867 results about "Linear regression" patented technology

The invention relates to the technical field of automatic medical treatment, provides an infusion process control scheme, and particularly relates to an infusion port risk prediction method and device and electronic equipment. According to the method, clinical-related medical data is constructed and screened through a multiple linear regression method to obtain clinical infusion risk-related data, and the clinical infusion risk-related data is coded and subjected to weight updating to obtain an input sequence; and performing risk prediction on the input sequence based on the prediction model to obtain a final risk prediction result.

The invention discloses a topology identification method based on intelligent measurement data of a low-voltage transformer area. According to the technical scheme, the topology identification method comprises the steps that 1, a master station issues a timing command to an intelligent measurement terminal and an electric meter; step 2, the master station issues a transformer area electric meter file to the intelligent measurement terminal, configures an acquisition task and issues the acquisition task to the intelligent measurement terminal; step 3, establishing an electric quantity sum linear regression model based on a Lasso algorithm by using daily frozen electric quantity data, and converting the identification problem of the transformer area user-transformer relation into regression coefficient solving; step 4, on the basis of user change relation identification, performing dimension reduction on historical voltage data by using a t-SNE algorithm to improve the identification efficiency of the model; and step 5, clustering the voltage data after dimension reduction by using agglomerated hierarchical clustering, identifying a user phase and a meter box to which the user belongs, and completing identification of a topological structure. According to the method, the power grid topology information of the transformer area can be automatically identified, the efficiency and the accuracy of topology identification are improved, and the method has far-reaching significance for construction of an intelligent power grid.

The invention relates to the technical field of power distribution network fault monitoring and forest fire early warning, and discloses a high-voltage line tree line fault ignition time early warning method and device and a medium, and the method comprises the steps: collecting electrical characteristic data and non-electrical characteristic data in a single-phase tree contact grounding fault process in real time; performing preprocessing and sliding window processing on the electrical characteristic data to screen out a key characteristic set; establishing a thermal-electric coupling model representing the exponential decay relation of the equivalent grounding resistance along with the rise of the tree body temperature; training a plurality of different types of data-driven models based on the key feature set to regressively predict the ignition time countdown; taking prediction results of the plurality of data driving models on the training set as input, and taking real ignition time as a target to train a linear regression model as a fusion device so as to obtain an ignition time countdown early warning model; and according to a final predicted value output by the ignition time countdown early-warning model, determining and outputting an early-warning level and a confidence coefficient. According to the invention, early warning can be realized hundreds of seconds before open fire appears.

The invention relates to the technical field of agricultural biological information, in particular to a rape phenotype-genotype correlation analysis system based on deep learning, which comprises a phenotype image acquisition module, an image preprocessing module, an image phenotype feature extraction module, a multi-dimensional phenotype integration module, a deep learning model training module and a correlation analysis module, the image phenotypic feature extraction module creatively fuses a differential geometry theory and a deep learning technology, maps rape phenotypic features from an Euclidean space to a Riemannian manifold space through a multi-scale differential manifold representation network, and calculates feature correlation based on a geodesic distance by using a manifold attention weighting mechanism. The multi-scale geodesic integration module is adopted to realize organic fusion of phenotypic characteristics of different scales, the association analysis module adopts a multi-level linear regression model to analyze key agronomic characters of the rape, an association relationship between phenotypes and genotypes is established, and the phenotypic characteristic extraction accuracy under a complex background is improved by 35.2% compared with that of a traditional method.

The invention discloses a calibration test method and system for a pressure sensor, and relates to the technical field of pressure sensor calibration. According to the method, original temperature and voltage data are collected in real time, ideal pressure is obtained through an ideal state equation, original output voltage is obtained through a preset relation model in combination with the temperature, then precise compensation of zero drift and sensitivity drift is completed in sequence through a multi-dimensional calibration coefficient, and further correction is performed through a linear regression relation polynomial, so that the accuracy of the zero drift and the sensitivity drift is improved. And finally, a detection pressure value is obtained. The output pressure value is closer to the real physical quantity, the application requirements of high-precision scenes such as medical equipment are met, and the calibration effect and the pressure measurement precision are improved.

