are the weight matrices of fruit size, fruit weight, and fruit growth rate, respectively; σ(·) is the activation function; b1, b2, and b3 are the bias constants of fruit size, fruit weight, and fruit growth rate, respectively.

[0087] The fruit growth data is input into the fruit growth analysis model to obtain a fruit growth evaluation result.

[0088] In one embodiment, historical fruit growth data and citrus growth patterns are collected, including information such as fruit size, weight, and growth rate. This data can be obtained through long-term monitoring and recording, covering the growth of fruits at different growth stages.

[0089] An initial fruit growth analysis model was constructed based on a deep neural network. Fruit size, fruit weight, and fruit growth rate were selected as input features, and the corresponding weight matrix and bias constant were set.

[0090] The initial fruit growth analysis model was trained using historical fruit growth data and citrus growth patterns. The weight matrix and bias constant were adjusted to minimize prediction error. The trained model was validated using cross-validation to evaluate its performance on different datasets.

[0091] The real-time collected fruit growth data is input into the trained fruit growth analysis model to obtain the fruit growth evaluation result G, which reflects the current fruit growth status.

[0092] Specifically, in this embodiment, by using a deep neural network, the model can capture the complex nonlinear relationship in the fruit growth process, and selects fruit size, weight and growth rate as input features, so that the model can fully reflect the growth status of the fruit.

[0093] The fruit growth analysis model is built based on a deep neural network, which can automatically learn the complex nonlinear relationships between features, thereby better capturing the laws of fruit growth.

[0094] The model simultaneously considers three characteristic variables: fruit size, fruit weight and fruit growth rate. These characteristics can comprehensively reflect the growth status of the fruit.

[0095] By training and adjusting the weight matrices W1, W2, W3 and the bias constants b1, b2, b3, the model can automatically learn the relative importance of different features to fruit growth, thereby making more accurate predictions.

[0096] Introducing an activation function σ(·) into the model output, such as the sigmoid function or the ReLU function, can introduce nonlinear transformations and enhance the expressive power of the model.

[0097] Step S4 includes:

[0098] Obtaining citrus meteorological factors affecting citrus, extracting meteorological characteristic variables of the citrus meteorological factors, constructing a decision tree model based on the meteorological characteristic variables, obtaining historical meteorological data, iteratively training the decision tree model using the historical meteorological data until an iteration stopping condition is met, thereby obtaining a trained decision tree model, and validating and evaluating the trained decision tree model through cross-validation to obtain a meteorological condition risk assessment model;

[0099] The calculation formula of the meteorological condition risk assessment model is:

[0100]

[0101] Among them, R is the degree of influence of meteorological conditions, w i is the weight coefficient of the i-th node in the decision tree, X i is the characteristic value of the i-th citrus meteorological factor, n is the number of nodes in the decision tree, and exp(·) is the exponential function;

[0102] The meteorological condition data is identified through the meteorological condition risk assessment model to obtain the degree of meteorological condition impact.

[0103] In a specific embodiment, n is set to 5, and meteorological factors affecting citrus growth are collected, including temperature, humidity, precipitation, wind speed, and light. These data can be collected regularly by meteorological monitoring equipment to ensure the timeliness and accuracy of the data.

[0104] Meteorological characteristic variables, such as soil temperature, air humidity, and precipitation, are extracted from the collected meteorological factors and used as input features of the decision tree model.

[0105] A decision tree model is constructed based on the extracted meteorological characteristic variables, and the model is trained using historical meteorological data to determine the weight coefficient and characteristic value of each node until the preset iteration stopping condition is reached.

[0106] The trained decision tree model is verified and evaluated through cross-validation method to ensure the accuracy and reliability of the model, and finally a meteorological condition risk assessment model is obtained.

[0107] The trained meteorological condition risk assessment model is used to identify meteorological condition data and obtain the meteorological condition impact degree R, which reflects the impact of meteorological factors on citrus growth.

