227 results about "Growth data" patented technology

The invention discloses a storage battery capacity attenuation trend prediction method, and belongs to the technical field of storage battery prediction. By collecting voltage, current and temperature data of each monomer in real time and combining historical capacity attenuation and internal resistance growth data, the method identifies a voltage and capacity difference value, evaluates cyclic stress non-uniform distribution, and determines a current sharing proportion and a load unbalance degree. Identifying an abnormal mode of new battery overload and aged battery deep discharge, constructing a mixing abnormal working condition identification mode, if the unbalance degree exceeds the standard, adaptively adjusting the charging and discharging time and the current switching frequency, establishing a load balance control framework, predicting the capacity attenuation rate and the residual cycle index of each monomer, and determining the capacity matching degree and the life matching degree; finally, a comprehensive residual life estimation value and a credibility interval are generated through fusion; the performance balance of the mixed battery pack is remarkably improved, the overall service life is prolonged, and the method is suitable for real-time monitoring of a battery management system.

The invention relates to the technical field of biological breeding, in particular to a biological breeding management method and system based on big data. The method comprises the following steps: acquiring original germplasm data; multi-source heterogeneous data integration is carried out on the original germplasm data, the data integration comprises genotype sequencing integration and historical phenotype integration, and integrated germplasm data is generated and stored in a database in a distributed mode; obtaining original growth data, performing data preprocessing, generating standardized germplasm resource comprehensive data including a crop image feature set and an environment response parameter set, and synchronizing the data to the database; and performing whole genome selection prediction on the integrated germplasm data based on the standardized germplasm resource comprehensive data to generate a genotype-phenotype prediction model. By integrating germplasm resources, field management, breeding management, phenotype management, genotype management and whole genome selection, the accuracy and adjustment flexibility of biological breeding management are improved.

The invention relates to the technical field of ecological environment protection, in particular to a mangrove forest carbon sink monitoring and metering method based on an unmanned aerial vehicle, a radar and an AI technology, and the method comprises the steps: 1, obtaining the laser radar data of a mangrove forest through a laser radar carried by the unmanned aerial vehicle; step 2, acquiring elevation data in a mangrove forest vegetation layer area, and obtaining point cloud data after topographic error correction; 3, obtaining a multispectral image of the mangrove forest, and obtaining a multispectral data matrix; 4, identifying forest growth data features, constructing a mangrove forest carbon sink prediction model, and predicting the mangrove forest carbon sink amount; step 5, marking the image region with the NDVI value higher than a preset NDVI threshold value as a blade over-dense region; and according to the area of the overdense leaf region and the multispectral data matrix, calculating a light depression factor by using a photosynthetic depression factor formula, determining the carbon sink deviation of the overdense leaf region by using a regional carbon sink deviation formula, and obtaining a real carbon sink value of the mangrove forest according to a carbon sink calculated value obtained by prediction.

The invention relates to the technical field of concrete curing, in particular to an intelligent control method and system for a precast concrete component curing kiln, and the method comprises the steps: obtaining historical and real-time environment parameters of the curing kiln, generating monitoring data with an identifier in combination with a component type and a current process stage, and matching a corresponding preset temperature and humidity regulation and control model. And optimizing the model by using a historical temperature and humidity response sample and component strength increase data to generate a personalized target temperature and humidity regulation and control model. And analyzing real-time parameters based on the model, outputting a control instruction containing a target temperature rise rate, a humidity threshold value and heat preservation time, and converting the control instruction into a kiln body adjustment control signal according to a dynamically adjusted execution rule. And finally, sending to an execution terminal to realize cooperative regulation and control of heating, humidification, ventilation and heat preservation systems. According to the method, multi-source data and process characteristics are fused, the adaptive learning ability is achieved, the maintenance environment can be accurately regulated and controlled, and intelligent, refined and efficient operation of the maintenance process is achieved.

