20 results about "Linear trend" patented technology

The invention discloses a material demand prediction method based on an ARIMA model and an LSTM model, and the method comprises the steps: constructing a time-serialized material demand data set, and carrying out the data preprocessing; time sequence analysis is carried out, linear trend modeling is carried out by using an ARIMA model, a preliminary prediction value is obtained, and a prediction residual error is calculated; constructing an LSTM neural network to perform nonlinear feature learning on the residual error, predicting a future residual error value, determining an optimal model by using an LSTM model evaluation index, and predicting future annual material consumption by using the optimal model; determining a calculation range of an annual purchase quantity, and establishing an annual recommendation purchase model; calculating an annual purchase quantity recommendation value of a plurality of years in the future, and forming an annual purchase schedule; and setting an early warning inventory value, and monitoring the inventory in real time. Compared with a traditional method depending on a single model, the method can more comprehensively capture tendency and volatility in material demands, and adapts to changes of material consumption modes in complex construction scenes.

The invention belongs to the technical field of power system operation and planning, and discloses a power grid net load fluctuation scene generation method, system and device based on an ARIMA and Copula combined model and a medium, so as to solve the problem of poor scene generation accuracy. The method comprises the following steps: decomposing and reconstructing an original time sequence of the net load of the power grid by using discrete wavelet transform; taking permutation entropy minimization as an optimization target, adopting a variable chromosome length hybridization genetic algorithm to divide time segments for the low-frequency linear subsequences, and respectively establishing ARIMA models to generate linear trend scenes; establishing a joint probability distribution model of the high-frequency fluctuation subsequences at adjacent moments based on a Copula function, and deducing conditional probability distribution in combination with a Bayesian formula to generate a fluctuation scene; and the linear trend scene and the fluctuation scene are superposed to form an initial net load scene set, and a k-means clustering algorithm is adopted to reduce the initial net load scene set to obtain a representative scene set.

This invention discloses a multivariate time series forecasting method based on prediction domain transformation and dual-path fusion. A time series forecasting model, PDT, is proposed. It utilizes data-adaptive prediction domain transformation to project the time series onto the energy-concentrated optimal latent space for prediction. A dual-path architecture is designed to simultaneously achieve efficient extrapolation of dominant linear trends and capture complex nonlinear dynamics. A masked channel dependency strategy and a lightweight linear encoder are used to adaptively fusion and filter inter-variable dependencies and deeply extract intra-channel features, respectively. The proposed time series forecasting model significantly solves the problem of insufficient nonlinear expressive power of traditional linear models and enhances noise resistance and computational efficiency in high-dimensional multivariate scenarios, comprehensively improving the prediction accuracy and robustness of the model in multivariate time series forecasting tasks.

The invention relates to an international sports match medal prediction method, system and device and a medium. According to the method, original historical data is subjected to standardization preprocessing to obtain structured basic data, then a weighted synthesis algorithm is adopted to quantify the historical comprehensive performance level of each country on each sports item, and a Holt linear trend prediction method is adopted to predict the performance level of a future match according to a historical trend. And then simulating a random process of medal distribution of each sports item through a Monte Carlo simulation method, and finally aggregating multiple simulation results to output stable prediction of the number of meals of each country in a statistical value form. Therefore, the technical effects that effective information is automatically extracted from massive heterogeneous data, the competition strength change trend is scientifically quantified, competition uncertainty factors are reasonably included, and then systematic, repeatable and stable medal list prediction is generated are achieved.

The invention discloses a multi-model fusion polar shift prediction method and system, and relates to the technical field of polar shift prediction. The method comprises the steps of firstly obtaining a polar shift historical time sequence; constructing a linear trend model for representing a polar shift linear change trend in the polar shift historical time sequence, a multi-channel singular spectrum analysis model for representing a polar shift change trend in a primary residual sequence, and a trained XGBoost model for representing a polar shift change trend in a secondary residual sequence; and performing polar shift prediction by using the linear trend model, the multi-channel singular spectrum analysis model and the trained XGBoost model. According to the method, the linear model, the multi-channel singular spectrum analysis model and the XGBoost model are fused, polar shift prediction is realized, and the precision and the stability of polar shift prediction are improved.