[0108] Specifically, this embodiment forms a comprehensive meteorological monitoring system by collecting multiple meteorological factors, which can timely identify potential meteorological risks, and use a decision tree model to analyze meteorological conditions, which can clearly show the impact of different meteorological factors on citrus growth. By analyzing real-time meteorological data, it can quickly assess the impact of meteorological conditions on citrus growth, provide timely risk warnings to growers, and promote rapid response.

[0109] The meteorological risk assessment model is based on a decision tree model. By introducing the exponential function exp, it can better reflect the nonlinear impact of meteorological factors on citrus fruit cracking risk. The exponential function amplifies the importance of meteorological factors and improves the model's ability to identify extreme meteorological conditions. The decision tree model automatically learns the complex relationship between meteorological factors and citrus fruit cracking risk, eliminating the need for manual rule-setting and improving the model's adaptability and generalization capabilities.

[0110] Step S5 includes:

[0111] A random forest model was constructed with soil moisture, fruit growth, and meteorological conditions as characteristic variables. The random forest model was trained based on multi-source historical data, which included historical soil moisture feature vectors, historical fruit growth feature vectors, and historical meteorological condition feature vectors. The trained random forest model was verified and evaluated through cross-validation to obtain a comprehensive risk judgment model.

[0112] The calculation formula of the comprehensive risk judgment model is:

[0113]

[0114] Among them, Q is the risk level of citrus fruit cracking, U is the number of decision trees in the random forest model, and θ u is the fitting parameter of the u-th decision tree, P(y=k|x,θ u ) is the predicted probability of the u-th decision tree for sample x belonging to the k-th risk level.

[0115] The soil moisture assessment status, the fruit growth assessment results and the degree of influence of the meteorological conditions are identified through a comprehensive risk judgment model to obtain the risk level of citrus fruit cracking of the monitored citrus and output the corresponding risk level.

[0116] In one embodiment, soil moisture, fruit growth assessment results, and the impact of meteorological conditions are selected as characteristic variables to construct a random forest model. These characteristic variables can fully reflect the environment and status of citrus growth.

[0117] A random forest model was constructed based on multi-source historical data, including historical soil moisture feature vectors, historical fruit growth feature vectors, and historical meteorological condition feature vectors. Each decision tree in the random forest model was trained using a different subset of features and samples to enhance the model's generalization capabilities.

[0118] The random forest model is trained using historical data and model parameters are adjusted to improve prediction accuracy. The trained random forest model is verified and evaluated using cross-validation to ensure model stability and reliability.

[0119] The trained random forest model is used to identify the soil moisture assessment status, fruit growth assessment results, and the impact of meteorological conditions to obtain the citrus fruit cracking risk level Q and output the corresponding risk level.

[0120] The risk level Q is usually a continuous value that represents the risk level of citrus fruit cracking. This value can be a number between 0 and 1, where 0 represents no risk and 1 represents extremely high risk.

[0121] Based on the risk level Q, multiple risk levels can be set to facilitate implant personnel to understand and take corresponding measures. Common risk level classifications may include:

[0122] Low risk (Q<0.3): indicates that the risk of fruit cracking is low and the current management measures basically meet the growth needs of citrus.

[0123] Medium risk (0.3≤Q<0.7): indicates a medium risk of fruit cracking. Planters are advised to pay attention to soil moisture and meteorological changes and adjust management measures appropriately.

[0124] High risk (Q≥0.7): indicates a high risk of fruit cracking and requires immediate prevention and control measures, such as increasing irrigation, adjusting fertilization, and turning on ventilation equipment.

[0125] Specifically, in this embodiment, by integrating multi-source data on soil moisture, fruit growth, and meteorological conditions, the random forest model can comprehensively assess the risk of citrus fruit cracking. By constructing multiple decision trees and voting, the random forest model can effectively reduce overfitting. By inputting soil moisture, fruit growth, and meteorological condition data in real time, it can quickly assess the risk level of citrus fruit cracking and provide timely risk warnings to growers.

[0126] The comprehensive risk assessment model is built on a random forest model. It integrates the prediction results of multiple decision trees, leveraging the independence and diversity of each tree to enhance the model's stability and accuracy. This integration method can effectively reduce the bias and variance of a single model.