The invention relates to the technical field of automobile batteries, in particular to a battery cell matching system with improved cycle life, comprising: a battery cell parameter acquisition module for acquiring initial capacity, internal resistance value and voltage curve of a to-be-matched battery cell for a new energy automobile; the dynamic attenuation modeling module is used for establishing a dynamic attenuation model according to capacity attenuation and internal resistance growth data in the battery cell recycling process; the matching optimization module is used for combining the output of the dynamic attenuation model with the current matching strategy and adjusting the battery cell group in the matching scheme to obtain a battery pack; the high-temperature correction module is used for dynamically adjusting a matching correction value of a high-temperature cycle attenuation coefficient according to the attenuation performance of the battery cells in the battery pack under high-temperature cycle; and the control module is used for determining a screening threshold value of the static parameters of the battery cells according to the maximum difference quantity of the initial capacities of the plurality of battery cells in the same battery pack, or determining a dynamic characteristic weight coefficient according to the internal resistance difference of the plurality of battery cells in the battery pack. According to the invention, the stability of battery cell matching is improved.

The invention discloses a corn drought resistance identification device and method, and belongs to the technical field of agricultural scientific research. According to the method, the problems of inaccurate environment control, long identification period, poor repeatability and the like of an existing method are solved, the drought resistance index of the corn is calculated through the random forest model, the drought resistance grade of the corn is divided, and the drought resistance index threshold value is set; the drought resistance indexes of the corn in different growth periods are analyzed, the drought resistance index threshold value is dynamically adjusted according to the corn growth data monitored in real time, and the drought resistance of the corn in different growth periods can be reflected more accurately; feature importance scores are calculated through a random forest model, feature variables are sorted, and it can be determined that certain features make the greatest contribution to the drought resistance index; based on feature importance analysis, suggestions on corn drought resistance research and breeding are provided, researchers can be helped to optimize breeding strategies, errors caused by subjective judgment are reduced, and the breeding efficiency of drought-resistant varieties is improved.

The invention discloses a collaborative management method for fishing and light complementation of a photovoltaic station, and relates to the technical field of photovoltaic station operation and maintenance. The method comprises the following steps: collecting data; screening important features by adopting a recursive feature selection method based on a random forest to obtain growth data, and calculating a feature weight of each feature by utilizing a TreeSHAP method; calculating the growth deviation degree corresponding to each feature in the growth data, calculating the aquatic organism efficacy index by combining with the feature weight, and drawing an efficacy curve; calculating a growth trend factor of the efficiency curve based on a least square method of a sliding window, and determining growth trend stability; the inertia weight of the particle swarm optimization algorithm is adjusted by using the growth trend stability, and the optimal state of the photovoltaic panel, namely the optimal photovoltaic panel inclination angle and the optimal reflective film angle, is obtained; photovoltaic panel power generation is predicted through the optimal photovoltaic panel inclination angle and the optimal reflective film angle, the optimal power generation amount of the photovoltaic station is obtained, the photovoltaic station is cooperatively managed according to the optimal photovoltaic panel inclination angle and the optimal reflective film angle, and the operation and maintenance state of the photovoltaic station is optimal.

The invention relates to the technical field of forestry intelligent monitoring, and particularly discloses an intelligent monitoring system for the growth condition of afforestation and greening nursery stocks based on deep learning, which is characterized in that multi-modal growth data of the nursery stocks are acquired through a multi-spectral imaging sensor, a three-dimensional laser scanning sensor and an environment monitoring sensor, and a standardized data set is formed through space-time alignment processing; extracting morphological structure and spectral response features through spatial domain and frequency domain parallel analysis, and generating multi-dimensional feature representation through cross-modal fusion; converting the parameters into growth state parameters by utilizing a feature recombination and space mapping technology; identifying an abnormal growth mode through time sequence dynamic analysis; and finally, adaptively adjusting the working parameters of the sensor according to the abnormal type to form closed-loop monitoring. The problem that an existing monitoring system cannot autonomously optimize a monitoring strategy according to the abnormal state is solved, and accurate monitoring and intelligent regulation and control of the nursery stock growth condition are achieved.