The invention discloses a method for determining pesticides in water by liquid chromatography and tandem mass spectrometry, which comprises the following steps: acquiring retention time data of different pesticide compounds in a liquid chromatography system, recording retention behavior characteristics of each pesticide compound under optimized chromatographic conditions by adjusting the composition proportion of a mobile phase and a gradient elution program, and determining the retention behavior characteristics of each pesticide compound under optimized chromatographic conditions. Obtaining a standardized retention time spectrum library data set; establishing a quantitative relation model between molecular structure parameters and chromatographic retention time by adopting a multiple linear regression algorithm through key parameters such as molecular weight, polar surface area and lipid-water partition coefficient in the molecular structure parameter matrix, and obtaining a structure-retention correlation prediction model; and constructing a classification prediction model between molecular structure parameters and environmental durability by adopting a support vector machine algorithm through activity index data in the biological activity prediction result, and determining the environmental durability grade of the pesticide compound according to molecular stability parameters and degradation half-life characteristic values.

The invention relates to the technical field of direct-current large-current power supply testing, in particular to a direct-current large-current power supply testing method and device. According to the invention, temperature, voltage and current are synchronously acquired through multiple channels, and data validity is marked; after smoothing processing, calculating compensation resistance by using a two-item temperature compensation formula; according to current characteristics, an abnormal threshold value is dynamically adapted in a current increasing / stabilizing / decreasing period, and a normal / early warning / shutdown instruction is output in a grading manner; performing correlation analysis on temperature-resistance and voltage-current characteristics to identify hidden defects; the weighted linear regression predicts future temperature / resistance and gives an early warning; data are stored in a layered mode, visualized display is achieved, and a report containing A / B / C grade rating is automatically generated. The device comprises a corresponding functional unit and a processor-memory architecture, and is suitable for large-current testing of multi-scene electrical equipment.

The invention relates to a wind noise modeling method and system based on machine learning and application, and belongs to the field of marine acoustics and environmental noise modeling, and the method comprises the steps of data preprocessing and feature extraction, hybrid network model building, hybrid network model training and verification. According to the method, a hybrid network model of multiple linear regression and a multi-layer perceptron is constructed based on a known physical mechanism of wind noise and two main noise generation mechanisms of surface turbulence and bubble oscillation, a linear relation and a non-linear relation are modeled respectively, outputs of the two models are integrated through a weighted fusion strategy, and a multi-layer perceptron model is constructed. Smooth transition modeling from a low-wind-speed linear relation to a high-wind-speed nonlinear relation is achieved.

The invention discloses an intelligent compilation method for a feasibility research report fusing a big data model and artificial intelligence, relates to the technical field of information, and solves the technical problem that it is difficult to obtain quantized values from multiple aspects of market scale, project income and technology to analyze project feasibility. Comprising the following steps: collecting and preprocessing multi-source data; an ARIMA model is constructed to output a market scale prediction value, and partition assignment is carried out; processing corresponding structured data through XGBoost, analyzing time series data through Transform, predicting ROI and CBR through a full connection layer after feature fusion, and weighting to generate a comprehensive income evaluation index prediction value; calculating a technical equipment performance evaluation index and a technical improvement evaluation index; constructing a linear regression model, and outputting a comprehensive technology evaluation index prediction value; and calculating a score according to the market scale partition value, the comprehensive income evaluation index prediction value and the comprehensive technical evaluation index prediction value, comparing the score with a threshold value, and automatically generating an analysis report.

The invention discloses a comprehensive performance evaluation method for a bridge structure, and belongs to the technical field of bridge engineering, and the method comprises the steps: obtaining time domain data of deflection and displacement, and frequency domain data of cable force and vibration according to dynamic monitoring data; rejecting the temperature effect by adopting a linear regression model to obtain corrected time domain and frequency domain data; applying a Weibull extreme value distribution model to perform fitting analysis, extracting a dead load effect, and obtaining a structural response data set caused by dead load; calculating a score value and an allocation weight of a single position index on the bridge, calculating a uniform coefficient, and calculating a non-uniform coefficient based on the structural response data set; and multiplying the uniform coefficient by the non-uniform coefficient to obtain a comprehensive score of the monitoring index, and comparing the comprehensive score with a preset standard interval to obtain a performance evaluation result. According to the method, the temperature effect can be systematically eliminated, the dead load effect is extracted, and comprehensive and accurate evaluation of the health state of the bridge structure is realized by combining scoring and a weight distribution mechanism.