The invention relates to the technical field of electricity market transaction, in particular to an electricity spot market price prediction and transaction optimization method. According to the technical scheme, the electric power spot market price prediction and transaction optimization method comprises a work flow of electric power spot market price prediction and transaction optimization; according to the method, the robustness and the prediction precision of a price prediction model in the face of market complexity and sudden events are remarkably improved, and a long-term and short-term memory network component can effectively capture and memorize a nonlinear dependency relationship from a historical price sequence and related multi-dimensional features by virtue of a gating mechanism; the model can understand a more abstract market dynamic mode, the autoregressive integral moving average model component is used for capturing inherent linear trends and short-term laws in a time sequence, and the model is allowed to dynamically allocate different weights for input information of past different time steps by setting an attention mechanism when prediction is performed each time.

A method and system for identifying the critical transformation risk of an ecosystem based on multi-source remote sensing vegetation indicators are disclosed. The method includes: S10: acquiring and preprocessing vegetation time-series data of the study area; S20: performing signal separation based on multiple time-series decomposition methods; S30: selecting the best separation result from the multiple time-series decomposition methods; S40: calculating statistical indicators for early warning of critical transformation; S50: performing linear trend analysis on the time series of early warning indicators for each pixel, and calculating the slope b of the indicator value changing over time using the least squares method; S60: directly summing the trend values ​​b of each indicator obtained in step S50 to obtain the pixel-scale consistency score CS; S70: constructing a multi-indicator covariance matrix; calculating the standard deviation MS of the Mahalanobis distance time series; S80: constructing a critical transformation risk index R, and judging the critical transformation risk of the ecosystem in the study area based on R.

The present application provides a permafrost region surface stability evaluation report generation method and device, analyzes the time series cumulative deformation data of the pixels in each observation time period to obtain the annual average deformation rate corresponding to each pixel, the seasonal fluctuation amplitude corresponding to the pixel, the trend significance coefficient corresponding to the linear trend of the pixel and the variation coefficient corresponding to the time series cumulative deformation data of the pixel; according to the annual average deformation rate, the seasonal fluctuation amplitude, the trend significance coefficient and the variation coefficient corresponding to each pixel, the risk level is determined, and then the surface stability evaluation report corresponding to the target region is generated. The combination of long-term plastic deformation of frozen soil and short-term elastic deformation of active layer can accurately lock the composite type extremely high risk area with fast settlement and large fluctuation, strict data quality control, double statistical constraints of trend significance coefficient and variation coefficient, and realize the algorithm level active denoising mechanism, which can eliminate the false positive of geological disasters from the root.

The present application relates to the technical field of hydrological data, and more particularly to a system and method for comprehensive analysis of hydrological element change trend. The method comprises the following steps: obtaining time series data of hydrological elements of a target basin, eliminating the influence of upstream reservoir regulation and storage in the time series data of hydrological elements, and obtaining a natural runoff sequence; performing linear trend analysis on the natural runoff sequence under different time scales to obtain a regression slope parameter; when a non-zero trend signal is detected in the regression slope parameter, triggering a non-parametric significance test to generate an effective trend signal; decomposing the natural runoff sequence into multi-level periodic characteristics by time-frequency wavelet; identifying the main oscillation period of the multi-level periodic characteristics, calibrating the dominant period and the average change period, and constructing a periodic oscillation atlas of hydrological elements. The present application integrates trend signals, periodic characteristics and persistence parameters to construct a trend comprehensive identification path, and realizes the characterization of the change characteristics of hydrological elements.

The application relates to the technical field of ecological environment monitoring and intelligent data analysis, and discloses a method and system for observing and positioning ecological quality in an alpine endorheic region, which comprises the following steps: arranging a three-dimensional observation network in the target alpine endorheic region, continuously collecting data in a hierarchical classification mode, forming a multi-ecosystem collaborative monitoring data set, standardizing the data set, and constructing a structured standard database; inputting time-series monitoring data in the standard database into a pre-trained correlation analysis model, respectively extracting dynamic spatial correlation features between observation field nodes, multi-scale periodic features of time-series data, and nonlinear trend features, and adaptively fusing the features to generate comprehensive feature vectors of the observation field nodes; simultaneously outputting quantitative results representing the collaborative relationship between different observation field nodes; and using a prediction model to perform short-term, medium-term and long-term trend prediction on core ecological indexes. The method realizes multi-period and confidence interval precise prediction of core ecological indexes of a multi-ecosystem in an alpine endorheic region.

