676 results about "Linear model" patented technology

The invention discloses an indoor temperature real-time regulation and control method of a heat distribution pipeline and a control system of the indoor temperature real-time regulation and control method, relates to the technical field of dynamic control of a heat distribution pipe network, and solves the problems of hydraulic oscillation and temperature control hysteresis caused by the fact that local valve regulation neglects whole-network coupling and a first-order linear model is difficult to describe multi-order thermal inertia and large heat capacity in the prior art. According to the scheme, on the basis of pipe network distributed PDE / lumped parameter hybrid modeling and in combination with extended Kalman filtering and unscented Kalman filtering on-line identification, a feedforward decoupling compensation item is generated through spectral decomposition, a self-adaptive multi-model predictive control and iterative learning compensation closed-loop structure is constructed, a control instruction is issued according to a pump-first and valve-second serialization strategy, and a self-adaptive multi-model predictive control and iterative learning compensation closed-loop structure is constructed. Meanwhile, the model weight and the prediction time domain are dynamically adjusted; according to the method, the global balance capability and the temperature tracking precision of heat distribution pipeline regulation and control are remarkably improved.

The invention relates to the technical field of fluid valve actuator equipment diagnosis, in particular to a fluid valve actuator monitoring and diagnosing system based on AI intelligence. According to the method, the target opening control instruction is generated through fluid medium characteristic data matching, and accurate regulation and control of the fluid valve are achieved; an opening degree change data set and a torque time sequence data set are captured in real time through a deployed sensor group, multi-dimensional dynamic fluctuation analysis is carried out, and intelligent recognition and screening of fluid valve abnormity are achieved; the medium temperature, the pipeline pressure and the medium flow velocity in the dynamic working condition parameters are substituted into a multiple regression linear model to generate reference torque and opening, and dynamic diagnosis of clamping stagnation abnormity is achieved by combining fusion calculation of a torque fluctuation coefficient and an opening deviation coefficient; through the correlation between the opening degree and the torque time sequence data and the traceability analysis of the current dynamic working condition parameters, the accurate positioning of the abnormal source is realized, and a systematic solution is provided for the state diagnosis and reliability evaluation of the fluid valve actuator.

The invention relates to an environment adaptive radio measurement calibration method and system. The method comprises the following steps: synchronously sampling temperature, humidity, air pressure and electromagnetic interference around radio equipment to obtain an environmental parameter vector; constructing a linear model and a nonlinear target model of a three-layer neural network structure according to the environmental parameter vector; solving an environmental parameter separation vector set by using the linear model and the nonlinear target model; executing one-dimensional iterative search optimization on the environment parameter separation vector set to obtain a target environment parameter vector; and inputting the target environment parameter vector into the nonlinear target model for calibration compensation calculation to obtain dynamic calibration compensation data. According to the method, the updating frequency is reduced when the environment is stable, the calibration precision is ensured, the system resource use is optimized, and the method is particularly suitable for wireless equipment with limited resources.

The invention provides an urban groundwater inrush risk assessment method based on a three-dimensional grid model and an XGBoost algorithm, and belongs to the technical field of geological disaster early warning. The method comprises the following steps: acquiring geological, environmental and historical disaster data in real time through a multi-source heterogeneous data interface; constructing a heterogeneous grid unit based on three-dimensional geological modeling, and dynamically binding parameters such as underground water level, fault activity and lithologic permeability coefficient; an XGBoost algorithm is combined with second-order Taylor expansion to optimize gradient calculation, and a nonlinear risk assessment model is constructed through Bayesian hyper-parameter tuning and SHAP value feature interpretive analysis; identifying contiguous dangerous areas based on spatial autocorrelation analysis and an adaptive threshold segmentation algorithm; and a risk thermodynamic diagram, geological section superposition and real-time early warning are realized through a 3D GIS platform. According to the method, the problems that a traditional method depends on a linear model, data is single and local risk description is insufficient are solved, and the accuracy, efficiency and engineering practicability of risk assessment are remarkably improved.

Systems, methods, and instrumentalities are disclosed for template based cross component linear model / multimode linear model (CCLM / MMLM) adjustment. In an example, a device, such as a video decoding device, or a video encoding device, may obtain a prediction model for predicting a coding block. The device may select an adjustment model, from multiple adjustment models, for adjusting the prediction model. The device may adjust the prediction model based on the selected adjustment model. The device may process (e.g., encode and / or decode) the coding block based on the adjusted prediction model.

