Artificial cultivation method for conopomorpha sinensis based on growth environment regulation and control

Abstract

The invention relates to the technical field of insect breeding, in particular to a conopomorpha sinensis artificial breeding assisting method based on growth environment regulation, and the method comprises the steps: constructing a conopomorpha sinensis growth environment prediction model according to a training data packet in combination with a deep learning network; acquiring real-time growth environment characteristic parameters of the conopomorpha sinensis growth area at a preset time node, and importing the real-time growth environment characteristic parameters into the growth environment prediction model to obtain a prediction result; if the prediction result is a normal growth environment prediction result, the conopomorpha sinensis does not need manual interference under the current environment condition; if the prediction result is the early-warning growth environment abnormity prediction result, generating environment regulation early-warning information; and if the prediction result is the restoration growth environment abnormity prediction result, generating activity restoration early warning information. By means of the technology, the scientificity and effectiveness of artificial breeding assistance of conopomorpha sinensis can be improved, and normal growth and population stability of conopomorpha sinensis in the breeding assistance process are guaranteed.

Description

Technical Field

[0001] The invention relates to the technical field of insect breeding, in particular to an artificial breeding method of litchi peduncle borer based on growth environment regulation. Background Art

[0002] As the main pest of litchi, the outbreak of litchi stem borer and the degree of damage are closely related to the growth environment. During the flowering and fruiting period of litchi reproductive growth, the larvae of litchi stem borer bore into litchi fruits, causing fruit drop, which in turn becomes a place for the larvae to develop and reproduce, forming a vicious cycle and causing huge losses to the litchi industry. During the vegetative growth stage of litchi in autumn and winter, the larvae of litchi stem borer harm the tender shoots and leaves of litchi, reducing the nutritional accumulation of litchi and thus affecting the yield of the following year. At present, there are generally two ways to obtain litchi stem borer in scientific research. One is to use fresh litchi pits for artificial breeding indoors, but due to the extremely low amount and low efficiency, it is difficult to meet the needs of scientific research; the other is to capture adults directly from the wild and bring them back indoors for experiments. Because more litchi stem borers can only be obtained during the litchi fruiting period, the above two methods limit the scientific research on litchi stem borers throughout the year, resulting in a lag in the research and development of its biological characteristics, damage mechanisms and advanced prevention and control technologies. Therefore, chemical control is still the main measure at present, and the resulting environmental pollution and biodiversity loss problems are becoming increasingly serious, which is inconsistent with the national policy orientation. Therefore, there is an urgent need for an efficient and controllable artificial breeding method that can obtain litchi stem borers all year round to meet the needs of scientific research and provide scientific and technological support for the effective prevention and control of litchi stem borers. Therefore, in an artificial net room, the natural environment is simulated, and litchi tree varieties that are easy to bear fruit are cultivated. Appropriate artificial intervention methods are used to make litchi trees bear flowers and fruits during the reproductive growth period and lush tender shoots during the vegetative growth period. At the same time, changes in environmental temperature and humidity are detected to prevent the occurrence of diseases. Efforts are made to create environmental conditions that are conducive to the growth and development of litchi stem borers, so that the litchi stem borers can maintain a high population size all year round under the environmental conditions of artificial breeding to meet the needs of scientific research. With the rapid development of big data technology and deep learning algorithms, a basis has been provided for accurately regulating the growth and development of litchi trees and predicting the trend of disease epidemics, and a new way has been provided for artificially regulating the growth environment of litchi pedicle borers. In view of this, the present invention proposes an artificial breeding method for litchi pedicle borers based on growth environment regulation. Summary of the invention

[0003] The invention overcomes the shortcomings of the prior art and provides an artificial rearing method for litchi peduncle borer based on growth environment regulation.

[0004] To achieve the above-mentioned purpose, the technical solution adopted by the present invention is:

[0005] The invention discloses an artificial breeding method for litchi peduncle borer based on growth environment regulation, comprising the following steps:

[0006] Obtaining the foster log information of litchi stem borer through a big data network, extracting features and analyzing the foster log information, and obtaining a training data packet; wherein the training data includes early warning growth environment characteristic parameters and restorative growth environment characteristic parameters;

[0007] Constructing a growth environment prediction model for litchi pedicle borers according to the training data packet and in combination with a deep learning network; obtaining real-time growth environment characteristic parameters of the litchi pedicle borer growth area at a preset time node, importing the real-time growth environment characteristic parameters into the growth environment prediction model, and obtaining a prediction result;

[0008] If the prediction result is a normal growth environment prediction result, it means that under the current environmental conditions, no human intervention is required for litchi stem borer;

[0009] If the prediction result is a warning growth environment abnormality prediction result, it means that under the current environmental conditions, after the litchi pedicle borer lives for a preset period of time, the litchi pedicle borer will have an abnormal population growth activity event, indicating that the current growth environment of the litchi pedicle borer needs to be artificially adjusted, and then an environmental adjustment warning information is generated;

[0010] If the prediction result is an abnormal recovery growth environment prediction result, it means that the litchi stalk borer has experienced abnormal population growth activity, which means that the litchi stalk borer needs to be artificially restored in terms of growth activity, and activity recovery warning information will be generated.