The invention relates to a road bougainvillea speetabilis intelligent fertilization supervision method and system, and the method comprises the steps: enabling collected multispectral image data and thermal imaging image data to be precisely matched with GPS coordinates, enabling soil component parameters to be aligned with the corresponding multispectral image data and thermal imaging image data, and forming a spatiotemporal data set, performing feature extraction on the multispectral image by adopting a convolutional neural network, constructing a growth attitude comprehensive analysis model, comparing a result output by the growth attitude comprehensive analysis model, namely a current growth state quantized value, with a preset standard based on historical growth data and a bougainvillea speetabilis growth rule, and calculating the missing amount of related fertilizers, a precise fertilization formula and a fertilization time node are determined by combining the current fertility condition of the soil, a control instruction is generated according to the fertilization formula and the time node, and the fertilization robot executes fertilization operation based on the control instruction and a preset path. The method has the effects of improving the fertilizer utilization rate, reducing the labor cost and improving the maintenance efficiency.

The invention is suitable for the technical field of three-dimensional planting management, and particularly relates to a three-dimensional planting field management method and system, and the method comprises the steps: dividing a three-dimensional agricultural planting region, cutting the planting region into a plurality of blocks, collecting the growth data of all agricultural plants through video monitoring equipment disposed in the planting region in advance, and storing the collected growth data in a database; identifying a risk point location, selecting an infrared anchor point, starting a laser generator integrated in the video monitoring equipment in advance, and projecting infrared laser to the infrared anchor point; sending a working instruction to the inspection equipment, shooting a snapshot of the planting area, reading the projection positions of the infrared anchor points, integrating all the infrared anchor points, generating an anchor point group, and positioning the real-time position of the inspection equipment when the real-time position coincides with the projection positions. By determining the management report, defective plants in three-dimensional planting can be found in time, health management of agricultural plants is enhanced, and intelligence and refinement of agricultural production are effectively improved.

The invention discloses an AI-based rice planting environment data informatization management system, and relates to the technical field of agricultural informatization. The method comprises the steps that an environment data acquisition module obtains multi-dimensional space-time marked environment parameters; the disturbance modeling module is used for analyzing environment change, identifying a disturbance structure and calculating a sudden change value; the growth stage parameter module is used for constructing a stage characteristic parameter set; the growth state modeling module is used for generating growth state data in combination with disturbance and a growth stage; the strategy generation module formulates a resource regulation and control strategy based on the growth state and the feedback information; the resource coupling analysis module is used for identifying coupling and interference relations among resources; the strategy constraint module is used for adjusting a resource strategy to generate a control strategy; and the control execution feedback module is used for executing the regulation and control instruction and feeding back growth data. Through high-precision space-time environment data fusion, deep coupling analysis of environment disturbance and growth states and resource regulation and control closed-loop optimization management, intelligent and precise monitoring and regulation and control of a rice planting environment are realized.

The invention discloses a tree growth prediction method and system based on a TCN time sequence convolution model, and the method comprises the steps: collecting the historical growth data of trees in a power transmission line corridor, and carrying out the data preprocessing; constructing a neural network model based on the TCN; dividing the preprocessed data into a training set and a test set, and training the TCN model by using the data of the training set; performing performance evaluation on the trained TCN model by using the test set; and deploying the trained TCN model in an actual application scene, and performing real-time growth prediction on the target tree species in the power transmission line corridor area. According to the method, through the model based on the time convolutional network (TCN), the growth condition of the tree species in the power transmission line corridor is accurately predicted. The TCN model captures the sequential relationship in the time sequence by using the convolution operation, thereby avoiding the problem of gradient disappearance or explosion easily occurring when the traditional recurrent neural network processes the long sequence, and improving the prediction accuracy.