The invention discloses an AI-based emotional text speech conversion method and device. The method comprises the following steps: acquiring speech segments and text records from historical data of a user; performing noise reduction processing and feature extraction according to the voice segments and the text records to obtain voice features; inputting the voice features into a pre-constructed emotional tendency model, and outputting emotional tendency and emotional intensity; according to the emotional tendency and the emotional intensity, adjusting a tone weight, a speech speed and a volume to obtain a speech parameter; extracting new voice features according to the voice parameters to perform scene emotion label matching, and determining voice adjustment parameters through a linear regression model; according to the emotional tendency and the new voice features, generating an emotional type through a pre-established emotional intention classification model, and calculating a voice parameter weight in combination with a pre-established emotional mapping table; and according to the voice adjustment parameter and the voice parameter weight, performing language synthesis to generate personalized voice. According to the method, personalized expression can be accurately generated according to the scene.

The invention discloses a crude oil production, supply, storage and transportation system monitoring and port adaptability evaluation method and system, and belongs to the technical field of energy and logistics. The method comprises the steps of collecting multi-source data and performing standardization processing; constructing a three-level distributed storage framework to perform integration storage of crude oil production, supply, storage and transportation full-chain data; in the aspect of transportation safety, a berth utilization rate and reasonable trafficability dynamic measurement and calculation model containing ship type proportions, dynamic time parameters and random fluctuation factors is constructed for different types of wharfs, and adaptability grading evaluation is carried out; in the aspect of reserve safety, the reasonable reserve scale of the port is calculated by adopting a multi-factor linear regression model, and a transport connection convenience correction coefficient and an industrial cluster radiation capability parameter are introduced to obtain a reserve layout rationality index; monitoring and evaluation results are displayed through a visual system architecture, and dynamic tracking and decision support of a nationwide crude oil production, supply, storage and transportation system are achieved. The collaborative management level of the crude oil production, supply, storage and transportation system can be effectively improved.

The invention discloses an electric power intelligent monitoring system based on multi-device cooperative linkage and an operation and maintenance method, and belongs to the technical field of electric power intelligent monitoring and operation and maintenance. Radiation data and historical decay records are collected through a sensor, a linear regression algorithm is adopted to fit and establish an energy spectrum purity attenuation model, and the dose loss proportion is calculated in real time; carrying out dynamic compensation on beam intensity and irradiation time based on the proportion, fusing real-time dose monitoring data by utilizing a Kalman filtering algorithm, and optimizing a beam intensity value to suppress noise interference; the irradiation time is further adjusted according to whether the dose loss exceeds a threshold value or not, and a dynamic compensation combination is formed; the effectiveness of the combination in each stage of the source life cycle is verified through the simulation module, and the verified compensation parameters are updated to the control system in real time, so that closed-loop regulation and control are realized. According to the invention, the whole process of dose output in the irradiation process is accurately and stably controlled, and the consistency and reliability of the irradiation process are remarkably improved.

The invention discloses a coal mine porosity calculation method for improving sequential Gaussian simulation under multi-source data fusion. On the basis of logging data preprocessing, logging curves of natural gamma, density, three-lateral resistivity and the like are synthesized, the high-organic-matter stratum is accurately discriminated, a volume physical model is constructed to explain the porosity of the rock stratum, a linear regression model is constructed to explain the porosity of the high-organic-matter stratum, and the porosity of the high-organic-matter stratum is accurately discriminated. After the two results are spliced and combined, abnormal sudden change points on a curve are eliminated through a curve smoothing algorithm, then the porosity curve, rock core data and the water temperature experiment result are normalized and mapped into a geological grid model, variation function analysis and parameter selection work are carried out, improved sequential Gaussian simulation and post-processing are carried out on the basis, and the geological grid model is obtained. And the reliability of the porosity attribute model is improved.

The invention relates to a rigidity measuring method for a servo joint of a robot, which comprises the following steps of: establishing a measuring coordinate system, carrying out coincidence calibration on a base coordinate system of the robot and a world coordinate system, and initializing the pose of the robot; target ball arrangement and non-load reference point acquisition are carried out, a target ball is rigidly bonded to a connecting rod close to a joint shaft to be detected, and target ball coordinate points in a non-load state are acquired and fitted to obtain a related reference; graded tail end load application and six-dimensional force synchronous acquisition are carried out, loading is carried out step by step according to incremental loads, and six-dimensional force data are recorded; robot joint torque is calculated, a kinematic model is called to obtain a force Jacobian matrix, and joint torque increment is calculated; then target ball displacement is collected under the load, and joint deformation is calculated; fitting the joint basic stiffness k through linear regression, evaluating a fitting effect by using a decision coefficient, and outputting the joint basic stiffness. The method can accurately measure the rigidity of the servo joint of the robot, and provides a reliable basis for evaluating the performance of the robot.