[0011] Preferably, the foster log information of litchi stem borer is obtained through a big data network, and feature extraction and analysis processing are performed on the foster log information to obtain a training data packet, specifically:

[0012] Acquire the fostering log information of litchi stem borers through a big data network, and acquire abnormal growth environment events during the fostering process of litchi stem borers according to the fostering log information;

[0013] At the same time, the growth characteristic state corresponding to the litchi pedicle borer when each abnormal growth environment event occurs is obtained according to the fostering log information; wherein the growth characteristic state includes an abnormal growth activity state and a normal growth activity state;

[0014] Statistical analysis is performed on the growth characteristic states corresponding to the litchi pedicle borer when each abnormal growth environment event occurs, and the total number of abnormal growth activity states and the total number of normal growth activity states of the litchi pedicle borer when each abnormal growth environment event occurs are obtained;

[0015] The total number of abnormal growth activity states of litchi pedicle borers when each abnormal growth environment event occurs is processed by ratio to the total number of normal growth activity states, so as to obtain the abnormal growth activity ratio of the litchi pedicle borer population when each abnormal growth environment event occurs;

[0016] Compare the abnormal proportion of population growth activity with a preset proportion threshold; and extract only the abnormal growth environment events corresponding to the abnormal proportion of population growth activity greater than the preset proportion threshold;

[0017] Obtaining, from the fostering log information, growth environment characteristic parameters corresponding to the growth area of ​​the litchi pedicle borer within a preset time period before the occurrence of an abnormal growth environment event corresponding to the abnormal growth activity ratio of the population being greater than a preset ratio threshold, and defining them as early warning growth environment characteristic parameters;

[0018] And obtaining the growth environment characteristic parameters corresponding to the growth area of ​​the litchi pedicle borer when an abnormal growth environment event corresponding to the abnormal growth activity ratio of the population is greater than a preset ratio threshold in the fostering log information, and defining them as recovery growth environment characteristic parameters;

[0019] A training data packet is generated according to each warning growth environment characteristic parameter and the restorative growth environment characteristic parameter.

[0020] Preferably, a growth environment prediction model for litchi pedicle borer is constructed based on the training data packet and in combination with a deep learning network, specifically:

[0021] Construct a deep learning network and obtain the early warning growth environment characteristic parameters and the restorative growth environment characteristic parameters of the training data package;

[0022] Using the early warning growth environment characteristic parameter as an early warning condition node, using the restorative growth environment characteristic parameter as a restorative condition node, and initializing a weight matrix according to the early warning condition node and the restorative condition node;

[0023] The initialized weight matrix is ​​input into the deep learning network, and the input weight matrix is ​​imported into each layer of the network to calculate the output by using the forward propagation algorithm, and then the weight matrix is ​​adjusted according to the error between the output and the true value by the back propagation algorithm, and the iterative operation is continuously performed until the error is less than the preset threshold, and the iteratively optimized weight matrix is ​​obtained;

[0024] A prediction model is constructed, and the iteratively optimized weight matrix is ​​embedded in the prediction model to obtain a growth environment prediction model for litchi peduncle borer.

[0025] Preferably, the real-time growth environment characteristic parameters of the litchi stem borer growth area are obtained at a preset time node, and the real-time growth environment characteristic parameters are imported into the growth environment prediction model to obtain the prediction result, specifically:

[0026] Acquire the real-time growth environment characteristic parameters of the litchi pedicle borer growth area at a preset time node, and standardize the real-time growth environment characteristic parameters to match the data format and characteristic scale in the training data package when building the model;

[0027] Importing the standardized real-time growth environment characteristic parameters into the growth environment prediction model, and organizing the standardized real-time parameters according to the input layer requirements when constructing the model;

[0028] Using the forward propagation mechanism of the deep learning network, starting from the input layer, matching calculations are performed through the hidden layers in sequence to obtain the matching degree between the real-time growth environment characteristic parameters and the nodes in the corresponding layer; the matching degree between the real-time growth environment characteristic parameters and the nodes in the corresponding layer is compared with a preset matching degree threshold;

[0029] If the matching degree is greater than a preset matching degree threshold, then obtaining node information of a node corresponding to a node whose matching degree is greater than the preset matching degree threshold; wherein the node information includes a warning condition node and a recovery condition node;

[0030] If the node information of the node corresponding to the matching degree greater than the preset matching degree threshold is a warning condition node, a warning growth environment abnormality prediction result is generated; if the node information of the node corresponding to the node with a matching degree greater than the preset matching degree threshold is a recovery condition node, a recovery growth environment abnormality prediction result is generated;

[0031] If the matching degree is not greater than the preset matching degree threshold, the standardized real-time parameters are passed to the next layer until the output layer is reached. If there is still no node corresponding to a matching degree greater than the preset matching degree threshold after reaching the output layer, a normal prediction result of the growth environment is generated.

[0032] The artificial rearing method of litchi peduncle borer also includes the following steps:

[0033] Obtaining a preset cultivation scheme for litchi pedicle borers, and obtaining characteristic parameters of a standard growth environment for litchi pedicle borers at a current time point according to the preset cultivation scheme;

[0034] Calculating the difference between each standard growth environment characteristic parameter and the corresponding real-time growth environment characteristic parameter to obtain a plurality of growth environment characteristic parameter difference values;

[0035] Preset growth environment parameter deviation thresholds, and respectively compare the difference between each growth environment characteristic parameter and the preset growth environment parameter deviation threshold;

[0036] Extract the real-time growth environment characteristic parameters corresponding to the growth environment characteristic parameter difference greater than the preset growth environment parameter deviation threshold, and mark them as the real-time growth environment characteristic parameters that need to be regulated; obtain the difference between the real-time growth environment characteristic parameters that need to be regulated and the standard growth environment characteristic parameters;

[0037] A recommended environment adjustment plan is generated based on the difference between the real-time growth environment characteristics that need to be regulated and processed and the standard growth environment characteristic parameters.

[0038] The artificial rearing method of litchi peduncle borer also includes the following steps:

[0039] Prepare in advance the activity recovery and control measures corresponding to various abnormal growth and activity events of litchi stem borer populations;

[0040] Construct a knowledge graph, and import the activity recovery control measures corresponding to various abnormal growth activity events of litchi stem borer populations that have been prepared in advance into the knowledge graph;

[0041] Obtaining abnormal events of population growth activity currently occurring in the litchi peduncle borer according to the abnormal prediction results of the restorative growth environment;

[0042] Importing the abnormal growth activity events of the current population of litchi peduncle borer into the knowledge graph for matching, and obtaining corresponding activity recovery and control measures;

[0043] The obtained activity recovery control measures are output as recommended activity recovery control measures.