A video signal processing method comprises the steps of: configuring a template including neighboring blocks of a current block; down-sampling luminance component samples of the neighboring blocks on the basis of a color format of a current picture including the current block; deriving a first linear model and a second linear model on the basis of the down-sampled luminance component samples; and on the basis of any one linear model from among the first linear model and the second linear model, predicting a chrominance component sample at a location corresponding to a location of a first sample from among the luminance component samples of the current block, wherein the any one linear model is determined by comparing a value of the first sample with a threshold value.

The invention relates to a quantitative control method for a nonlinear system, computer equipment and a storage medium. The method comprises the following steps of: performing sector bounded processing on the output of a static logarithmic quantizer to establish an expression for quantizing control input; constructing a nonlinear system with disturbance, time lag and parameter uncertainty, and constructing a T-S fuzzy linear model of the nonlinear system according to the expression; constructing a target function according to the T-S fuzzy linear model, and constructing an optimization problem on an infinite time domain by taking a system state constraint, a control input constraint and a disturbance constraint as conditions so as to realize minimum quantization control input under maximum disturbance; and performing stability analysis on the T-S fuzzy linear model by constructing an LKF, and solving an optimization problem under the condition of meeting the stability analysis to obtain quantitative control input. By adopting the method, stable control on a nonlinear system can be effectively realized.

The invention relates to a time sequence prediction method based on Transform, and belongs to the field of artificial intelligence time sequence prediction. The method comprises the following steps: preprocessing multivariable time series data, and performing season trend decomposition to obtain a season term and a trend term; predicting the trend term by adopting a linear model; a multi-layer perceptron (MLP) and a graph convolutional network (GCN) are adopted to model correlation among seasonal item variables; using convolution kernels of different scales to extract multi-scale features of seasonal items; constructing a position code and a timestamp code; a Transform-based encoder is constructed, and a sparse attention mechanism is adopted to capture a global dependency relationship of the time sequence; capturing a short-term dependency relationship of the time sequence by adopting a local attention mechanism; initializing a decoder input; and a decoder based on a Transform is constructed to realize prediction.

The invention relates to the technical field of industrial mechanism models, in particular to a shock absorber performance optimization control method based on model fusion, which comprises the following steps: extracting a low-frequency disturbance state variable and inputting the variable-topology industrial mechanism model to generate a nominal reference state trajectory; utilizing a depth state observation network fused with energy passivity constraint to calculate a non-linear model mismatch compensation amount and an adaptive weighting parameter; performing dynamic fusion on the nominal reference state trajectory and the compensation amount based on the adaptive weighting parameter to generate a generalized state estimation value; and executing dynamic multi-objective optimization based on the generalized state estimation value, and generating mixed mode control input acting on an execution end. According to the invention, through adaptive fusion of a mechanism model and a data driving method, physical consistency, calculation real-time performance and robustness of a control process are considered.

The invention discloses a hydraulic engineering operation and management cooperative control system based on digital twinning, and the system comprises the following steps: a data assimilation module which collects observation station, remote sensing and meteorological data and assimilates the data to obtain an assimilation estimator; the dictionary and lifting module is used for generating a lifting variable sequence according to hydrodynamic force prior; the structure identification and uncertainty module is used for identifying a lifting linear model and recursively predicting uncertainty under the constraints of conservation, monotonicity, dissipation and spectral radius; the terminal security module constructs a positive invariant terminal security set according to a support function; the opportunity constraint module is used for setting a total out-of-limit probability budget and generating an opportunity constraint substitution set; the Koopman rolling optimization module is used for solving the control sequence and executing the first control quantity; and the online updating and correcting and deformation and verification module is used for executing residual error triggering low-rank correction, projection return and support surface parameter online deformation. According to the invention, risk-controllable and stable and efficient collaborative scheduling is realized.