[0127] Assign different fitting parameters θ to each decision tree u , the weights are dynamically adjusted according to their performance on the training set, so that the decision trees with better performance contribute more to the final risk assessment, thereby improving the predictive ability of the overall model.

[0128] By combining the predicted probabilities of sample x belonging to different risk categories k, we can more comprehensively assess the risk level of citrus fruit cracking and provide a more detailed risk analysis.

[0129] Step S6 includes:

[0130] The irrigation volume of the irrigation system is set according to the risk level of citrus fruit cracking. When the risk level of citrus fruit cracking is higher than the preset high-risk threshold, the irrigation volume is increased; when the risk level of citrus fruit cracking is lower than the preset low-risk threshold, the irrigation volume is reduced.

[0131] Adjust the amount of fertilizer applied by the fertilizing device. Based on the risk assessment results, when the risk level of citrus fruit cracking is higher than the preset high-risk threshold, increase the amount of fertilizer applied; when the risk level of citrus fruit cracking is lower than the preset low-risk threshold, reduce the amount of fertilizer applied.

[0132] Control the opening and closing of ventilation equipment. When the risk level of citrus fruit cracking is higher than the preset high-risk threshold, turn on the ventilation equipment to reduce the humidity inside the fruit; in low-risk situations, turn off the ventilation equipment to maintain the growth environment.

[0133] In one embodiment, the irrigation system's irrigation rate is set based on the risk assessment results of citrus fruit cracking. When the monitored risk level is higher than a preset high-risk threshold, the irrigation rate is increased to ensure that the soil maintains appropriate moisture. When the risk level is lower than a preset low-risk threshold, the irrigation rate is reduced to prevent fruit cracking caused by excessive moisture.

[0134] Based on the risk assessment results, the fertilizer application rate of the fertilization device is adjusted. When the risk level of citrus fruit cracking is higher than the preset high-risk threshold, the fertilizer application rate is increased to promote healthy fruit growth; when the risk level is lower than the preset low-risk threshold, the fertilizer application rate is reduced to avoid nutrient overload.

[0135] The ventilation system is controlled to open and close according to the risk level of citrus fruit cracking. When the risk level is higher than the preset high-risk threshold, the ventilation system is turned on to reduce the internal humidity of the fruit and reduce the risk of fruit cracking. In low-risk situations, the ventilation system is turned off to maintain a suitable growing environment.

[0136] Through the monitoring system, the effects of irrigation, fertilization and ventilation can be tracked in real time, and management strategies can be adjusted in a timely manner to ensure the optimal growing environment for citrus.

[0137] Specifically, this embodiment achieves precise management of the citrus growth environment by dynamically adjusting irrigation, fertilization and ventilation measures according to the risk level of citrus fruit cracking, reduces the risk of fruit cracking, and improves the growth quality of the fruit. Adjusting management measures based on real-time risk assessment results can effectively avoid waste of resources, such as reducing unnecessary use of water and fertilizer. Through real-time monitoring and feedback mechanisms, growers can quickly respond to environmental changes.

[0138] Step S7 includes:

[0139] When it is monitored that the risk level of citrus fruit cracking exceeds the preset threshold, an early warning message will be sent to the growers via SMS, email or APP push, reminding them to take corresponding prevention and control measures, such as increasing irrigation frequency, adjusting fertilization plans, and turning on ventilation equipment.

[0140] In one embodiment, based on historical data and expert experience, preset thresholds for citrus fruit cracking risk are set, including high and low risk thresholds. These thresholds serve as a benchmark for risk assessment so that appropriate prevention and control measures can be taken in a timely manner.

[0141] When the monitored risk of citrus fruit cracking exceeds a preset high-risk threshold, an early warning mechanism will be automatically triggered. The warning information can be sent to growers via text message, email, or mobile application (mobile phone APP).

[0142] Early warning information should include current risk assessment results, recommended prevention and control measures (such as increasing irrigation frequency, adjusting fertilization plans, turning on ventilation equipment, etc.), as well as relevant weather forecasts and environmental change information to help growers make timely decisions.