[0044] The present invention also discloses an artificial rearing system for litchi stalk borers based on growth environment regulation, the artificial rearing system for litchi stalk borers comprising a memory and a processor, the memory storing a program for an artificial rearing method for litchi stalk borers, and when the program for an artificial rearing method for litchi stalk borers is executed by the processor, the steps of any one of the methods for artificial rearing of litchi stalk borers are implemented.

[0045] The present invention also discloses a computer-readable storage medium, which includes a program for the artificial rearing method of litchi stalk borer. When the program for the artificial rearing method of litchi stalk borer is executed by a processor, the steps of any one of the methods for artificial rearing of litchi stalk borer are implemented.

[0046] The present invention solves the technical defects existing in the background technology, and has the following beneficial effects: by constructing a prediction model through big data analysis and deep learning, the environmental conditions of the growth area of ​​​​the litchi stalk borer can be monitored in real time and predicted, and abnormal growth environment conditions can be warned in advance so as to make environmental adjustments in time to avoid abnormal growth activity of the litchi stalk borer population; when abnormal growth activity has already occurred in the population, it can also be discovered in time and prompted to carry out growth activity recovery treatment, thereby improving the scientificity and effectiveness of artificial breeding of litchi stalk borer, saving unnecessary human intervention costs, and ensuring the normal growth and population stability of litchi stalk borer during the breeding process. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0048] Figure 1 This is a flow chart of the overall method of the artificial rearing method of the litchi pedicle borer;

[0049] Figure 2 This is a partial method flow chart of the artificial rearing method of litchi peduncle borer;

[0050] Figure 3 This is the system block diagram of the artificial breeding system for litchi stem borer. DETAILED DESCRIPTION

[0051] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0052] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0053] like Figure 1 As shown, the present invention discloses an artificial breeding method for litchi stem borer based on growth environment regulation, comprising the following steps:

[0054] S102: obtaining the fostering log information of the litchi stem borer through a big data network, performing feature extraction and analysis processing on the fostering log information, and obtaining a training data packet; wherein the training data includes early warning growth environment characteristic parameters and restorative growth environment characteristic parameters;

[0055] S104: constructing a growth environment prediction model for litchi pedicle borers according to the training data packet and in combination with a deep learning network; obtaining real-time growth environment characteristic parameters of the litchi pedicle borer growth area at a preset time node, importing the real-time growth environment characteristic parameters into the growth environment prediction model, and obtaining a prediction result;

[0056] S106: If the prediction result is a normal growth environment prediction result, it means that under the current environmental conditions, no human intervention is required for the litchi peduncle borer;

[0057] S108: If the prediction result is a warning growth environment abnormality prediction result, it means that under the current environmental conditions, after the litchi pedicle borer lives for a preset time, the litchi pedicle borer will have an abnormal population growth activity event, indicating that the current growth environment of the litchi pedicle borer needs to be artificially adjusted, and then an environmental adjustment warning information is generated;

[0058] S110: If the prediction result is a restorative growth environment abnormality prediction result, it means that the litchi stalk borer has experienced an abnormal population growth activity event, indicating that the litchi stalk borer needs to be artificially restored in terms of growth activity, and activity recovery warning information is generated.

[0059] It should be noted that the big data network is first used to obtain the foster log information of litchi pedicle borers. These log information contains various data related to the foster log information. Then, the foster log information is feature extracted and analyzed to obtain a training data packet. The early warning growth environment characteristic parameters and the recovery growth environment characteristic parameters in this training data packet are key data for subsequent model construction. According to the obtained training data packet, a growth environment prediction model of litchi pedicle borers is constructed in combination with a deep learning network. The deep learning network has strong learning and prediction capabilities and can mine complex relationships in the data. The real-time growth environment characteristic parameters of the litchi pedicle borer growth area are obtained at the preset time node. These real-time parameters reflect the actual situation of the current litchi pedicle borer growth environment. These real-time parameters are imported into the growth environment prediction model, and the model will give a prediction result. If the prediction result is a normal growth environment prediction result, it means that the current environmental conditions are suitable for the growth of litchi pedicle borers and no human intervention is required, which helps save manpower and resources, and also shows that the current environmental management measures are effective. If the prediction result is an abnormal early warning growth environment prediction result, this is an early warning signal. It shows that under the current environmental conditions, after a period of time (preset time), the litchi stalk borer will have an event of abnormal population growth activity. At this time, the current growth environment of the litchi stalk borer needs to be artificially adjusted, and environmental adjustment warning information needs to be generated. For example, the occurrence of abnormal population growth activity can be avoided by adjusting environmental factors such as temperature, humidity, and light. If the prediction result is a restorative growth environment abnormality prediction result, it means that the litchi stalk borer has had an event of abnormal population growth activity. At this time, the litchi stalk borer needs to be artificially restored in growth activity, and activity recovery warning information needs to be generated. For example, special nutritional supplements or environmental improvement measures need to be taken to restore its population growth activity.

[0060] In summary, by building a prediction model through big data analysis and deep learning, it is possible to monitor the environmental conditions of the litchi stem borer's growth area in real time and make predictions. It can warn of abnormal growth environment conditions in advance so that environmental adjustments can be made in time to avoid abnormal growth activity of the litchi stem borer population; when the population has abnormal growth activity, it can also be discovered in time and prompt for growth activity recovery treatment, thereby improving the scientificity and effectiveness of artificial breeding of litchi stem borers, saving unnecessary human intervention costs, and ensuring the normal growth and population stability of litchi stem borers during the breeding process.