The invention discloses an irrigation decision determination method considering a multi-source water conversion process, and relates to the technical field of agricultural water conservancy irrigation, and the method comprises the steps: obtaining a multi-spectral image and a thermal infrared image of crops in a target irrigation region, calculating a vegetation index and a leaf area index, and combining with the actually measured growth vigor data of the crops, a growth vigor model of crops is constructed by establishing a mapping relation, the rainfall, the irrigation volume, the canal system infiltration replenishment volume and the groundwater capillary rise volume of a target irrigation area are obtained, and a soil moisture income and expenditure model is constructed based on a water circulation process. According to the method, a plurality of paths such as rainfall, irrigation, canal system leakage and groundwater capillary rise are considered through the constructed soil water volume income and expenditure model, crop physiological moisture inflection points are accurately recognized by simulating the root zone volumetric moisture content and growth vigor response curve, threshold control irrigation based on crop requirements is achieved, and the crop yield is improved. And the multi-source input and conversion process of the farmland hydrological system is comprehensively reflected.

The invention discloses a hybrid model seedling growth prediction method based on ant colony algorithm optimization, and relates to the field of seedling cultivation, and the method comprises the following steps: collecting and preprocessing seedling growth data, and constructing a standardized data set; performing hybrid model training by using the standardized data set; the hybrid model training comprises the following steps: screening out key features by using an XGBoost sub-model, outputting a seedling growth prediction result through a GRU-KAN fusion sub-model, and optimizing hyper-parameters of the GRU-KAN fusion sub-model through an ant colony optimization algorithm; and inputting growth parameters of a to-be-predicted seedling into the optimized hybrid model, and outputting a corresponding seedling growth prediction result. According to the method, through a collaborative architecture of XGBoost and GRU-KAN and intelligent parameter adjustment of an ant colony optimization algorithm, adaptive feature screening, deep fusion and hyper-parameter efficient optimization of multi-source heterogeneous data are completed, and automatic and high-precision prediction of the seedling growth process is realized.

The invention discloses a live pig dynamic nutrition model construction method and system based on multiple breeding modes, and relates to the technical field of intelligent precise feeding with livestock breeding and information technology crossing, and the method comprises the following steps: S1, collecting macroscopic growth data and physiological indication data of live pigs through a sensor network deployed in a breeding environment; the macroscopic growth data comprises body weight, feed intake, water intake and environment temperature and humidity. According to the live pig dynamic nutrition model construction method and system based on the multiple breeding modes, by constructing a collaborative mechanism of data credibility grading, dual-core model cross validation and cyclic feedback optimization, the technical limitation that a traditional nutrition model depends on macroscopic data and neglects individual physiological state changes is effectively solved. According to the system, multi-source sensor data fusion analysis is utilized, and dynamic threshold judgment and a machine learning algorithm are combined, so that accurate perception and adaptation of real nutritional requirements of live pigs are realized, and the problem of disjunction between nutrition supply and physiological requirements is avoided from the source.

The invention discloses a hydraulic valve body unit machining process and device, and relates to the technical field of valve body machining, and the hydraulic valve body unit machining process comprises the steps that a hydraulic valve body unit is placed in an electrolyte tool container, a preset working voltage is set, and an electrolysis environment is formed; a plurality of observation points are arranged on the surface of the hydraulic valve body unit, and real-time thickness data of the oxidation film are obtained; calculating oxidation film growth data of the hydraulic valve body unit according to the oxidation film real-time thickness data of the plurality of observation points; comparing the oxide film growth data with a standard threshold range, and if the oxide film growth data accords with the standard threshold range, adjusting working parameters of an electrolyte tool container according to a comparison result of the oxide film growth data and the standard threshold range; and electrolytic machining is stopped, and the hydraulic valve body unit subjected to anodic oxidation is obtained. The oxidation film growth data are monitored in real time in the anodic oxidation process, and the process parameters are adjusted in real time, so that the oxidation film preparation quality of the valve body workpiece can be effectively improved.

