57 results about "Covariate" patented technology

The invention discloses an electric power total-factor unified load prediction method and system based on a time sequence large model, and relates to the technical field of machine learning, and the method comprises the steps: triggering a prediction process through a timed task, and obtaining historical load data and external covariable data; constructing a time sequence sample required by training and prediction according to a time window, and performing batch pulling and processing according to a fixed number of days when the data size exceeds a single-batch threshold value; performing data standardization and data cleaning on the data, and generating structured time sequence input; loading a time sequence large model as a unified prediction engine, and inputting a load sequence and an external covariable into a covariable fusion component for collaborative modeling; and outputting fine-grained prediction results of the electric power elements in the target time period at the same time under a single model framework, and post-processing and storing the prediction results. Through the technical scheme of the invention, the model fragmentation and maintenance cost is reduced, the generalization and stability are improved, and the operation stability and the engineering availability are improved.

A method and an apparatus for forecasting a future time target variate, a computer device, and a storage medium are disclosed. Specifically, a method for forecasting a future time target variate is disclosed. The method includes: dividing a history target variate and covariates of target variates into a plurality of patches channel-wise, and performing linear embedding, wherein the covariates of target variates comprise a covariate of the history target variate and a covariate of the future time target variate; performing position embedding on a plurality of tokens and a learnable forecasting token; wherein the plurality of tokens are linear embedded from the plurality of patches; inputting the plurality of tokens and the learnable forecasting token into an encoder; and forecasting and outputting the future time target variate based on the encoder and the learnable forecasting token. According to the foregoing manner, in a case of zero-shot training, the future time target variate can be accurately forecast through limited history target variates and covariate information of the target variates, thereby having a large inspiration effect on the field of basic models in time series forecasting.

The invention discloses a wind power prediction covariable offset adaptive method based on a conditional energy model. According to the method, the wind power generation power prediction model is constructed by utilizing the gated cycle unit network, and offline training of the wind power generation power prediction model is completed by adopting a sample weighting mechanism, so that the robustness of the wind power generation power prediction model to distribution change is enhanced. A conditional de-noising score matching strategy is adopted to learn the distribution difference of data in a training stage and a prediction stage through a conditional energy model, and a sample weight used for measuring the covariable offset degree is obtained based on the model. Data samples flowing in real time are stored in a replay buffer area, incremental learning is carried out on a condition energy model and a wind power generation prediction model through an online updating mechanism, and the prediction performance is kept stable. The method can effectively improve the power prediction precision and operation scheduling capability of the wind power plant under complex meteorological conditions, and has good engineering practical value and deployment flexibility.

The invention belongs to the field of water environment monitoring and water quality prediction, and particularly discloses a causal discovery and deep learning coupled water quality prediction method and system, and the method comprises the steps: carrying out the causal screening of dynamic covariables through employing a causal discovery algorithm based on a neural network, and recognizing causal dynamic covariables; and inputting the causal dynamic covariable, the multi-scale water quality index characteristics and the static covariable data into a trained probability time sequence prediction model, and outputting a probabilistic prediction result of the water quality index concentration of the target water area in a plurality of time steps in the future. According to the method, pseudo-correlation variables are eliminated, the non-stationarity is processed through multi-scale analysis, purer and richer multi-scale information is provided for the probability time sequence prediction model, and the accuracy of point prediction is remarkably improved.

The application discloses a data classification prediction method based on covariate and semantic offset and related devices, relates to the technical field of graph data machine learning and data classification prediction, and comprises the following steps: obtaining labeled in-distribution graph data and unlabeled open-world graph data, and uniformly performing standardization processing to obtain standardized graph data; constructing a graph learning model, and outputting a logit based on the graph learning model; calculating an energy score based on the logit, and mapping to obtain a semantic consistency estimation value; constructing a total loss function comprising an in-distribution classification loss, an in-distribution energy upper bound constraint and an open-world energy lower bound constraint, and training a model; obtaining open-world graph data to be measured, inputting standardized graph data of the open-world graph data to be measured into the trained graph learning model, calculating an energy score based on an output logit, combining a preset energy threshold, and performing classification prediction or rejection prediction. The application can accurately distinguish covariate offset and semantic offset, and improves the accuracy of data classification prediction.