[0061] Preferably, the foster log information of litchi stem borer is obtained through a big data network, and feature extraction and analysis processing are performed on the foster log information to obtain a training data packet, specifically:

[0062] Acquire the fostering log information of litchi stem borers through a big data network, and acquire abnormal growth environment events during the fostering process of litchi stem borers according to the fostering log information;

[0063] At the same time, the growth characteristic state corresponding to the litchi pedicle borer when each abnormal growth environment event occurs is obtained according to the fostering log information; wherein the growth characteristic state includes an abnormal growth activity state and a normal growth activity state;

[0064] Statistical analysis is performed on the growth characteristic states corresponding to the litchi pedicle borer when each abnormal growth environment event occurs, and the total number of abnormal growth activity states and the total number of normal growth activity states of the litchi pedicle borer when each abnormal growth environment event occurs are obtained;

[0065] The total number of abnormal growth activity states of litchi pedicle borers when each abnormal growth environment event occurs is processed by ratio to the total number of normal growth activity states, so as to obtain the abnormal growth activity ratio of the litchi pedicle borer population when each abnormal growth environment event occurs;

[0066] Compare the abnormal proportion of population growth activity with a preset proportion threshold; and only extract the abnormal growth environment events corresponding to the abnormal proportion of population growth activity greater than the preset proportion threshold;

[0067] Obtaining, from the fostering log information, growth environment characteristic parameters corresponding to the growth area of ​​the litchi pedicle borer within a preset time period before the occurrence of an abnormal growth environment event corresponding to the abnormal growth activity ratio of the population being greater than a preset ratio threshold, and defining them as early warning growth environment characteristic parameters;

[0068] And obtaining the growth environment characteristic parameters corresponding to the growth area of ​​the litchi pedicle borer when an abnormal growth environment event corresponding to the abnormal growth activity ratio of the population is greater than a preset ratio threshold in the fostering log information, and defining them as recovery growth environment characteristic parameters;

[0069] A training data packet is generated according to each warning growth environment characteristic parameter and the restorative growth environment characteristic parameter.

[0070] It should be noted that the fostering log information of litchi pedicle borers is obtained through the big data network. These log information are important data sources for comprehensively recording the fostering process of litchi pedicle borers. Then, based on these log information, abnormal growth environment events during the fostering process are found. For example, it may include sudden changes in temperature, abnormal fluctuations in humidity, changes in food supply, and other situations that affect the growth environment of litchi pedicle borers. At the same time, the corresponding growth characteristic state of litchi pedicle borers at the time of each abnormal growth environment event is determined based on the fostering log information. Here, the growth characteristic state is divided into abnormal growth activity state (such as slow growth, decreased reproduction rate, etc.) and normal growth activity state. This step helps to establish the connection between the growth environment and the growth state of litchi pedicle borers. Then, the growth characteristic state of litchi pedicle borers under each abnormal growth environment event is statistically analyzed. The total number of abnormal growth activity states and the total number of normal growth activity states of litchi pedicle borers at each abnormal growth environment event are calculated. For example, if in a certain abnormal growth environment event with temperature fluctuations, after multiple fostering records are counted, it is found that the abnormal growth activity state of litchi pedicle borers occurs 5 times and the normal growth activity state occurs 3 times.

[0071] By performing ratio processing on the total number of abnormal growth activity states and the total number of normal growth activity states, the abnormal proportion of population growth activity of litchi pedicle borers when each abnormal growth environment event occurs is obtained. This proportion reflects the possibility of abnormal growth activity of litchi pedicle borers under a specific abnormal growth environment event. The abnormal proportion of population growth activity is compared with the preset proportion threshold, and only the abnormal growth environment events corresponding to the abnormal proportion of population growth activity greater than the preset proportion threshold are extracted. This screening process can focus on those abnormal growth environment events that have a greater impact on the growth activity of litchi pedicle borers, exclude some events with less impact, and thus improve the pertinence of subsequent analysis. The growth environment characteristic parameters corresponding to the growth area of ​​litchi pedicle borers in the preset time period before the occurrence of the abnormal growth environment event corresponding to the abnormal proportion of population growth activity greater than the preset proportion threshold are obtained in the foster care log information, and are defined as early warning growth environment characteristic parameters. These parameters can be used as a basis for early warning of possible problems in the growth of litchi pedicle borers. For example, if the humidity continues to drop before an abnormal growth environment event with a high proportion of abnormal population growth activity, this humidity parameter may be a warning growth environment characteristic parameter. At the same time, the growth environment characteristic parameters corresponding to the growth area of ​​the litchi stalk borer when the abnormal growth environment event corresponding to the abnormal proportion of population growth activity greater than the preset proportion threshold are obtained, and defined as the restorative growth environment characteristic parameters. These parameters are helpful to determine the environmental factors that need to be adjusted when the growth activity of the litchi stalk borer has already shown abnormalities. Finally, a training data packet is generated based on these screened warning growth environment characteristic parameters and restorative growth environment characteristic parameters. This data packet will be used for subsequent operations such as building a growth environment prediction model for the litchi stalk borer.

[0072] This step accurately defines the early warning growth environment characteristic parameters and the restorative growth environment characteristic parameters through in-depth mining and analysis of the litchi pedicle borer fostering log information, and then generates a training data packet. This method can focus on the abnormal growth environment events that have a greater impact on the growth activity of the litchi pedicle borer, and effectively screen out the key growth environment characteristic parameters. This helps to build a more accurate prediction model, so as to more accurately predict the changes in the growth environment of the litchi pedicle borer, give early warning of possible growth activity abnormalities, and when an abnormality has occurred, it can take effective countermeasures based on the restorative growth environment characteristic parameters to improve the success rate and efficiency of litchi pedicle borer fostering.