The invention provides a predetermined time distributed optimal inclusion control method for an unmanned aerial vehicle cluster, and belongs to the technical field of collaborative decision of an unmanned cluster system. The method comprises the steps that unmanned aerial vehicles in an unmanned aerial vehicle cluster are divided into virtual leader unmanned aerial vehicles and follower unmanned aerial vehicles; according to the single integral linear model of the virtual leader unmanned aerial vehicle, generating a preset formation pattern through a preset time distributed optimization theory; constructing a nonlinear model of a follower unmanned aerial vehicle in the unmanned aerial vehicle cluster; and designing an inclusion control protocol of the follower unmanned aerial vehicle based on a distributed inclusion control algorithm according to the nonlinear model of the follower unmanned aerial vehicle, so that the follower unmanned aerial vehicle converges to a preset formation. Through a distributed optimal formation control algorithm and a distributed inclusion control algorithm, a virtual leader unmanned aerial vehicle formation generation strategy within a preset time is realized, follower unmanned aerial vehicles converge to a preset formation, and thus unmanned aerial vehicle cluster security cooperative control under a complex sea condition is completed.

An intra prediction method by using cross component liner prediction mode (CCLM), includes: determining a luma block corresponding to a current chroma block; obtaining luma reference samples of the luma block based on determining L available chroma reference samples of the current chroma block, wherein the obtained luma reference samples of the luma block are down-sampled luma reference samples; calculating linear model coefficients based on the luma reference samples and chroma reference samples that correspond to the luma reference samples; and obtaining a prediction for the current chroma block based on the linear model coefficients and values of a down-sampled luma block of the luma block.

The ELM-based frequency safety constraint linearization optimization scheduling method comprises the following steps: establishing a fire-wind-light multi-energy complementary frequency dynamic response model, and deducing a time domain analytical expression of a frequency safety index by taking a frequency maximum change rate and a frequency maximum deviation as the frequency safety index of a system; considering a frequency lowest point security constraint and a frequency maximum change rate security constraint, and establishing a frequency security constraint-containing optimization scheduling model; a maximum bearable disturbance power linearization mapping method based on ELM is provided, nonlinear frequency security constraints are converted into a series of linear constraint sets, and linearization processing of the frequency security constraints is achieved; and piecewise linearization is carried out on quadratic terms in the established objective function containing the frequency security constraint optimization scheduling model, the overall model is converted into a mixed integer linear model, and a Gurobi solver is used to solve the linearly processed model. The scheduling method can ensure that the system frequency is safe and stable, so that the established optimal scheduling model gives consideration to both economy and frequency safety.

The invention relates to a dynamic grid-based foundation pit temperature field-seepage-stress field coupling calculation method, which comprises the following steps of: establishing a nonlinear dynamic bidirectional coupling constitutive relationship of a temperature field, a seepage field and a stress field containing a two-dimensional foundation pit cofferdam, and obtaining a three-field full coupling equation set; in combination with a GPU acceleration solver, solving the three-field full-coupling equation set by adopting a sub-domain decoupling-asynchronous iteration strategy; and performing real-time early warning on the cofferdam construction period according to a solving result. The method has the advantages that compared with a traditional linear model, the precision of the established three-field full-coupling equation set is improved, and the interaction influence relation of three fields can be well simulated.

The invention relates to the technical field of prediction models, in particular to a battery energy storage life prediction method and system based on multiple time scales, and the method comprises the following steps: deploying three groups of short-term, medium-term and long-term parallel data windows, obtaining battery capacity and internal resistance parameters, sliding window initialization parameters, synchronous baseline calibration, and Kalman filtering correction parameters, and extracting a variation amplitude and rate, an output capacity mean value and an internal resistance variance, calculating a residual sequence, performing three-time linkage on the capacity residual exceeding a threshold to trigger an abnormity, performing weighted fusion after the abnormity, and outputting the residual life. According to the invention, through multi-time scale parallel data window dynamic monitoring, sliding window synchronous calibration base line, Kalman filtering dynamic correction parameters, capacity and internal resistance residual error linkage determination mechanism establishment, dual-attenuation coupling model fusion multi-dimensional attenuation characteristics and multi-scale collaborative analysis of battery aging, the limitation of a linear model is broken through; the anomaly detection sensitivity and robustness are enhanced, the life prediction accuracy and timeliness are improved, and the misjudgment risk is reduced.