The application relates to a multi-source covariate irrigation load short-period prediction method, a device and a medium, and relates to the fields of power system load prediction and intelligent analysis of agricultural irrigation energy. The application is to solve the problems that the existing irrigation load short-period prediction method has limited modeling capability for exogenous factors, cannot effectively utilize user difference information, and is prone to lag or drift in prediction. The application performs correlation evaluation on candidate multi-source covariates and historical global irrigation load, and then selects effective covariates according to the evaluation results; extracts a user feature vector based on the effective covariates; constructs a historical input sequence by combining the historical global irrigation load and the effective covariates, and constructs a future known covariate by combining the known future prior information in the prediction interval; constructs input data by combining the user feature vector, the historical input sequence and the future known covariate; inputs the input data into a load prediction model adopting a TimeXer framework, and outputs an irrigation load prediction result in the prediction interval.

The invention belongs to the field of intelligent monitoring, and discloses an active and passive fusion clean room environment intelligent monitoring method, which constructs an active and passive fusion monitoring system based on covariable statistics. A two-parameter variation function is established by introducing real-time wind field difference as a key covariable, and the defect that a traditional model ignores flow field dynamics is overcome; a comprehensive blind area guiding index is constructed by using a deviation between a basic information field generated by a fixed sensor and a correction information field formed by robot fusion data and combining an inherent coverage risk and a joint space disparity degree, and an absolute monitoring blind area is accurately identified; therefore, the control mode of fixed route inspection to blind area driving type active detection is converted. According to the method, on the premise that the hardware cost is not increased, the full-domain monitoring coverage rate, precision and sudden pollution response speed of the clean room are remarkably improved.

Various embodiments can reduce variance in a target metric (e.g., experiment outcome). Embodiments can use historical pre-experiment outcomes to predict a metric (forecasts) that is expected for a future measurement time. The forecasts can then be used as a covariate to reduce the variance of the target metric. When predicting forecasts, various embodiments can use an automation pipeline that can generate better and quicker forecasts. When there are multiple covariates that may be considered to reduce variance in the target metric, various embodiments can use a closed from solution for determining optimal coefficient of each covariate.

The invention discloses a visibility time sequence prediction method based on a DeepAR model and a weather prediction field, mainly relates to the technical field of visibility time sequence prediction, and is used for solving the problems that in an existing scheme, weather prediction field feature fusion is insufficient, only a single value is output, and weather prediction field features cannot be spliced by an existing DeepAR model. Comprising the steps of generating a comprehensive feature vector according to a feature result, historical visibility time sequence observation data and a preset additional covariable, inputting the comprehensive feature vector into a hidden layer of an initial DeepAR model, and completing construction of a future meteorological field extraction module in the DeepAR model; a DeepAR model with a future meteorological field extraction module is trained and optimized based on historical visibility time sequence observation data, and a trained DeepAR model is obtained; and performing visibility time sequence prediction by using the trained DeepAR model.

The invention provides a dynamic Copula extreme wave energy prediction method based on wind speed covariant full-link driving. The dynamic Copula extreme wave energy prediction method is applied to safety evaluation of ocean engineering and a wave energy power generation system. Firstly, B-spline-logarithmic normal edge distribution with wind speed as a covariable is constructed through data preprocessing, and dynamic changes of wave height and wave period parameters along with the wind speed are described; secondly, selecting an optimal Copula function, coupling a Copula parameter with the wind speed through Logistic mapping, and establishing a dynamic dependency structure model; then, dividing a low wind speed interval, a medium wind speed interval and a high wind speed interval, and verifying the fitting precision of the model in each interval and extreme wave energy over-limit probability prediction performance; and finally, integrating edge distribution and dynamic Copula, and constructing a wind speed driven wave element joint probability prediction model. Compared with a traditional static model, the method can accurately capture the regulation and control effect of the wind speed on the wave height-wave period dependency relationship, remarkably improves the prediction accuracy of an extreme wave energy event, and provides reliable support for ocean engineering safety and wave energy power station operation.

An estimation device according to one aspect of the present disclosure comprises: an input unit that inputs observation data including a covariate in a predetermined space and an occurrence position of an event observed in the space, and kernel function data related to a kernel function used in a Gaussian process; and an estimation unit that estimates, on the basis of the observation data and the kernel function data, a posterior prediction distribution followed by an intensity function that outputs an occurrence probability of the event with the covariate as an input. The estimation unit approximates a theoretical lower limit value of a peripheral likelihood for the observation data as a function of a hyperparameter of the Gaussian process and maximizes the theoretical lower limit value using a steepest descent method, to thereby estimate the posterior prediction distribution followed by the intensity function.