The invention discloses a forage grass drought loss assessment method, and relates to the technical field of forage grass drought loss assessment, and the method comprises the following steps: S1, collecting forage grass basic data in a drought influence period, the data comprising meteorological data, growth data, quality detection data and field management data; s2, preprocessing the basic data to obtain standardized data; s3, constructing a two-dimensional loss evaluation model, wherein the model comprises a yield loss calculation sub-model and a quality loss calculation sub-model; s4, respectively inputting the standardized data into the two sub-models to obtain a forage grass yield loss rate and a quality loss coefficient; and S5, in combination with the yield loss rate and the quality loss coefficient, weight coupling is set according to the forage grass type to calculate the total loss rate, and evaluation is completed. According to the forage grass drought loss evaluation method, the quality loss is taken into evaluation, and the problem that the feeding cost is high due to neglect of the quality is avoided.

The invention relates to the technical field of automobile batteries, in particular to a battery cell matching system with improved cycle life, comprising: a battery cell parameter acquisition module for acquiring initial capacity, internal resistance value and voltage curve of a to-be-matched battery cell for a new energy automobile; the dynamic attenuation modeling module is used for establishing a dynamic attenuation model according to capacity attenuation and internal resistance growth data in the battery cell recycling process; the matching optimization module is used for combining the output of the dynamic attenuation model with the current matching strategy and adjusting the battery cell group in the matching scheme to obtain a battery pack; the high-temperature correction module is used for dynamically adjusting a matching correction value of a high-temperature cycle attenuation coefficient according to the attenuation performance of the battery cells in the battery pack under high-temperature cycle; and the control module is used for determining a screening threshold value of the static parameters of the battery cells according to the maximum difference quantity of the initial capacities of the plurality of battery cells in the same battery pack, or determining a dynamic characteristic weight coefficient according to the internal resistance difference of the plurality of battery cells in the battery pack. According to the invention, the stability of battery cell matching is improved.

The invention discloses a method for constructing a sugarcane biomass 3D prediction model based on a dynamic video, and the method comprises the following steps: (1) actual measurement and investigation of sugarcane growth data: investigating the plant height, stem height, stem thickness, leaf area and stem leaf biomass of sugarcane every month until the sugarcane is mature; (2) carrying out 3D Gaussian spattering phenotype data extraction, data acquisition and data preprocessing; (3) 3D model construction and data verification, including three-dimensional reconstruction, point cloud data processing, phenotype analysis, growth trend tracking and biomass 3D prediction model establishment; and establishment of a biomass 3D prediction model: specifically, based on the key indexes extracted from the point cloud data and the actually measured biomass true value, the biomass prediction model established by adopting a random forest regression algorithm is the biomass 3D prediction model. The problems that in the prior art, data collection is prone to being interfered by the environment, the model generalization ability is insufficient, and the mapping relation between the form and the biomass is indefinite are solved.

The intelligent apple planting management and expert decision-making system comprises a sensor data acquisition module, an image data acquisition module, an edge calculation module, an expert knowledge module, an intelligent analysis module, an expert decision-making module, a wireless communication module, an equipment control module, a growth data tracing analysis module and a user interaction control module. Environmental and growth data are collected through a sensor and an image data acquisition module, and are transmitted to an edge calculation module through wireless communication. The system combines artificial intelligence technology analysis and expert knowledge, evaluates the growth condition of fruit trees, provides a scientific planting strategy, supports automatic control of agricultural operations such as irrigation and fertilization, realizes full-cycle tracing and optimization of growth data, and aims to provide more accurate production management parameters for planters, provide decision basis for scientific and standardized production, and improve the production efficiency. Apple quality is improved, diseases are reduced, planting management is optimized, and orchard production benefits are increased.

The invention discloses a beef cattle breeding decision-making method and device based on environmental factor analysis and a medium, and the method comprises the steps: extracting environmental factor time sequence features, and distributing corresponding environmental factor weights to the environmental factor time sequence features; determining interaction items corresponding to the environmental factors and influence effect values of the interaction items and the environmental factors on beef cattle growth, and constructing an interaction matrix according to the influence effect values; inputting the environment data and the growth data into an LSTM-Transform model, and outputting the predicted weight of the beef cattle in a future preset time period according to the environment factor weight and the interaction matrix; calculating the heat stress index of the beef cattle according to the growth data, and calculating the weight gain income of the beef cattle according to the predicted weight; and according to the heat stress index, the weight gain income and the carbon emission, constructing an objective function for deciding a beef cattle breeding strategy, and based on an NSGA-II algorithm, according to the objective function, outputting the beef cattle breeding strategy.