[0073] Preferably, a growth environment prediction model for litchi pedicle borer is constructed based on the training data packet and combined with a deep learning network, such as Figure 2 As shown, specifically:

[0074] S202: constructing a deep learning network, and obtaining early warning growth environment characteristic parameters and restorative growth environment characteristic parameters of a training data packet;

[0075] S204: using the early warning growth environment characteristic parameter as an early warning condition node, using the restorative growth environment characteristic parameter as a restorative condition node, and initializing a weight matrix according to the early warning condition node and the restorative condition node;

[0076] S206: Input the initialized weight matrix into the deep learning network, use the forward propagation algorithm to import the input weight matrix into each layer of the network to calculate the output, and then use the back propagation algorithm to adjust the weight matrix according to the error between the output and the true value, and continuously perform iterative operations until the error is less than a preset threshold, thereby obtaining an iteratively optimized weight matrix;

[0077] S208: constructing a prediction model, and embedding the iteratively optimized weight matrix into the prediction model to obtain a growth environment prediction model for litchi peduncle borer.

[0078] It should be noted that a deep learning network is first constructed, which is a powerful machine learning model structure that can automatically learn complex patterns in data. Then, the warning growth environment characteristic parameters and the recovery growth environment characteristic parameters in the training data packet are obtained. These parameters are obtained through a detailed analysis of the foster log information and contain key information related to the growth environment of the litchi pedicle borer. The warning growth environment characteristic parameters are used as warning condition nodes, and the recovery growth environment characteristic parameters are used as recovery condition nodes. The weight matrix is ​​initialized according to these condition nodes. The weight matrix plays a vital role in the deep learning network. It determines the importance of the input data when it is transmitted between the layers in the network. By converting different types of characteristic parameters into condition nodes to initialize the weight matrix, an initial direction can be provided for the learning of the network. The initialized weight matrix is ​​input into the deep learning network, and the input weight matrix is ​​imported into each layer of the network to calculate the output using the forward propagation algorithm. The forward propagation algorithm is the process of calculating the activation value of neurons from the input layer to the output layer in sequence according to the structure of the network. Then, the weight matrix is ​​adjusted according to the error between the output and the true value through the back propagation algorithm. The back propagation algorithm is an effective method for calculating gradients. By calculating the derivative of the error with respect to the weight matrix, each element in the weight matrix is ​​adjusted to make the network output closer to the true value. This iterative operation is performed continuously until the error is less than the preset threshold. This process is the core of deep learning network learning. Through multiple iterations, the network continuously adjusts the weight matrix to better fit the data. A prediction model is constructed, and the iteratively optimized weight matrix is ​​embedded into this prediction model to obtain a prediction model for the growth environment of litchi pedicle borers. This prediction model can use the learned weight matrix to process the input data related to the growth environment of litchi pedicle borers, thereby predicting the growth environment of litchi pedicle borers.

[0079] By using the characteristic parameters in the training data packet to build a deep learning network, after a series of operations such as initializing the weight matrix and iteratively optimizing the weight matrix, a prediction model for the growth environment of the litchi pedicle borer is finally constructed. This method can fully mine the information in the training data, and through the powerful learning ability of the deep learning network, the prediction model can accurately predict the growth environment of the litchi pedicle borer based on the input growth environment characteristic parameters, providing an effective technical means for early warning of abnormal growth environment of the litchi pedicle borer and taking corresponding intervention measures, thereby helping to improve the scientificity and success rate of litchi pedicle borer breeding assistance.

[0080] Preferably, the real-time growth environment characteristic parameters of the litchi stem borer growth area are obtained at a preset time node, and the real-time growth environment characteristic parameters are imported into the growth environment prediction model to obtain the prediction result, specifically:

[0081] Acquire the real-time growth environment characteristic parameters of the litchi pedicle borer growth area at a preset time node, and standardize the real-time growth environment characteristic parameters to match the data format and characteristic scale in the training data package when building the model;

[0082] Importing the standardized real-time growth environment characteristic parameters into the growth environment prediction model, and organizing the standardized real-time parameters according to the input layer requirements when constructing the model;

[0083] Using the forward propagation mechanism of the deep learning network, starting from the input layer, matching calculations are performed through the hidden layers in sequence to obtain the matching degree between the real-time growth environment characteristic parameters and the nodes in the corresponding layer; the matching degree between the real-time growth environment characteristic parameters and the nodes in the corresponding layer is compared with a preset matching degree threshold;

[0084] If the matching degree is greater than a preset matching degree threshold, then obtaining node information of a node corresponding to a node whose matching degree is greater than the preset matching degree threshold; wherein the node information includes a warning condition node and a recovery condition node;

[0085] If the node information of the node corresponding to the matching degree greater than the preset matching degree threshold is a warning condition node, a warning growth environment abnormality prediction result is generated; if the node information of the node corresponding to the node with a matching degree greater than the preset matching degree threshold is a recovery condition node, a recovery growth environment abnormality prediction result is generated;

[0086] If the matching degree is not greater than the preset matching degree threshold, the standardized real-time parameters are passed to the next layer until the output layer is reached. If there is still no node corresponding to a matching degree greater than the preset matching degree threshold after reaching the output layer, a normal prediction result of the growth environment is generated.

[0087] It should be noted that after obtaining the real-time growth environment characteristic parameters of the growth area of ​​the litchi pedicle borer at the preset time node, standardization processing must be performed first. This is because the data in the training data package used when building the growth environment prediction model has a specific format and characteristic scale. Standardization processing can make the real-time growth environment characteristic parameters consistent with the training data, ensuring that the model can correctly process the input data. The standardized real-time growth environment characteristic parameters are imported into the growth environment prediction model, and the data is organized according to the requirements of the input layer when the model is built. This step is to allow the data to enter the deep learning network for calculation in the way expected by the model. Using the forward propagation mechanism of the deep learning network, starting from the input layer, matching calculations are performed through the hidden layers in sequence. In this process, the real-time growth environment characteristic parameters are interactively calculated with the nodes in each layer, so as to obtain the matching degree between the real-time growth environment characteristic parameters and the nodes in the corresponding layer. This matching degree reflects the similarity between the real-time data and the characteristic patterns represented by each node in the model. The matching degree is compared with the preset matching degree threshold. If the matching degree is greater than the preset matching degree threshold, the node information of the node corresponding to the matching degree greater than the preset matching degree threshold is obtained. The node information here includes early warning condition nodes and recovery condition nodes. If the corresponding node information is a warning condition node, a warning growth environment abnormality prediction result is generated, which means that according to the current real-time growth environment characteristic parameters, abnormal growth environment may occur in the future, and measures need to be taken in advance to prevent it. If it is a restorative condition node, a restorative growth environment abnormality prediction result is generated, indicating that the current growth environment may have an adverse effect on the growth of litchi pedicle borers and requires restorative intervention. If the matching degree is not greater than the preset matching degree threshold, the standardized real-time parameters are passed to the next layer, and this process continues until the output layer is reached. If there is still no node corresponding to a matching degree greater than the preset matching degree threshold after reaching the output layer, a normal growth environment prediction result is generated, indicating that the current growth environment is suitable for the growth of litchi pedicle borers and no additional intervention is required.