[0143] After receiving the early warning information, the growers can quickly take corresponding prevention and control measures to ensure that the growing environment of citrus is adjusted in time to reduce the risk of fruit cracking.

[0144] Specifically, this embodiment sets risk thresholds and establishes a monitoring mechanism to promptly detect potential fruit cracking risks, ensuring that growers can take preventive measures before problems occur and reduce losses. It utilizes modern communication technologies (such as text messages, emails, and APP push) to achieve real-time information transmission to growers, allowing growers to quickly obtain important information. The recommended prevention and control measures contained in the early warning information provide growers with scientific decision-making support, helping them make reasonable management adjustments when facing risks. By establishing a comprehensive monitoring and early warning system, growers can more effectively manage risks, optimize resource allocation, and improve overall production efficiency and economic benefits.

[0145] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for preventing and controlling citrus fruit cracking based on intelligent monitoring and regulation, characterized in that: The following steps are involved: S1, collecting original soil moisture data, original fruit growth data and original meteorological condition data of the monitored citrus through a sensor network, and preprocessing the original soil moisture data, original fruit growth data and original meteorological condition data to obtain soil moisture data, fruit growth data and meteorological condition data; S2, constructing a soil moisture analysis model based on a logistic regression model, identifying the soil moisture data through the soil moisture analysis model, and obtaining a soil moisture assessment state; S3, constructing a fruit growth analysis model according to a deep neural network, identifying the fruit growth data through the fruit growth analysis model, and obtaining a fruit growth evaluation result; S4, constructing a meteorological condition risk assessment model using a decision tree model, identifying the meteorological condition data through the meteorological condition risk assessment model, and obtaining the degree of impact of the meteorological condition; S5, constructing a comprehensive risk judgment model through a random forest model, identifying the soil moisture assessment state, the fruit growth assessment result and the impact degree of the meteorological conditions through the comprehensive risk judgment model, and obtaining the risk degree of citrus fruit cracking of the monitored citrus; S6, based on the risk level of citrus fruit cracking, controlling the irrigation system and fertilization device of the monitored citrus, adjusting the irrigation amount, fertilization amount and ventilation equipment; S7: When it is detected that the risk of citrus fruit cracking exceeds a preset threshold, an early warning is sent to the growers to take preventive measures.

2. A method for preventing and controlling citrus fruit cracking based on intelligent monitoring and regulation as claimed in claim 1, characterized in that: Step S1 includes: The original soil moisture data is collected by a soil moisture sensor, and the original soil moisture data is filtered and normalized to obtain soil moisture data. The original fruit growth data is collected by a fruit growth monitor, and the original fruit growth data is smoothed and standardized to obtain fruit growth data. The original meteorological condition data is collected by a meteorological monitoring device, and the original meteorological condition data is filled with missing values ​​and detected for outliers to obtain meteorological condition data.

3. A method for preventing and controlling citrus fruit cracking based on intelligent monitoring and regulation as claimed in claim 1, characterized in that: Step S2 includes: Acquire historical soil moisture data and citrus growth demand data, and construct a soil moisture analysis model based on a logistic regression model, the historical soil moisture data and the citrus growth demand data; Soil moisture content and soil temperature were selected as characteristic variables and regression coefficients were determined. The soil moisture analysis model was trained using historical soil moisture data and citrus growth demand data, and the regression coefficients were adjusted to minimize the prediction error. The trained soil moisture analysis model was validated using the cross-validation method to evaluate the performance of the model on different data sets and obtain the soil moisture analysis model; The soil moisture data is input into the soil moisture analysis model to obtain a soil moisture assessment state.

4. A method for preventing and controlling citrus fruit cracking based on intelligent monitoring and regulation as claimed in claim 3, characterized in that: Step S3 includes: An initial fruit growth analysis model is constructed based on a deep neural network, historical fruit growth data and citrus growth patterns are obtained, and the initial fruit growth analysis model is further trained using the historical fruit growth data and the citrus growth patterns to obtain a fruit growth analysis model.