This application relates to the field of artificial intelligence technology, and more particularly to a multi-scale drought index prediction method based on a hybrid model. It models the linear trend and periodic variations of the SPEI drought index time series using a SARIMA model; mines nonlinear features from the SARIMA model residual sequence using a long short-term memory network; introduces a lightweight spatial attention mechanism to adaptively weight the importance of different spatial grid locations; and constructs a dynamic residual weight adjustment mechanism to adaptively adjust the fusion ratio of linear prediction results and nonlinear correction results based on the actual contribution of residual correction to improving prediction accuracy. The aim is to address the problem of how to improve the reliability of drought prediction results using a hybrid model.

The present application relates to geophysics and geodetic data processing technical field, disclose the GNSS time series analysis and modeling method considering variable amplitude, the method comprises the following steps: S1, obtain the GNSS coordinate time series;S2, time series is modeled as the Gaussian process defined by mean function and covariance function;S3, construct the mean function describing long-term linear trend;S4, construct the product kernel by the periodic kernel and the non-periodic smooth kernel multiplication as the covariance function, to unify the periodicity and time-varying amplitude of the signal modeling;S5, the maximum likelihood estimation method is used to solve the hyperparameter in model;S6, the time series is decomposed by using the optimized model, and the time-varying amplitude periodic signal is obtained.The present application can integrally model the periodic signal and its time-varying amplitude by constructing the Gaussian process product kernel, so as to realize the accurate separation of signal component.

The invention relates to the technical field of ecological environment monitoring and intelligent data analysis, and discloses a method and system for observing and positioning the ecological quality of a high and cold internal flow area, and the method comprises the steps: laying a three-dimensional observation network in a target high and cold internal flow area, carrying out the layered and classified continuous data collection, forming a multi-ecosystem cooperative monitoring data set, and carrying out the standardization processing, constructing a structured standard database; inputting the time sequence monitoring data in the standard database into a pre-trained correlation analysis model, respectively extracting dynamic space correlation features among observation field nodes, multi-scale periodic features of the time sequence data and nonlinear trend features, and performing adaptive fusion to generate comprehensive feature vectors of the observation field nodes; meanwhile, a quantitative result representing the cooperative relation between the nodes of different observation fields is output; and performing short-term, middle-term and long-term trend prediction on the core ecological indexes by using the prediction model. And multi-period accurate prediction with a confidence interval of the core ecological index of the multi-ecological system in the high and cold internal flow area is realized.

The invention relates to the technical field of electronic digital data processing, and discloses an industrial wastewater quality prediction method which comprises the following steps: acquiring historical time sequence data of key water quality parameters including an organic pollutant index, a nutritive salt index and a physical and chemical index, and preprocessing; performing periodic feature coding and linear trend feature coding on an input timestamp, and splicing all features into an enhanced feature sequence; inputting the enhanced feature sequence into a recurrent neural network for time sequence coding, and extracting a long-term dependency relationship of historical data; based on time sequence coding output, introducing an adaptive attention mechanism capable of learning a scaling factor to calculate the attention weight of each time step, and carrying out adaptive weighted fusion on the features; and processing the weighted and fused features, and outputting a plurality of key water quality parameter predicted values of a plurality of time steps in the future. The problems of insufficient utilization of time sequence features and weak multi-parameter correlation modeling are solved, and the purposes of multi-parameter joint prediction and adaptive attention adjustment are achieved.

The invention relates to the technical field of data analysis, and provides an error quadratic sum decomposition method and system for distinguishing seasonal influence of global average sea temperature, and the method comprises the steps: carrying out the processing based on a prediction time sequence of global average sea temperature over the years provided by a preset sea temperature prediction model and an observation time sequence of global average sea temperature over the years obtained through observation; according to the method, the error sum of squares is obtained through calculation and analysis, the error sum of squares is accurately decomposed into the multi-year mean deviation sum of squares, the multi-year linear trend deviation sum of squares and the residual sum of squares, the error source of the sea temperature forecasting model can be finely analyzed, and the interpretability is improved.