The invention discloses a traditional folk song singing voice acoustics and physiology multi-modal analysis method and system, and the method comprises the steps: carrying out the synchronous calibration of a voice recording device, a voice monitoring device and a respiration sensor through a unified clock source, obtaining a multi-modal signal data flow, and obtaining a time-aligned original data set; according to the original data set subjected to time alignment, Fourier transform is adopted to process the voice signal part, acoustic feature vectors are determined, and a physiological signal sequence is extracted from the original data set subjected to time alignment; if the acoustic feature vectors are matched with the timestamps of the physiological signal sequences, Pearson coefficients between the acoustic feature vectors and the timestamps are calculated through correlation analysis, and quantitative correlation strength is judged; for the part of which the quantitative correlation strength is higher than a fourth preset threshold value, fitting a mapping relationship between the acoustic characteristics and the physiological mechanism by adopting a linear regression model to obtain a parameterized singing technique model; and according to the parameterized singing technique model, analyzing the relationship between traditional singing voice acoustics and physiological multiple modes.

The invention belongs to the field of medical treatment, and relates to a dynamic birth process progress automatic monitoring method, which comprises the following steps: S1, continuously detecting and screening image frames containing intact pubis symphysis and fetal head skull halo; s2, performing standard section screening on the continuous ultrasonic image frames by using a USPNet model based on a Swin Transform architecture; s3, automatically marking a birth process progress angle and a fetal head exposure position through an AI segmentation model, and calculating a fetal head exposure value; s4, driving the three-dimensional fetal head and pelvis model based on the real-time progress angle and fetal head exposure position data, and dynamically rendering and updating a relation graph of the three-dimensional fetal head and pelvis; s5, dynamically displaying the three-dimensional relational graph of the fetal head and the pelvis; and S6, establishing a fetal head descending and uterine contraction signal time sequence curve, and estimating the descending amplitude of the fetal head in the future set time based on a linear regression or curve fitting prediction algorithm. According to the application, continuous, dynamic and accurate monitoring and visual display in the uterine contraction period can be realized, doctors are assisted in scientific decision making, and the delivery management level is improved.

The embodiment of the invention provides a mutual inductor state monitoring system and method for an intelligent laboratory standard electric energy meter, and relates to the technical field of electric energy meters. The mutual inductor state monitoring system comprises a main control unit and a data acquisition unit, wherein the data acquisition unit is used for acquiring a signal measurement value and a working temperature of a secondary side of a mutual inductor; the main control unit is configured to execute the following calculation modules: a linear regression calculation module used for performing regression analysis on the signal measurement value acquired by the data acquisition unit so as to estimate a signal true value; the temperature compensation module is used for removing the influence of the working temperature on the real value of the signal based on the standard value matrix; and the state judgment module is used for calculating an abnormal score of the signal true value after the influence of the working temperature is removed based on the isolated tree model, and judging the misalignment state of the mutual inductor based on the abnormal score and a preset misalignment boundary threshold, so that the monitoring precision of the mutual inductor in the standard electric energy meter can be improved.

The method comprises receiving one or more bounds for each coefficient of independent variables. Further, a steepness value is selected for controlling the velocity of weight updates for each coefficient. Subsequently, an update vector with a length equal to the number of independent variables may be created. Further, the method may comprise iterating until convergence. Each iteration may include computing a gradient for each independent variable based on the gradient, updating values of each coefficient based on the computed gradient, the computed multiplier, a learning rate, and the update vector. Further, the method involves monitoring the CLR optimization process for convergence based on whether a change in the value of the cost function is below a predefined threshold or a maximum number of iterations is reached. Further, the optimized CLR model may be utilized for at least one application within Revenue Growth Management (RGM).

The invention relates to a data-driven power distribution network topology and line parameter identification method, which comprises the following steps of: 1, constructing a linear power flow model, obtaining an initial estimation value of a network parameter of a power distribution network based on an initial identification method of linear regression, and further obtaining an initial identification result of a power distribution network admittance matrix; and step 2, introducing the voltage phase angle change between nodes, and correcting the initial identification result of the admittance matrix of the power distribution network obtained based on the step 1 through an accurate power flow iteration model to obtain a final topology identification result. According to the method, the limitation of a traditional method is broken through, and topology and parameter estimation completely without phase angle measurement is realized through a strategy of combining linear regression and iterative calculation under the condition that the voltage phase angle and topological information do not need to be known in advance.