The invention discloses a plant extraction method and system based on image processing, and relates to the technical field of image processing. The method comprises the following steps: acquiring and standardizing a plant raw material image; extracting feature vectors of the images by using a neural network convolutional layer algorithm, comparing the feature vectors with a plant raw material standard feature database, and screening out qualified images of the plant raw materials; qualified images are processed through multispectral image analysis and an image segmentation algorithm, and an effective component spatial distribution diagram is obtained; constructing a partial least squares regression linear model, extracting spectral feature vectors of the effective components, inputting the spectral feature vectors into the model to obtain concentration values, and calculating quantitative data of the effective components; establishing an extraction process rule base, selecting a process parameter combination according to a spatial distribution diagram and quantitative data, and extracting to obtain a primary extracting solution and plant residues; and calculating an effective component residual error rate by matching and mapping to a qualified image coordinate, and if the effective component residual error rate is greater than a preset threshold, performing secondary extraction to obtain a secondary extracting solution. And quantitative basis is provided for extracting process parameters through the spatial distribution diagram of the effective components.

The invention relates to an electric vehicle wireless charging system LPV-Hammerstein model identification method based on an alternating least square iteration strategy, and belongs to the technical field of power electronic system modeling and parameter identification. The method comprises the following steps: firstly, constructing an LPV-Hammerstein model structure based on a WPT system circuit topology and a working mechanism, and then collecting discrete data of a system input control signal U, an output DC voltage Vo and a scheduling variable Req; then, alternately fixing other parameters by adopting an alternate least square iterative algorithm, and converting a to-be-estimated parameter problem into a linear least square sub-problem to be solved; and when a convergence condition is satisfied, a final LPV-Hammerstein model parameter estimation result is obtained. According to the method, the common nonlinear and parameter change coupling characteristics in the wireless charging system of the electric vehicle can be effectively processed, a systematized and efficient way is provided for obtaining a high-precision system dynamic model, and the limitation that the system cannot be accurately described by a traditional linear model or a simple nonlinear model is overcome.

The invention belongs to the field of near-inertia internal wave modeling, and particularly relates to a near-inertia internal wave modeling method based on a reverse echo observation array, which comprises the following steps: arranging a reverse echo measurement device array in an observation area, and acquiring echo signals from the seabed to the sea surface; performing data cleaning and band-pass filtering on the signals, and extracting near-inertia internal wave signals; calculating the depth and the relative vorticity of a mixed layer by combining historical temperature-salinity data; calculating a wind input energy flux by using the wind stress data; calculating a net level energy flux by adopting a boundary integration method; judging whether the boundary energy flux influence is smaller than a threshold value or not, and if not, adjusting array layout; and finally establishing a multivariable linear model containing wind energy input, mixed layer depth, relative vorticity and energy flux. According to the method, the temporal-spatial resolution and prediction precision of near-inertia internal wave modeling are effectively improved, the problems that in a traditional method, the boundary effect is remarkable, and multi-factor comprehensive modeling is insufficient are solved, and the method is suitable for long-term efficient monitoring of a large-range marine environment.

The invention discloses an electromagnetic target recognition method based on interpretable multi-task learning, a storage medium and equipment, and the method comprises the steps: firstly constructing a multi-task learning network facing modulation recognition and individual recognition, inputting an original signal, setting a hyper-parameter, and carrying out the training of the multi-task learning network; constructing a generator for outputting a disturbance signal, segmenting an original signal and generating an original mask, generating a disturbance mask through an inversion element, and further generating the disturbance signal; inputting the disturbance signals into the trained multi-task learning network to obtain probability distribution of classification results under different tasks; taking a prediction probability of a multi-task model to a disturbance signal as a supervision label, training a local linear model on an original mask and a disturbance mask, and taking a model weight Wg as a contribution degree of each subsequence to explanation through optimization; and performing normalization processing on the Wg, mapping the weight back to the original signal length, and visually marking a time-frequency domain signal region which plays a key role in model decision based on a constellation diagram.

The application discloses a curved surface structure zero state form finding method, comprising the following steps: (1) a model is established based on a design configuration, and deformation data is obtained by simulating the whole construction process; (2) a zero state configuration is deduced reversely: key node deformation values are extracted, arches are started, a simulation is re-performed after a linear model is fitted and adjusted, and iteration correction is performed until deformation error meets the requirement, and a zero state installation configuration is obtained; and (3) construction data is generated by extracting component line shapes and node coordinates. The application innovatively adopts an integral curvature adjustment strategy to replace a traditional node coordinate local adjustment mode, and through reverse deduction of 'construction simulation-control point arching-curved surface fitting-grid mapping', the problem of curved surface discontinuity caused by the traditional reverse superposition method is solved, construction precision and efficiency are significantly improved, and the application is suitable for intelligent construction of complex curved surface space grid structures.