[0088] By obtaining the real-time growth environment characteristic parameters of the litchi pedicle borer's growth area, importing them into the growth environment prediction model after preprocessing, using the forward propagation mechanism of the deep learning network to calculate the matching degree, and generating different prediction results based on the comparison results of the matching degree and the preset threshold. This method can timely and accurately evaluate the current growth environment conditions of the litchi pedicle borer, give early warning of possible abnormal growth environment conditions (early warning prediction results), promptly detect abnormal growth environment conditions that have already occurred (restoration prediction results), or confirm that the growth environment is normal and no intervention is required, thereby providing a scientific basis for the artificial breeding of litchi pedicle borers, helping to improve the efficiency and success rate of litchi pedicle borer breeding, and also helping to rationally arrange resources and avoid unnecessary intervention measures.

[0089] The artificial rearing method of litchi peduncle borer also includes the following steps:

[0090] Obtaining a preset cultivation scheme for litchi pedicle borers, and obtaining characteristic parameters of a standard growth environment for litchi pedicle borers at a current time point according to the preset cultivation scheme;

[0091] Calculating the difference between each standard growth environment characteristic parameter and the corresponding real-time growth environment characteristic parameter to obtain a plurality of growth environment characteristic parameter difference values;

[0092] Preset growth environment parameter deviation thresholds, and respectively compare the difference between each growth environment characteristic parameter and the preset growth environment parameter deviation threshold;

[0093] Extract the real-time growth environment characteristic parameters corresponding to the growth environment characteristic parameter difference greater than the preset growth environment parameter deviation threshold, and mark them as the real-time growth environment characteristic parameters that need to be regulated; obtain the difference between the real-time growth environment characteristic parameters that need to be regulated and the standard growth environment characteristic parameters;

[0094] A recommended environment adjustment plan is generated based on the difference between the real-time growth environment characteristics that need to be regulated and processed and the standard growth environment characteristic parameters.

[0095] It should be noted that the preset cultivation scheme of the litchi pedicle borer is first obtained, which stipulates the cultivation method and environmental conditions of the litchi pedicle borer under ideal conditions. According to this preset cultivation scheme, the standard growth environment characteristic parameters of the litchi pedicle borer at the current time node can be obtained. For example, the preset cultivation scheme stipulates a specific temperature range, humidity range, light intensity, etc. as the ideal environmental parameters for the normal growth of the litchi pedicle borer. The difference between each standard growth environment characteristic parameter and the corresponding real-time growth environment characteristic parameter is calculated. The real-time growth environment characteristic parameter is the parameter of the current litchi pedicle borer growth environment obtained by actual measurement. By calculating the difference, the degree of deviation between the actual environment and the ideal environment can be intuitively understood. The growth environment parameter deviation threshold is preset, and this threshold is used to determine whether the deviation between the actual environment and the ideal environment reaches the degree that needs to be adjusted. The difference of each growth environment characteristic parameter is compared with the preset growth environment parameter deviation threshold. The real-time growth environment characteristic parameter corresponding to the growth environment characteristic parameter difference greater than the preset growth environment parameter deviation threshold is extracted, and marked as the real-time growth environment characteristic parameter that needs to be regulated. This step can screen out those environmental parameters that may have a greater impact on the growth of litchi pedicle borers and need to be adjusted. Obtain the difference between the real-time growth environment characteristic parameters that need to be regulated and processed and the standard growth environment characteristic parameters. Generate a recommended environmental adjustment plan based on this difference. For example, if the humidity difference is -10% (that is, the actual humidity is 10% lower than the standard humidity), then the recommended environmental adjustment plan may be to increase the humidity. The specific adjustment range can be determined according to the size of the difference, such as increasing the humidity by spraying, etc., so that the humidity is close to the standard humidity value.

[0096] By comparing the standard growth environment characteristic parameters in the preset cultivation plan of litchi pedicle borer with the actual real-time growth environment characteristic parameters, the real-time growth environment characteristic parameters that deviate greatly from the ideal state and need to be adjusted are screened out, and the recommended environment adjustment plan is generated based on the difference between them and the standard value. This method can accurately identify the problems existing in the current growth environment of litchi pedicle borer, provide a clear direction and specific adjustment plan for adjusting the growth environment of litchi pedicle borer, help maintain the stability of the growth environment of litchi pedicle borer, and improve the success rate and quality of litchi pedicle borer cultivation.

[0097] The artificial rearing method of litchi peduncle borer also includes the following steps:

[0098] Prepare in advance the activity recovery and control measures corresponding to various abnormal growth and activity events of litchi stem borer populations;

[0099] Construct a knowledge graph, and import the activity recovery control measures corresponding to various abnormal growth activity events of litchi stem borer populations that have been prepared in advance into the knowledge graph;

[0100] Obtaining abnormal events of population growth activity currently occurring in the litchi peduncle borer according to the abnormal prediction results of the restorative growth environment;

[0101] Importing the abnormal growth activity events of the current population of litchi peduncle borer into the knowledge graph for matching, and obtaining corresponding activity recovery and control measures;

[0102] The obtained activity recovery control measures are output as recommended activity recovery control measures.