The invention provides a mixed data regression model-based pm2.5 influence factor analysis method and system, and the method comprises the steps: building a mixed data regression model of which covariables are component data and numerical data and dependent variables are functional data: the mixed data regression model is a pm2.5 concentration curve of a city, is the functional data, is the proportion of first yield, second yield and third yield of the city, and is called the proportion of third yield for short; the data is component data, logarithm of per capita GDP, average temperature and numerical data, is a to-be-estimated component type coefficient changing along with time, is distributed to a component type covariable at any moment, is a to-be-estimated function type coefficient and is a function type residual error; obtaining robust M-estimation based on equidistant logarithmic ratio transformation, functional basis expansion and an iterative reweighted least square method; and according to the estimated values, analyzing the influence of the three-yield ratio, the per capita GDP and the average temperature of each city on the pm 2.5. The method can be used for analyzing the pm 2.5 influence factors.

The embodiment of the invention provides a soil bacterial community structure evaluation method and system based on an intermediate infrared spectrum transfer function and application, and relates to the technical field of soil microbial ecology. The method comprises the following steps: S1, collecting a soil sample; s2, acquiring a mid-infrared diffuse reflection spectrum of the soil sample to be detected; s3, acquiring an environment covariant corresponding to the soil sample, wherein the environment covariant comprises at least one of soil physicochemical properties, climate data and topographic parameters; s4, splicing the mid-infrared diffuse reflection spectrum obtained in the step S2 and the environment covariable obtained in the step S3 to form a fusion feature vector as an input feature set; and S5, inputting the input feature set into a pre-trained nonlinear machine learning regression model, and outputting the relative abundance and / or OTU richness of at least one bacteriophage in the soil sample. According to the method, a technical route of combining a mid-infrared (MIR) spectrum and a spectral transfer function (STF) is adopted and is used for predicting the soil bacterial community structure.

The application discloses a multi-time-scale power consumption dynamic analysis method, device and equipment, relates to the technical field of power load analysis, and comprises the following steps: acquiring a historical daily power consumption sequence and corresponding multi-source covariants; decomposing the historical daily power consumption sequence to obtain a trend component, a seasonal component and a residual component; constructing a correction operator to perform primary correction on the residual component; quantifying a lag effect through a kernel function based on industry operation capacity, and reconstructing the trend component; constructing an impact variable based on a period application capacity to perform secondary correction on the primary corrected residual component; and fusing prediction results of a first prediction model and a second prediction model to obtain a full-society power consumption prediction value, and mapping the full-society power consumption prediction value into power supply and power sale business indexes. The application realizes accurate quantification of industrial structural changes and collaborative analysis of multi-time-scale influencing factors, and significantly improves the accuracy of medium-and short-term power consumption prediction and the degree of connection with power grid business.

This invention discloses a covariate-enhanced time-series large-scale model for electricity price forecasting, comprising: S1: acquiring and preprocessing multi-source time-series data, the data including at least historical day-ahead electricity price sequences, historical day-ahead dispatch disclosure data sequences, and historical meteorological data sequences, and acquiring the dispatch disclosure and / or weather forecast data of the next day as future covariates; S2: constructing a backbone network based on a general time-series pre-trained model, and on the basis of the backbone network, integrating a historical covariate fusion module, a future covariate injection module, and a principal-covariate nonlinear interaction module in a pluggable manner to form a covariate-enhanced time-series large-scale model; S3: the forecasting phase; and S4: evaluating the forecast results. This invention can improve forecasting performance and operational indicators by structurally fusing historical dispatch disclosures, meteorological data, and electricity load, and reasonably introducing known future covariates, while ensuring causal constraints.

The application relates to a sample dynamic supplement method, system, device and medium based on environmental covariates. The method calculates the similarity between unknown points and the remaining sample point set and quantifies the uncertainty, then constructs a hierarchical sampling framework according to the uncertainty mean, dynamically allocates the supplement sample points, and finally combines the supplement points with the original sample set to form an updated sample set. Based on the full use of historical sample data, the method can significantly improve the representativeness and statistical balance of the updated sample set in geographical space, so that when compared with the traditional hierarchical sampling method, the map product precision evaluation result can obtain a better and more accurate performance, and the evaluation deviation problem caused by the non-representativeness of the sample set can be effectively overcome.