This invention discloses a dynamic prediction method for gas turbine gas path system faults, comprising: acquiring multi-source operating parameters and model output parameters during gas turbine operation; constructing hysteresis features and extracting sliding statistical features from the parameters to form a time-series feature vector; based on an integrated regression prediction model, using the time-series feature vector to predict the state parameters of the compressor, combustion chamber, and turbine online. Using the original baseline efficiency as a reference, the predicted efficiency parameters are normalized to construct a health index reflecting the degree of component degradation; within a sliding time window, a piecewise linear trend analysis of the health index is performed, and the current state is used as an anchor point to extrapolate the future degradation trend; for a preset health threshold, the remaining time for the health index to reach the threshold is calculated, and the uncertainty of the threshold arrival time is quantified using a block bootstrapping method, outputting a prediction interval to achieve early warning of gas turbine degradation status.

The invention relates to the field of seismic data processing, and particularly discloses a method for automatically extracting geophysical abnormal waveforms by utilizing multiple algorithms, which comprises the following steps of: S1, data preprocessing: acquiring an abnormal interference waveform data curve of historical observation data of a geophysical station network in a target area, carrying out linear trend removal and direct current component removal preprocessing on single measurement item data to obtain a corrected abnormal interference waveform curve, and obtaining preprocessed data; s2, primary algorithm identification: identifying forms of obvious order change, kick, daily variation amplitude change, steps, distortion and trend turning deviating from a normal waveform data curve in the preprocessed data by adopting slope change, first-order difference and second-order difference methods; and S3, carrying out secondary algorithm identification. By adopting the technical scheme of the invention, the abnormal waveform data trend turning change, multiple number and abnormal tiny change can be identified, different station test items are adapted, and the kick response delay time is short.

The invention discloses an integrated energy system load prediction method based on CDA-LSTNet, and relates to the technical field of energy prediction. The method comprises the following steps: firstly, constructing a multi-dimensional time sequence input matrix containing historical load and screened environmental parameters; then, constructing an LSTNet model for cross-dimension attention optimization, and extracting local short-time dependence and long-time time sequence features of the data by using a convolutional layer and a circulating layer; innovatively introducing a cross dimension attention layer (CDA), calculating association weights among different feature dimensions, and performing adaptive reconstruction to explicitly capture a dynamic coupling relationship among multiple variables; capturing a linear trend by using an autoregression layer; and finally fusing nonlinear and linear components to output a predicted value. According to the method, the problem that a traditional model ignores the multi-dimensional coupling characteristic between input variables is effectively solved, and the precision and robustness of load prediction are remarkably improved under the multi-energy-flow complex working condition.

The application discloses a kind of gas turbine gas path system fault dynamic prediction method, comprising: obtaining the multiple-source operating parameters and model output parameters in the operation process of gas turbine, lag feature construction and sliding statistical feature extraction are carried out to the parameters, and time sequence characteristic vector is formed;Based on integrated regression prediction model, the state parameters of compressor, combustion chamber and turbine are predicted online using the time sequence characteristic vector.The original benchmark efficiency is used as reference, the efficiency parameters are normalized, and the health index reflecting the degradation degree of the component is constructed;The health index is analyzed by piecewise linear trend in the sliding time window, and the future degradation trend is extrapolated with the current state as the anchor point;The remaining time for the health index to reach the threshold value is calculated for the preset health threshold value, and the threshold value arrival time is quantified by block bootstrap method, the prediction interval is output, and early warning of the degradation state of gas turbine is realized.

The present application relates to a kind of space radiation benchmark precision prediction method based on climate element dataset, first collect the satellite observation data of specific area and carry out regional average and month average processing, generate the regional monthly average dataset eliminating random noise;Then using difference method eliminates the seasonality in time series data, using least square method eliminates the linear trend in time series data, obtain the dataset that can characterize natural variability size;Again, the standard deviation of this dataset is obtained natural variability size σ var , using autoregressive model obtains natural variability correlation time τ var ;Finally, σ var , τ var And other parameters are substituted into climate precision prediction model to obtain benchmark load detection precision σ cal The present application uses difference method, least square method and autoregressive model method to process climate element time series dataset, so as to obtain space radiation benchmark load precision.