The invention discloses an air-ground cooperative system model prediction formation control method in an underground pipe gallery environment, and the method comprises the steps: constructing an air-ground cooperative system nonlinear model, and constructing an underground pipe gallery model; designing a high-order interference observer to estimate current unknown external disturbance, and establishing an interference prediction model to estimate future unknown external disturbance; a rolling optimization control strategy based on state prediction is designed, control signals of all followers are optimized according to the current state of the system and the predicted future state in each control period, and the tracking precision is ensured while the tracking precision is improved. Collision between followers, collision between the followers and the wall of the pipe gallery and obstacle avoidance between the followers and static and dynamic obstacles in the pipe gallery are achieved; and solving is carried out based on a distributed model prediction formation control algorithm combined with a Lyapunov stability theory, so that each follower keeps an expected formation configuration while tracking a leader trajectory. And the driving safety of the formation in the underground pipe gallery environment is ensured.

Systems and methods that enable the efficient and adaptive allocation of resources dedicated to a virtualized resource-based computation (e.g., one or more information processing tasks) are provided. In one embodiment, a reward model is generated based on a set of statistical distributions, for example, in response to receiving a request to launch a set of VCRs. Thereafter, an expected reward is predicting for each configuration of a set of configurations based on the reward model and one or more parameters of the corresponding configuration. The expected reward indicates an efficiency in distribution or allocation of physical computation resources to the set of VCRs. A configuration of the set of configurations is selected based on the predicted expected reward for the configuration. The set of VCRs are then configured with the selected configuration.

The invention is suitable for the technical field of intelligent medical treatment, and provides an intelligent data analysis assistant based on mesh meta analysis, which comprises a processor, a data acquisition module, a preprocessing module, an analysis modeling module, a statistical analysis module, a visualization module, a generation module, an AI auxiliary analysis module and a storage module are arranged in the processor; the analysis modeling module is used for automatically constructing a treatment network diagram on the basis of multi-element mesh Meta analysis and in combination with a Bayesian multi-level analysis model and a mixed linear model, displaying direct and indirect comparison relationships among different treatments and providing selection of a fixed effect model and a random effect model; the assistant helps a user to quickly understand complex data by automatically constructing a treatment network diagram, displaying direct and indirect comparison relationships among different treatments, providing selection of a fixed effect model and a random effect model, automatically calculating direct comparison and indirect comparison results of treatment effects, providing consistency check, improving retrieval analysis ability and helping the user to quickly understand the complex data.

The invention discloses a surface temperature downscaling method and system based on feature interaction optimization and a multi-model screening mechanism, and belongs to the technical field of surface temperature data processing, and the method comprises the following steps: screening spectral bands, remote sensing indexes and topographic features; resampling the topographic feature parameters to a resolution matched with the optical wave band and the remote sensing index; variables significantly related to the surface temperature are screened through correlation analysis, interaction features are constructed, and high-contribution features are dynamically screened based on an SHAP value; constructing a regression model based on double frameworks of a linear model and a machine learning model, and verifying the regression model; sHAP optimization features extracted from Sentinel-2 data are fused with SRTM topographic data, a verified regression model is input, and surface temperature data with the resolution of 10 meters are generated; according to the method, the high-order nonlinear relation is mined through combined feature interaction, the prediction RMSE in the uniform earth surface area is greatly reduced, the data scale effect is reserved, the prediction is more suitable for the physical process, and the space consistency is better.

The invention discloses an NUMA (Non Uniform Memory Access) architecture perceived high-scalability learning index construction method and system and an operation method. According to the construction method, a mixed node tree structure is adopted to organize data, the structure comprises internal nodes accurately searched by a linear model and leaf nodes for storing data, the leaf nodes are composed of ordered gap nodes and segment nodes, and the segment nodes support dynamic evolution from an ordered stage to a semi-ordered stage based on data density. Conflict data is managed using hierarchical benchmark buckets. According to the system, a self-adaptive node evolution mechanism based on an operation cost model is realized, index performance is optimized through node reconstruction triggered by foreground write operation and background hot and cold node compression, a multi-thread concurrency control strategy is adopted, and a memory management and thread scheduling framework perceived by NUMA is adopted. The operation method provides an efficient point query and point insertion operation process. The data layout can be dynamically optimized, the operation cost is reduced, and the method is suitable for high-performance database and memory management scenes.