The application provides a kind of bayesian disease risk high-precision mapping method and system, it is related to spatial statistical analysis technical field.The application collects multiple source covariates, constructs integrated covariate sub-model and generates out-of-sample prediction and in-sample prediction;establishes bayesian regional spatio-temporal hierarchical correlation model, takes sub-model prediction result as covariate, expresses prevalence rate as covariate linear combination by logit function, residual error uses three-dimensional Gaussian process and utilizes random partial differential equation approximation;high-resolution grid risk map is generated by posterior sampling, and is population-weighted to administrative unit.The application can realize multi-factor driven disease risk high-precision spatio-temporal estimation and mapping, and is suitable for public health monitoring, disease early warning and precise intervention.

The invention discloses a high-resolution extreme rainfall intensity prediction method and device, electronic equipment and a medium, and relates to the technical field of meteorological water temperature and water resource engineering, and the method comprises the steps: obtaining original data and observation data, and carrying out the time index and spatial index unification processing of the original data and the observation data; inputting the original data and the corresponding covariables into a trained random forest model to obtain conditional distribution functions under different quantile levels; the random forest model is obtained by adopting a quantile random forest method to model a training sample set formed by combining the observation data and covariables; the covariables represent climate background features; monotone mapping is carried out on the original data based on the conditional distribution function, and a corrected rainfall sequence is obtained; wherein systematic deviation correction is carried out on medium and low intensity rainfall below a preset quantile threshold value; carrying out tail distribution correction on the high-intensity rainfall above a preset quantile threshold value; and carrying out extreme value modeling based on the corrected rainfall sequence, and outputting extreme rainfall intensity curves of different recurrence periods in the future. By adopting the method, the accuracy and temporal-spatial resolution of extreme rainfall event prediction can be improved.

This invention relates to a runoff prediction method based on a local attention enhancement model, belonging to the field of time series prediction. The method includes: acquiring and preprocessing data, inputting the processed data into a trained prediction model to obtain the predicted runoff sequence; wherein the runoff prediction model includes a variable selection module, a local information enhancement module, an attention module, and a model extraction module. This invention's prediction model considers the long runoff cycle, irregular trends, and varying degrees of influence of different covariates on the results. It utilizes a variable selection module to weight covariates, a local information enhancement module to capture local information of the runoff sequence, enabling data at individual time points to obtain short-term trend characteristics, and a self-attention module to obtain similarity and attention information between trends. Furthermore, by appropriately setting the depth of the encoder and decoder and the model extraction module, more accurate predictions are achieved with the same available memory.

The invention discloses a cross-regional settlement traffic prediction and deviation management and control method and system, and the method comprises the steps: collecting bandwidth settlement traffic time sequence data and external covariable data of a target region, and carrying out the preprocessing; constructing a double-flow depth time sequence prediction model, and predicting the settlement flow based on the preprocessing result and the double-flow depth time sequence prediction model to obtain predicted settlement flow; calculating a real-time deviation index of the actual settlement flow and the predicted settlement flow; monitoring a real-time deviation index, and performing layered drilling analysis under the condition that the real-time deviation index exceeds a dynamic threshold value and an early warning condition is met to obtain a traceability result; and executing a traffic governance strategy based on the traceability result, and performing iterative calibration on the double-flow depth time sequence prediction model by taking the governed traffic feature data as a correction sample to form a continuously self-optimized closed loop. According to the invention, high-precision short-term prediction and deviation pre-management and control of settlement flow can be realized.

The invention discloses a multi-element type soil nutrient prediction method coupling multi-source covariables and an in-situ spectrum. The method comprises the following steps: step 1, soil in-situ hyperspectral measurement and data acquisition; 2, collecting and preparing a soil sample; step 3, acquiring covariable data; step 4, acquiring soil nutrient data; step 5, in-situ spectrum data pretreatment: carrying out pretreatment operation on the soil in-situ spectrum; step 6, detecting and eliminating abnormal values of the spectral data; step 7, feature dimension reduction and screening; 8, constructing a multi-target regression prediction model of the soil nutrient content; step 9, model prediction performance evaluation and comparison; and step 10, quantifying the contribution of the multi-source covariable in model prediction. The method provided by the invention overcomes the defects of large prediction deviation and weak generalization ability caused by multi-element soil and multi-target nutrient modeling in the prior art, and can realize efficient, lossless and accurate prediction of the total nutrient content and the effective nutrient content of the multi-element soil under the provincial scale.