The invention provides a forest sample plot monitoring system, the monitoring system comprises a three-dimensional live-action modeling subsystem, an Internet of Things monitoring perception subsystem and a data management analysis subsystem, the three-dimensional live-action modeling subsystem is used for collecting sky point cloud data and ground point cloud data, and constructing a forest sample plot three-dimensional model; the Internet of Things monitoring and sensing subsystem is used for collecting tree growth associated information collected by a sensor and gathering and transmitting the tree growth associated information to the data management and analysis subsystem; and the data management and analysis subsystem is used for combining the forest sample plot three-dimensional model with the tree growth associated information, and analyzing to obtain forest accumulation variation and tree growth data. Through high-precision, high-efficiency and automatic reconstruction of the three-dimensional real scene of the forest sample plot, automatic, real-time and continuous monitoring of key growth parameters such as diameters at breast height of all sample trees in the sample plot is realized, and the intelligent level and operation efficiency of forest monitoring are improved.

The invention relates to the technical field of field character detection, in particular to a crop breeding-based field character detection method, which comprises the following steps of: determining an environment penalty value according to detected color data and soil data of field crops, determining a yield reward value according to the detected field crop growth data and the nutrition data stored in the crop management platform, and determining a fertilizer utilization value according to the detected field soil element data; performing weighted summation to generate a fertilization adjustment reward function, wherein a dynamic coefficient is determined according to crop breeding stage big data fitting of a crop management platform; the fertilization amount reinforcement learning model is updated regularly based on the fertilization adjustment reward function, and the fertilization amount reinforcement learning model determines the breeding multi-element recommended fertilization amount based on the Critic network and the Actor network. According to the method, collaborative closed-loop detection and field detection data mining of stress resistance, yield increase and fertilizer saving of the reinforcement learning model in combination with breeding are realized.

The application discloses an online semantic momentum stabilization system and method using a Welford algorithm and an optimistic concurrent control. In view of the semantic centroid calculation deviation and timing conflict problems caused by data noise, distribution drift and distributed out-of-order writing in the process of continuous incremental input of a large-scale vector database, the application combines a Welford online iterative algorithm to realize high-precision incremental updating of the semantic centroid mean and variance under constant space complexity. Meanwhile, a distributed optimistic concurrent control mechanism is used to implement version revision number checking on semantic segments and block obsolete data coverage caused by cross-node out-of-order arrival. The application guarantees the numerical stability and timing consistency of semantic feature expression in a dynamic growing data environment.

The invention discloses a student internal drive cultivation method and system, and relates to the field of education management, and the method comprises the steps: obtaining basic information data, academic performance data and practice data of students; according to the academic performance data and the practice data, performing primary evaluation on the internal driving degree of the student to obtain a primary evaluation result; determining a training target of the student according to a primary evaluation result; constructing an internal driving education model according to the basic information data, scene teaching and course teaching on the basis of the training target; and generating internal driving force growth data based on the internal driving force education model, and performing secondary evaluation on the internal driving force growth data to obtain a student internal driving force cultivation result. According to the method, the internal driving force education model is constructed, so that the internal driving force of the students is cultivated, and the internal driving force of the students is widely improved.

The invention discloses a crop growth period monitoring method and system based on an AI algorithm model, and belongs to the technical field of agricultural information, and the method specifically comprises the steps: collecting the weather, growth cycle and soil data of a target crop in a source place and an introduction place; shooting images of the target crop at different growth stages by using an unmanned aerial vehicle, and performing feature marking; by loading and training a target detection model, identifying a growth stage of the crop; inputting real-time invisible light wave band single-channel radiometric calibration image data into the target detection model to obtain growth state and growth period information of crops; combining the introduction environment and the basic growth data, utilizing a time sequence analysis algorithm to predict the growth period of the crops, analyzing the prediction result, comparing the difference between the origin and the introduction site, and adjusting agricultural management measures to adapt to the influence of environmental conditions and genetic factors on the growth period of the crops.