The invention discloses a compaction quality detection method based on vibration signal interval cross-correlation optimization, which comprises the following steps: firstly, carrying out grid division on a field compaction area and arranging sampling points, collecting vibration signals through an acceleration sensor on a road roller and measuring the compaction degree of each point after compaction, then carrying out area division, preprocessing vibration signals in a specified interval, and carrying out cross-correlation optimization on the vibration signals in the specified interval; and calculating the cross correlation of adjacent intervals. And determining a signal rule degree and a cross correlation peak reference value, removing corresponding ICMV values of the interval lower than the reference value, and establishing a linear regression model of the actual compactness and the ICMV values. And adjusting the interval range and performing iteration, and selecting the model with the highest fitting precision as a final compaction quality analysis model. According to the method, extra detection means are not needed, and the cross correlation of the vibration signals of the road roller is used for detecting the section with large variability; by eliminating variability values, the accuracy of the linear model for evaluating the compactness is improved; meanwhile, the influence of the interval size on the vibration signal variability is also considered.

A method and apparatus for video coding are disclosed. According to the method for the decoder side, encoded data associated with a current block comprising a first-colour block and a second-colour block are received. An inherited model parameter set is determined from a previously coded block coded in a first CCLM related mode, wherein the inherited model parameter set comprises a first scaling parameter associated with the first CCLM related mode. A final inherited model parameter set is derived if an update value for the inherited model parameter set is determined, where the final inherited model parameter set is determined based on the first scaling parameter and the update value. Then, the encoded data associated with the second-colour block are decoded using prediction data based on an updated CCLM related model associated with the final inherited model parameter set. A method and apparatus for the encoder side are also disclosed.

A method for deriving a cross component or convolution model when encoding or decoding a current block of a current picture of a video is provided. The video coder retrieves a set of reconstructed component samples for a first set of pixel positions that are required for deriving a component prediction model of the current block. The video coder generates a set of padded component samples for a second set of pixel positions that are required for deriving the component prediction model based on the retrieved set of reconstructed component samples. The video coder derives the component prediction model by using the retrieved set of reconstructed component samples and the generated set of padded component samples. The video coder encodes or decodes the current block by using the component prediction model to generate a predictor for the current block.

The invention provides a gradient coil model correction method for ultra-low field magnetic resonance, and relates to the technical field of gradient coil control in medical imaging equipment, and the method comprises the steps: collecting a real voltage signal d (n) and a real current signal x (n) of a gradient coil in real time, and n represents a sampling moment; a gradient coil model is constructed, the gradient coil model comprises a linear model constructed based on an adaptive filtering algorithm and a nonlinear model constructed based on a neural network algorithm, and a first prediction voltage signal # imgabs0 # output by the linear model and a second prediction voltage signal # imgabs1 # output by the nonlinear model are determined according to the real current signal x (n); determining a mixed prediction voltage signal # imgabs4 # of the gradient coil model according to the first prediction voltage signal # imgabs2 # and the second prediction voltage signal # imgabs3 #; and updating the weight of the linear model and the weight of the nonlinear model according to the real voltage signal d (n) and the mixed predicted voltage signal # imgabs5 # so as to adapt to gradient coil model correction under complex working conditions.

The invention relates to a time sequence prediction method based on time-frequency double-domain decomposition and multi-cycle feature fusion, and the method comprises the following steps: carrying out the normalization operation of input historical data, and carrying out different feature extraction operations for different data sets. Specifically, a trend term is firstly extracted using a moving average in a time domain. Then, a trend term is extracted by applying a selection mode of an adaptive spectrum rarefaction mechanism in a frequency domain, and meanwhile, noise is isolated into a residual term; for a multi-period data set, trend terms are decomposed in a recursive manner, and period modes are separated from short to long time scales. And for each decomposed feature item, learning and predicting by using a prediction module based on a linear model or a multi-layer perceptron, and finally fusing prediction components of each feature and performing inverse normalization operation to obtain a prediction result. According to the method, novel characteristic decomposition processing is carried out on the input data, more details are provided for future prediction, and a better prediction effect is achieved.