This invention provides a method and equipment for park operation decision analysis, relating to the field of smart park operation management technology. The invention constructs a candidate solution scenario library through structured definitions of covariates and intervention variables, covering various operational intervention scenarios to be evaluated; relying on generative causal Bayesian networks to integrate causal inference and generative modeling capabilities, it achieves reliable causal prediction of counterfactual solutions without historical observation data; it outputs predicted effect values ​​and uncertainty indicators to quantify prediction credibility and decision risk; in high-uncertainty hypothetical scenarios, through proactive surveys and incremental updates, it achieves effect prediction of counterfactual and hypothetical solutions to assist decision analysis, reduce investment risk and decision costs, and improve the robustness and efficiency of operational decisions. This invention addresses the problem of unreliable prediction and difficulty in proactive decision-making for counterfactual and hypothetical solutions without historical data in park operations, achieving a closed-loop, end-to-end park operation decision analysis.

This invention discloses a method, apparatus, device, and medium for power load forecasting, relating to the field of power load forecasting. The method includes: preprocessing and embedding features into input data containing load sequences, time covariates, resolution identifiers, and subject identity identifiers; projecting various features onto the same hidden layer dimension and concatenating them; performing frequency domain transformation and adaptive gated filtering on the concatenated features, restoring the time domain, and capturing the correlation between variables through encoding; performing preliminary prediction on the encoded features; substituting the prediction results into a quadratic programming problem containing grid ramp rate constraints; obtaining a corrected result satisfying the constraints through a differentiable optimization layer; calculating the gradient based on the implicit function theorem and performing backpropagation; and inverse normalizing the corrected result; constructing a prediction interval based on the output quantile prediction values; calculating the coverage error; and updating the scaling factor to adjust the width of the prediction interval. This significantly improves the accuracy, engineering applicability, and operational stability of the prediction.

The application discloses a method, device and medium for analyzing the dependent structure of interval quantile, and relates to the technical field of computers.The method comprises the following steps: determining input data corresponding to a task scene, and determining a multi-dimensional covariate and a response variable according to the input data; dividing the response variable to obtain a plurality of sub-intervals; determining a projection correlation coefficient according to the multi-dimensional covariate and the response variable, wherein the projection correlation coefficient comprises a projection bias and a quantile projection correlation coefficient; determining the projection bias according to the projection correlation coefficient, determining a test statistic according to the projection bias, comparing the test statistic with a pre-set threshold, and determining data dependency. The application can adapt to multiple scenes because of the diverse input data sources; the response variable can be flexibly divided to obtain multiple sub-intervals; the projection correlation coefficient is determined by the multi-dimensional variable, which is scientific and reasonable; the test statistic is constructed by the projection bias and compared with the threshold, which can efficiently determine the data dependency, avoid complex approximation, and reduce the calculation cost.

The embodiment of the invention provides a data processing method and related equipment. The method comprises the following steps: acquiring a covariable of each first observation unit in a first unit group and a covariable of each second observation unit in a second unit group; the first observation unit is an observation unit which is subjected to intervention operation, and the second observation unit is an observation unit which is not subjected to intervention operation; the covariable comprises observation index data of the observation unit before intervention operation; performing prediction according to the observation index data of each first observation unit before intervention operation to obtain a prediction index result of each first observation unit; performing prediction according to the observation index data of each second observation unit before intervention operation to obtain a prediction index result of each second observation unit; according to the prediction index result of the first observation unit and the prediction index result of the second observation unit, the first observation unit and the second observation unit are matched, and matching between the observation units can be efficiently carried out.