The invention relates to the field of seedling cultivation, in particular to a forestry seedling cultivation system and method.The forestry seedling cultivation method comprises the steps that seedling growth data of a plurality of nursery areas are collected to extract growth state characteristics of seedlings in the nursery areas; in combination with the growth state characteristics and the gap area uniformity of the adjacent nursery stocks, the growth quality characterization value of the nursery stocks in the nursery region is evaluated, and the nursery region is marked; in response to a marking result, analyzing growth interference characterization parameters aiming at the nursery stock according to soil state characteristics in the nursery region, and judging whether the nursery region accords with environment-related interference conditions or not in combination with the range of the diameter-height ratio of the corresponding nursery stock; and evaluating and analyzing the cultivation and growth conditions corresponding to the nursery stocks in the nursery area. According to the method, the growth quality of the seedlings is comprehensively evaluated, the overall seedling rate and the cultivation quality of the forestry seedlings are improved, the industrial development requirements are met, the seedling cultivation management accuracy is improved, and the management difficulty is reduced.

The invention belongs to the technical field of forestry engineering, and provides a castanopsis hystrix improved variety and eucalyptus mixed forest litter resource regulation and control method, which comprises the following steps: setting a quadrat region based on a castanopsis hystrix and eucalyptus mixed forest, and dividing an ecological conflict region; calculating a litter load threshold value of the ecological conflict area and dividing a high-load area; regulating and controlling litters in the high-load area, and performing demand analysis based on the regulated and controlled high-load area to obtain a decomposition demand coefficient; judging whether the high-load area is a strong decomposition area or not, performing secondary regulation and control on the strong decomposition area, obtaining a periodic decomposition rate and judging whether the periodic decomposition rate accords with expectation or not; if so, acquiring related growth data of the castanopsis hystrix in the regulation and control region twice, and performing growth analysis on the castanopsis hystrix to obtain an average growth index; evaluating the growth state of the castanopsis hystrix based on the average growth index; the method is beneficial for improving the pertinence and effectiveness of ecological management of the mixed forest.

The present invention discloses a method for analyzing the distance between a power transmission line and a tree line based on monocular vision, which belongs to the field of visual analysis and distance measurement technology. The method comprises obtaining the length of a preset transmission line section, marking the length as an inspection length, collecting relevant data of trees on both sides of the inspection length, marking the trees as preset trees, wherein the relevant data includes height data of the trees on both sides and distance data between the ends of the branches of each tree and the preset transmission line section; collecting growth changes of the preset trees in time periods based on the distance data, wherein the time periods are divided into different temperature sections, and a database is constructed based on the collected growth changes; the present invention performs monocular vision collection on the preset trees and analyzes the growth data of different temperature sections, thereby providing more reliable data support for the distance changes between the preset trees and the preset transmission line section in future time periods in addition to the monocular vision data, thereby reducing the potential risks between the tree branches and the preset transmission line section.

The invention discloses an environmental parameter prediction method based on GRU-LSTM-Attention, and the method comprises the following steps: 1, collecting experimental bacterium growth data, dividing a time period, and sorting the experimental bacterium growth data into a data set; 2, preprocessing the data set, and constructing a training data set and a test data set; step 3, a GRU-LSTM-A model is constructed; 4, a GRU-LSTM-A model is trained; 5, carrying out the performance evaluation of the trained GRU-LSTM-A model, and obtaining a prediction model; and step 6, predicting the test data set by using the prediction model to obtain environmental parameters. According to the environmental parameter prediction method based on GRU-LSTM-Attention, the problems that in an existing environmental parameter prediction technology, due to the fact that an RNN model or an LSTM model is singly used, the operation efficiency is low, and the prediction result is not accurate enough are solved.