[0103] It should be noted that first, the activity recovery control measures corresponding to various population growth activity abnormal events of litchi pedicle borer are prepared in advance. This means that corresponding solutions are formulated in advance for different population growth activity abnormalities that may occur in litchi pedicle borer, such as slow growth and decreased reproduction rate. These measures may include adjusting environmental factors (such as temperature, humidity, light, etc.), changing food supply, or adopting specific biological intervention methods. Then, a knowledge graph is constructed. The knowledge graph is a semantic network that can represent various entities (here, population growth activity abnormal events) and relationships (activity recovery control measures) in a structured manner. The pre-prepared activity recovery control measures are imported into the knowledge graph so that the knowledge graph can store these knowledge to deal with different abnormal events. According to the abnormal prediction results of the restorative growth environment, the abnormal population growth activity events of litchi pedicle borer that are currently occurring are obtained. This step is based on the previous prediction of the growth environment of litchi pedicle borer. When the abnormal restorative growth environment is predicted, it is determined which growth activity abnormal event has occurred in the current litchi pedicle borer population. The current population growth activity abnormal event is imported into the knowledge graph for matching. Since the knowledge graph has stored the corresponding relationship between various abnormal events and activity recovery control measures, the corresponding activity recovery control measures can be obtained through matching. Finally, the obtained activity recovery control measures are output as recommended activity recovery control measures. This recommended activity recovery control measure is a specific solution to the current abnormal growth activity of the litchi stem borer population, which can be directly used to guide actual recovery operations.

[0104] By prefabricating active recovery control measures and constructing a knowledge graph, it is possible to quickly and accurately match the corresponding active recovery control measures and output recommended solutions based on the predicted results of abnormal growth and activity events of litchi pedicle borer populations. This method improves the efficiency and accuracy of responding to abnormal growth and activity of litchi pedicle borer populations, provides effective technical support for timely restoration of normal growth of litchi pedicle borer, and helps to ensure the health and stability of litchi pedicle borer populations during artificial breeding.

[0105] The present invention also discloses an artificial breeding system for litchi stem borers based on growth environment regulation, such as Figure 3As shown, the artificial rearing system for litchi stalk borer includes a memory 41 and a processor 62. The memory 41 stores a program for the artificial rearing method for litchi stalk borer. When the program for the artificial rearing method for litchi stalk borer is executed by the processor 62, the steps of any one of the methods for artificial rearing of litchi stalk borer are implemented.

[0106] The present invention also discloses a computer-readable storage medium, which includes a program for the artificial rearing method of litchi stalk borer. When the program for the artificial rearing method of litchi stalk borer is executed by a processor, the steps of any one of the methods for artificial rearing of litchi stalk borer are implemented.

[0107] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0108] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0109] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0110] Those skilled in the art can understand that: all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, and other media that can store program codes.

[0111] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0112] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for artificially raising litchi peduncle borer based on growth environment regulation, characterized in that: The following steps are involved: Obtaining the foster log information of litchi stem borer through a big data network, extracting features and analyzing the foster log information, and obtaining a training data packet; wherein the training data includes early warning growth environment characteristic parameters and restorative growth environment characteristic parameters; Constructing a growth environment prediction model for litchi pedicle borers according to the training data packet and in combination with a deep learning network; obtaining real-time growth environment characteristic parameters of the litchi pedicle borer growth area at a preset time node, importing the real-time growth environment characteristic parameters into the growth environment prediction model, and obtaining a prediction result; If the prediction result is a normal growth environment prediction result, it means that under the current environmental conditions, no human intervention is required for litchi stem borer; If the prediction result is a warning growth environment abnormality prediction result, it means that under the current environmental conditions, after the litchi pedicle borer lives for a preset period of time, the litchi pedicle borer will have an abnormal population growth activity event, indicating that the current growth environment of the litchi pedicle borer needs to be artificially adjusted, and then an environmental adjustment warning information is generated; If the prediction result is an abnormal recovery growth environment prediction result, it means that the litchi stalk borer has experienced abnormal population growth activity, which means that the litchi stalk borer needs to be artificially restored in terms of growth activity, and activity recovery warning information will be generated.

2. The method for artificially supporting litchi peduncle borer based on growth environment regulation according to claim 1, characterized in that: The foster log information of litchi stem borer is obtained through the big data network, and the feature extraction and analysis processing of the foster log information are performed to obtain a training data packet, which is specifically: Acquire the fostering log information of litchi stem borers through a big data network, and acquire abnormal growth environment events during the fostering process of litchi stem borers according to the fostering log information; At the same time, the growth characteristic state corresponding to the litchi pedicle borer when each abnormal growth environment event occurs is obtained according to the fostering log information; wherein the growth characteristic state includes an abnormal growth activity state and a normal growth activity state; Statistical analysis is performed on the growth characteristic states corresponding to the litchi pedicle borer when each abnormal growth environment event occurs, and the total number of abnormal growth activity states and the total number of normal growth activity states of the litchi pedicle borer when each abnormal growth environment event occurs are obtained; The total number of abnormal growth activity states of litchi pedicle borers when each abnormal growth environment event occurs is processed by ratio to the total number of normal growth activity states, so as to obtain the abnormal growth activity ratio of the litchi pedicle borer population when each abnormal growth environment event occurs; Compare the abnormal proportion of population growth activity with a preset proportion threshold; and extract only the abnormal growth environment events corresponding to the abnormal proportion of population growth activity greater than the preset proportion threshold; Obtaining, from the fostering log information, growth environment characteristic parameters corresponding to the growth area of ​​the litchi pedicle borer within a preset time period before the occurrence of an abnormal growth environment event corresponding to the abnormal growth activity ratio of the population being greater than a preset ratio threshold, and defining them as early warning growth environment characteristic parameters; And obtaining the growth environment characteristic parameters corresponding to the growth area of ​​the litchi pedicle borer when an abnormal growth environment event corresponding to the abnormal growth activity ratio of the population is greater than a preset ratio threshold in the fostering log information, and defining them as recovery growth environment characteristic parameters; A training data packet is generated according to each warning growth environment characteristic parameter and the restorative growth environment characteristic parameter.