The invention discloses a multi-model fusion space-time water quality prediction method. The method comprises the following steps: S1, collecting multi-source time sequence data of a key monitoring unit of a sewage plant; s2, cleaning the collected original time sequence data; s3, extracting long and short term dependence characteristics of the water quality indexes by adopting an LSTM module; a Transform module is adopted to extract global attention features; an XGBoost model is adopted to extract nonlinear regression features and prediction results; s4, constructing a plurality of model fusion strategies, and performing integrated evaluation on prediction results of the prediction models, including static weighted fusion, adaptive attention weight fusion and integrated learner fusion; and S5, performing performance evaluation on each fusion strategy by adopting the same evaluation index, automatically selecting the fusion strategy which shows the optimal on the verification set, and performing prediction output on the key water quality index at the future moment. The prediction precision of the key water quality indexes and the intelligent level of system operation are improved.

The invention relates to the technical field of dam safety in water conservancy projects, in particular to a dam osmotic pressure monitoring and analyzing method based on sensing data. According to the method, the arrangement density of the monitoring points in the key structure area of the dam body is determined, the seepage pressure severity levels of the historical seepage pressure areas are analyzed to determine the arrangement density of the corresponding monitoring points, and the monitoring points are arranged on the surface of the dam body in combination with the boundary contours of the key structure area and the historical seepage pressure areas; initial osmotic pressure and ring influence factors of each monitoring point are collected, and environment similar points are clustered to form monitoring sub-regions; constructing an osmotic pressure correction compensation model of linear regression, and eliminating environmental influence to output an actual osmotic pressure value; and comparing the actual osmotic pressure of the monitoring points in the same monitoring sub-region to determine abnormal monitoring points, and determining and feeding back the positions of the osmotic pressure monitoring points in combination with a historical osmotic pressure time sequence trend. According to the method, the layout of monitoring points can be optimized, data errors are reduced, misjudgment and missed judgment of abnormal recognition are reduced, and the reliability of dam safety monitoring is guaranteed.

The invention relates to the technical field of agricultural remote sensing and precise irrigation, in particular to a method and equipment for inverting leaf water potential of grapes before dawn based on multispectral remote sensing, and the method comprises the steps: measuring leaf water potential of representative grape plants in each planting test plot before dawn in different growth periods, and collecting canopy multispectral images in the noon period; preprocessing the remote sensing original image data; calculating a vegetation index; on the basis of the model fitting data set and the training feature set, constructing basic prediction models in different growth periods by adopting a unary linear regression algorithm, a multiple linear regression algorithm and a partial least square regression algorithm respectively; determining a PLSR inversion model according to the decision coefficient and the root-mean-square error; and carrying out pre-dawn leaf water potential prediction through the PLSR inversion model. Through the method, high-precision inversion of the water potential of the leaves before dawn based on the noon canopy multispectral data can be realized, and reliable technical support is provided for accurate management of vineyard moisture and intelligent irrigation.

The invention relates to the technical field of micro-grid load prediction, and discloses a multi-factor efficient fusion micro-grid load short-term prediction model optimization method. According to the method, micro-grid load historical data and multi-source factor data including information such as meteorology and electricity price are collected, and one-hot coding and normalization processing are carried out. Load data is decomposed into high-frequency and low-frequency components by using a variational mode decomposition technology, and prediction is carried out by respectively adopting a long-short-term memory network integrated with an attention mechanism and a linear regression model. And by calculating the difference between the source domain data and the target domain data in the model hidden layer feature space, the domain adaptive capacity of the model is enhanced, and accurate short-term prediction of the microgrid load is realized.

The invention relates to the technical field of smart power grids, in particular to a distributed photovoltaic-oriented smart power grid load prediction method and system, and the method comprises the steps: collecting the total load data, distributed photovoltaic power generation power data and meteorological data of a target region and an adjacent region, and carrying out the preprocessing; extracting spatial correlation characteristics and time sequence evolution characteristics of the photovoltaic power generation power data, and fusing to generate a low-dimension spatial-temporal characteristic matrix; splicing the spatial-temporal feature matrix with the total load data and the meteorological data to form a multi-modal fusion feature vector, and inputting the multi-modal fusion feature vector into a machine learning prediction model to generate an initial load prediction result; and introducing real-time updated ultra-short-term weather forecast data, carrying out dynamic correction on an initial prediction result through a linear regression error correction model, and outputting a final load prediction value. According to the method, the space-time precision and the real-time adaptive capacity of load prediction in a distributed photovoltaic scene can be remarkably improved, and the method has good practicability and engineering generalizability.