5. A method for preventing and controlling citrus fruit cracking based on intelligent monitoring and regulation as claimed in claim 4, characterized in that: The calculation formula of the fruit growth analysis model is: G=σ(W3·σ(W2·σ(W1·X+b1)+b2)+b3); Among them, G is the fruit growth evaluation result, X is the fruit growth data, are fruit size, fruit weight, and fruit growth rate, respectively; W1, W2, and W3 are the weight matrices of fruit size, fruit weight, and fruit growth rate, respectively; σ(·) is the activation function; b1, b2, and b3 are the bias constants of fruit size, fruit weight, and fruit growth rate, respectively; The fruit growth data is input into the fruit growth analysis model to obtain a fruit growth evaluation result.

6. A method for preventing and controlling citrus fruit cracking based on intelligent monitoring and regulation as claimed in claim 5, characterized in that: Step S4 includes: Obtain citrus meteorological factors that affect citrus, extract meteorological characteristic variables of the citrus meteorological factors, build a decision tree model based on the meteorological characteristic variables, obtain historical meteorological data, iteratively train the decision tree model through the historical meteorological data until an iteration stop condition is reached, obtain a trained decision tree model, verify and evaluate the trained decision tree model through cross-validation, and obtain a meteorological condition risk assessment model; The meteorological condition data is identified through the meteorological condition risk assessment model to obtain the degree of meteorological condition impact.

7. A method for preventing and controlling citrus fruit cracking based on intelligent monitoring and regulation as claimed in claim 6, characterized in that: Step S5 includes: A random forest model is constructed with soil moisture, fruit growth, and meteorological conditions as characteristic variables. The random forest model is trained based on multi-source historical data, which includes historical soil moisture feature vectors, historical fruit growth feature vectors, and historical meteorological condition feature vectors. The trained random forest model is verified and evaluated through cross-validation to obtain a comprehensive risk judgment model.

8. A method for preventing and controlling citrus fruit cracking based on intelligent monitoring and regulation as claimed in claim 7, characterized in that: The calculation formula of the comprehensive risk judgment model is: Among them, Q is the risk level of citrus fruit cracking, U is the number of decision trees in the random forest model, and θ u is the fitting parameter of the u-th decision tree, P(y=k|x,θ u ) is the predicted probability of the u-th decision tree for sample x belonging to the k-th risk level; The soil moisture assessment status, the fruit growth assessment results and the degree of influence of the meteorological conditions are identified through a comprehensive risk judgment model to obtain the risk level of citrus fruit cracking of the monitored citrus and output the corresponding risk level.

9. The method for preventing and controlling citrus fruit cracking based on intelligent monitoring and regulation according to claim 1, characterized in that: Step S6 includes: According to the risk level of citrus fruit cracking, the irrigation volume of the irrigation system is set. When the risk level of citrus fruit cracking is higher than the preset high risk threshold, the irrigation volume is increased; when the risk level of citrus fruit cracking is lower than the preset low risk threshold, the irrigation volume is reduced; Adjust the amount of fertilizer applied by the fertilization device. According to the risk assessment results, when the risk level of citrus fruit cracking is higher than the preset high-risk threshold, increase the amount of fertilizer applied; when the risk level of citrus fruit cracking is lower than the preset low-risk threshold, reduce the amount of fertilizer applied; Control the opening and closing of ventilation equipment. When the risk level of citrus fruit cracking is higher than the preset high-risk threshold, turn on the ventilation equipment to reduce the humidity inside the fruit; in low-risk situations, turn off the ventilation equipment to maintain the growth environment.

10. The method for preventing and controlling citrus fruit cracking based on intelligent monitoring and regulation according to claim 1, characterized in that: Step S7 includes: When it is monitored that the risk level of citrus fruit cracking exceeds the preset threshold, an early warning message will be sent to growers via SMS, email or APP push, reminding them to take corresponding prevention and control measures, such as increasing irrigation frequency, adjusting fertilization plans, and turning on ventilation equipment.

Images