3. The method for artificially supporting litchi peduncle borer based on growth environment regulation according to claim 1, characterized in that: A growth environment prediction model for litchi pedicle borer is constructed based on the training data packet and combined with a deep learning network, specifically: Construct a deep learning network and obtain the early warning growth environment characteristic parameters and the restorative growth environment characteristic parameters of the training data package; Using the early warning growth environment characteristic parameter as an early warning condition node, using the restorative growth environment characteristic parameter as a restorative condition node, and initializing a weight matrix according to the early warning condition node and the restorative condition node; The initialized weight matrix is ​​input into the deep learning network, and the input weight matrix is ​​imported into each layer of the network to calculate the output by using the forward propagation algorithm, and then the weight matrix is ​​adjusted according to the error between the output and the true value by the back propagation algorithm, and the iterative operation is continuously performed until the error is less than the preset threshold, and the iteratively optimized weight matrix is ​​obtained; A prediction model is constructed, and the iteratively optimized weight matrix is ​​embedded into the prediction model to obtain a growth environment prediction model for litchi peduncle borer.

4. The method for artificially supporting litchi peduncle borer based on growth environment regulation according to claim 1, characterized in that: At a preset time node, the real-time growth environment characteristic parameters of the litchi stem borer growth area are obtained, and the real-time growth environment characteristic parameters are imported into the growth environment prediction model to obtain the prediction result, specifically: Acquire the real-time growth environment characteristic parameters of the litchi pedicle borer growth area at a preset time node, and standardize the real-time growth environment characteristic parameters to match the data format and characteristic scale in the training data package when building the model; Importing the standardized real-time growth environment characteristic parameters into the growth environment prediction model, and organizing the standardized real-time parameters according to the input layer requirements when constructing the model; Using the forward propagation mechanism of the deep learning network, starting from the input layer, matching calculations are performed through the hidden layers in sequence to obtain the matching degree between the real-time growth environment characteristic parameters and the nodes in the corresponding layer; the matching degree between the real-time growth environment characteristic parameters and the nodes in the corresponding layer is compared with a preset matching degree threshold; If the matching degree is greater than a preset matching degree threshold, then obtaining node information of a node corresponding to a node whose matching degree is greater than the preset matching degree threshold; wherein the node information includes a warning condition node and a recovery condition node; If the node information of the node corresponding to the matching degree greater than the preset matching degree threshold is a warning condition node, a warning growth environment abnormality prediction result is generated; if the node information of the node corresponding to the node with a matching degree greater than the preset matching degree threshold is a recovery condition node, a recovery growth environment abnormality prediction result is generated; If the matching degree is not greater than the preset matching degree threshold, the standardized real-time parameters are passed to the next layer until the output layer is reached. If there is still no node corresponding to a matching degree greater than the preset matching degree threshold after reaching the output layer, a normal prediction result of the growth environment is generated.

5. The method for artificially raising litchi peduncle borer based on growth environment regulation according to claim 1, characterized in that: The artificial rearing method of litchi peduncle borer also includes the following steps: Obtaining a preset cultivation scheme for litchi pedicle borers, and obtaining characteristic parameters of a standard growth environment for litchi pedicle borers at a current time point according to the preset cultivation scheme; Calculating the difference between each standard growth environment characteristic parameter and the corresponding real-time growth environment characteristic parameter to obtain a plurality of growth environment characteristic parameter difference values; Preset growth environment parameter deviation thresholds, and respectively compare the difference between each growth environment characteristic parameter and the preset growth environment parameter deviation threshold; Extract the real-time growth environment characteristic parameters corresponding to the growth environment characteristic parameter difference greater than the preset growth environment parameter deviation threshold, and mark them as the real-time growth environment characteristic parameters that need to be regulated; obtain the difference between the real-time growth environment characteristic parameters that need to be regulated and the standard growth environment characteristic parameters; A recommended environment adjustment plan is generated based on the difference between the real-time growth environment characteristics that need to be regulated and processed and the standard growth environment characteristic parameters.

6. The method for artificially raising litchi peduncle borer based on growth environment regulation according to claim 1, characterized in that: The artificial rearing method of litchi peduncle borer also includes the following steps: Prepare in advance the activity recovery and control measures corresponding to various abnormal growth and activity events of litchi stem borer populations; Construct a knowledge graph, and import the activity recovery control measures corresponding to various abnormal growth activity events of litchi stem borer populations that have been prepared in advance into the knowledge graph; Obtaining abnormal population growth activity events currently occurring in the litchi peduncle borer according to the abnormal prediction result of the restorative growth environment; Importing the abnormal growth activity events of the current population of litchi peduncle borer into the knowledge graph for matching, and obtaining corresponding activity recovery and control measures; The obtained activity recovery control measures are output as recommended activity recovery control measures.

7. An artificial breeding system for litchi stem borer based on growth environment regulation, characterized in that: The artificial rearing system for litchi stalk borer includes a memory and a processor, wherein the memory stores a program for the artificial rearing method for litchi stalk borer. When the program for the artificial rearing method for litchi stalk borer is executed by the processor, the steps of the artificial rearing method for litchi stalk borer as described in any one of claims 1 to 6 are implemented.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a program for the artificial rearing method of litchi stalk borer. When the program for the artificial rearing method of litchi stalk borer is executed by a processor, the steps of the method for artificial rearing method of litchi stalk borer as described in any one of claims 1 to 6 are implemented.

Images