The invention discloses a virtual power plant load prediction method and system based on multi-model combination, and relates to the technical field of virtual power plants, and the method comprises the steps: carrying out the preprocessing of obtained multi-source heterogeneous data; clustering the preprocessed historical load data, and dividing the historical load data into at least two load feature groups according to different seasons or power utilization modes; respectively utilizing the trained linear regression model, the trained LSTM neural network model and the trained support vector machine model to correspondingly obtain three preliminary prediction results; determining a weight coefficient of the model by using a genetic algorithm to obtain a combined prediction result; and carrying out reverse normalization processing on the combined prediction result, and outputting a load prediction value of the virtual power plant in a future preset time period. The problem that a traditional single model is poor in adaptability to complex relevance, dynamic change and external interference is solved.

The invention relates to a method for establishing an SIL dynamic evaluation model based on Weibull distribution, and belongs to the technical field of safety protection systems. The method comprises the following steps: (1) establishing a dynamic cumulative failure probability function model based on Weibull distribution according to obtained failure data through a reliability test; (2) calculating Weibull distribution parameters by using a median rank method, a linear regression fitting method and a least square method; (3) drawing a function model image by using MATLAB software; and (4) in combination with a fault tree method, establishing an SIL dynamic evaluation model based on Weibull distribution. According to the SIL dynamic evaluation model for the final executive component of the safety instrument system based on Weibull distribution, the dynamic change trend of the final executive component PFDavg along with time obtained by the model basically accords with the reality, the accuracy of SIL evaluation is improved, and scientific method support is provided for design personnel of the safety instrument system.

The invention relates to the technical field of power system load prediction, and discloses a short-term power load prediction method considering load volatility and double-frequency modeling, which comprises the following steps: acquiring power load historical data of a target area in a preset historical period and same-period influence factor data, performing time mark alignment and screening on the data, and obtaining a power load prediction result; obtaining a processed data set, decomposing a load sequence in the processed data set, performing feature enhancement on the input data through a feature extraction model, and outputting an enhanced feature sequence; the overall trend of the load is accurately anchored through a linear regression model, and the dependency relationship of a residual error sequence is fully mined through a time sequence prediction model; and a final load prediction result is obtained through double-channel output superposition reconstruction. The method has the advantages that multi-dimensional factors such as weather and date are comprehensively considered, and the precision and robustness of short-term power load prediction are remarkably improved by using the STL load decomposition technology and the double-frequency model.

A processing method of a display screen brightness compensation includes: obtaining a sample data set of the display screen, a real environment temperature and a real image average grayscale value corresponding to the real environment temperature, the sample data set including an environment temperature sample set, a image average grayscale sample set corresponding to the environment temperature sample set and a simulated temperature sample set; performing a linear regression processing on the sample data set to generate a regression coefficient set; processing the real environment temperature and the real image average grayscale value according to the regression coefficient set to generate a temperature to be compensated of the display screen; and determining a brightness compensation parameter according to the temperature to be compensated, and performing a brightness compensation processing on the display screen according to the brightness compensation parameter.

The invention discloses a composite material storage life prediction method based on variable transformation and a genetic algorithm, and relates to the technical field of composite material performance evaluation and life prediction, and the method comprises the steps: obtaining performance degradation data at different temperatures through an accelerated aging test, and building a semi-empirical performance degradation model; specific variable transformation is introduced, a nonlinear optimization problem is converted into a linear regression problem, partial model parameters are rapidly solved by adopting a least square method, and meanwhile, key nonlinear parameters are globally optimized by using a genetic algorithm, so that the defect that a traditional method is easy to fall into local optimum is overcome. And establishing a temperature-service life relationship by combining with an archenius model, and realizing storage life prediction at normal temperature. The method also integrates sensitivity analysis and uncertainty quantification functions, and provides reliability evaluation for a prediction result. The method fully considers the influence of the environment temperature, meets the requirement of efficient prediction, and improves the precision and stability of life prediction.