The invention provides a contrast analysis-based rain and snow enhancement benefit evaluation method, system and device, and a medium. The method comprises the steps of obtaining a first data set and a second data set of an influence area, obtaining a third data set of a contrast area, and obtaining a meteorological data set; determining an index mean difference based on the first data set and the second data set; determining a standardized effect quantity based on the second data set and the third data set; after a covariable effect value is determined through covariable analysis, the standardized effect quantity is updated to be a target effect quantity, a comprehensive contribution index is determined based on the index mean value difference of all the ecological indexes and the target effect quantity, and an ecological benefit evaluation result is determined. According to the technical scheme of the embodiment of the invention, multi-dimensional analysis is realized through a plurality of ecological indexes, a longitudinal analysis result of rain and snow enhancement is represented through the index mean value difference, a transverse analysis result is represented through the target effect quantity, and covariable analysis is introduced, so that the comprehensive contribution index can eliminate climate fluctuation, and the evaluation accuracy of ecological benefits is improved.

The invention belongs to the technical field of airport capacity prediction, and particularly relates to an airport capacity prediction method based on a causal forest, and the method comprises the steps: collecting historical operation data of a target airport, forming basic data, and carrying out the preprocessing; dividing meteorological types corresponding to the airport meteorological condition data into favorable meteorological conditions and unfavorable meteorological types; training a causal forest model; obtaining an individual processing effect of each unfavorable weather type, and calculating an average processing effect under each unfavorable weather type; obtaining airport theoretical capacity; and giving airport meteorological condition data of a certain to-be-predicted time period, obtaining a meteorological type corresponding to the to-be-predicted time period by using a classification algorithm, and estimating the airport capacity of the to-be-predicted time period according to the airport theoretical capacity and the average processing effect corresponding to the meteorological type in the to-be-predicted time period. According to the method, the causal effect of the processing variable on the target variable under the interference of various covariables can be estimated, so that a more accurate and stable capacity prediction result is provided.

The invention relates to the technical field of machine learning, and particularly provides a time sequence prediction method and system based on multi-scale feature reconstruction and multi-expert fusion, and the method comprises the steps: obtaining and preprocessing multi-source time sequence data, and constructing a structured feature matrix; performing multi-scale feature reconstruction on the sequence in the matrix by using three groups of one-dimensional convolutional networks with different receptive fields to obtain reconstructed feature sample sets with local, medium and global scales; performing time sequence slicing on the sample set, and calculating statistical characteristics in each slice; for the sample slices of each scale, respectively obtaining three prediction components of basic linear extrapolation, covariant linear correction and nonlinear residual through parallel branches; based on the statistical characteristics, dynamically fusing the three components through a gating network to obtain branch prediction results of all scales; and finally, fusing all branch prediction results through a full-connection network, and outputting a final prediction value. According to the method, the accuracy, robustness and practicability of time sequence prediction are effectively improved.

A method and system for analyzing urban green development efficiency, belonging to the field of smart city construction, is presented. It addresses the limitations of existing green productivity measurement methods in handling high-dimensional control variables and potential nonlinear relationships, as well as the inaccurate identification of causal effects. The method includes: acquiring a panel dataset of the target city; calculating a green efficiency value, stripped of environmental factors and random noise, using a data envelopment analysis model adjusted by stochastic frontier analysis; constructing and executing a dual machine learning model using the green efficiency value as the outcome variable, a pre-defined treatment variable as the treatment variable, and a high-dimensional control variable as a covariate; obtaining an estimate of the treatment effect of the treatment variable on the outcome variable through cross-fitting and residual regression; and generating analytical results on urban green development efficiency based on the green efficiency value and the estimated treatment effect. It is primarily used in the field of green productivity measurement.

The invention discloses a power load prediction method, and relates to the technical field of power load prediction. Comprising the following steps: decomposing an original load sequence to obtain a trend component and a residual component; decomposing the residual component to obtain a low-frequency approximate component and a high-frequency detail component; inputting the high-frequency detail component into an industrial load disturbance identification sub-network to obtain a clean high-frequency detail component and an industrial disturbance component; inputting the trend component and the low-frequency approximate component into a sparse-gated hybrid expert time sequence large model to obtain trend prediction; inputting the clean high-frequency detail component and the covariable into a multivariable small parameter model to obtain first fluctuation prediction; performing prior driving prediction on the industrial disturbance component to obtain second fluctuation prediction; fusing the trend prediction, the first fluctuation prediction and the second fluctuation prediction to obtain load prediction; and if facing a cross-regional deployment scene, performing fine tuning on the multivariable small parameter model. The technical problems of low prediction precision and poor robustness in the prior art are solved.