32results about How to "Shorten forecast time" patented technology

The invention belongs to the technical field of ocean engineering, and discloses a column structure response forecasting method and system based on a residual connection neural network. According to the method, a physical information neural network (PINN) and a Kolmogorov-Arnold (KAN) theorem are combined, and the precision and the calculation efficiency of a prediction model are remarkably improved through a multi-layer nonlinear mapping framework and residual connection optimization. According to the method, physical equation constraint loss and environmental parameter input are introduced, so that the model can more accurately simulate dynamic response characteristics in an extreme marine environment, and the limitation of a traditional method in nonlinear factor processing and real-time calculation is effectively solved. Experiments show that the method has remarkable advantages in the aspects of accuracy, generalization ability and calculation efficiency, provides reliable and efficient technical support for ocean platform structure design and operation state monitoring, and has important theoretical value and wide application prospects.

The invention belongs to the technical field of hydrological prediction, and discloses a parallel architecture runoff prediction method fusing characteristic decomposition and an error correction strategy, and the method comprises the steps: collecting a runoff time sequence of a target region, and carrying out the decomposition of the runoff time sequence through employing a VMD optimized through an SSA algorithm; screening a subsequence most related to runoff change as a runoff prediction factor; inputting the runoff prediction factors into a parallel CNN-GRU model to extract spatial-temporal characteristics, and optimizing parameters of the parallel CNN-GRU model by using an SSA algorithm; inputting a runoff prediction factor into the optimized parallel CNN-GRU model for preliminary prediction, and predicting an error value based on the optimized parallel CNN-GRU model; and superposing the prediction error value to the initial prediction value to correct the prediction deviation, and obtaining a final runoff prediction value. According to the method, by combining the parallel CNN-GRU architecture, the SSA optimization algorithm and the error correction mechanism, the precision of regional runoff prediction is remarkably improved.

The invention belongs to the technical field of power system load prediction, and particularly relates to a short-term power load prediction method based on a hybrid model, and the method comprises the steps: S1, carrying out the preprocessing of original data; s2, searching an optimal combination [k, alpha] by adopting a firefly and sparrow algorithm; s3, decomposing the original data by adopting variational mode decomposition; s4, determining a demarcation point by adopting an adaptive algorithm, and dividing each signal component; s5, selecting features by adopting the maximum information coefficient, and establishing an input matrix; s6, substituting each signal component into the model for prediction; and S7, reconstructing the prediction result, and outputting a final prediction value. According to the method, a combined prediction model based on feature screening and frequency division prediction is provided, signal components are extracted from original data, the signal components are separated into a high-frequency part and a low-frequency part, after feature screening is conducted on the signal components, the signal components are substituted into the model for prediction, and finally all signal component prediction results are reconstructed. And obtaining a final prediction value.

The invention provides a crustal stress prediction method based on multi-modal deep learning, and the method comprises the steps: S11, obtaining seismic data, horizon data and logging data of an oil and gas reservoir region, and carrying out the preprocessing of the obtained seismic data and horizon data; s12, extracting deep features from the seismic data by using a convolutional neural network, performing sequence modeling on the horizon data by using a bidirectional long-short term memory network, and capturing time sequence features of horizon changes; s13, constructing a multi-modal ground stress prediction model; and S14, applying the multi-modal crustal stress prediction model to the new seismic data and horizon data for crustal stress prediction, and generating a report for displaying a prediction result. By fusing the seismic data and the horizon data, constructing the multi-modal deep learning model and comprehensively utilizing the information of the multi-source data, more comprehensive crustal stress distribution information can be provided, comparison verification with the logging data can be carried out, the reliability of the prediction result is further improved, and higher-precision crustal stress prediction is realized.

The invention belongs to the technical field of electric power measurement, and provides a voltage transformer error prediction method and system based on spatial-temporal feature fusion, and the method comprises the steps: obtaining the state data of a voltage transformer; respectively extracting spatial features and time features of the obtained state data based on the graph structure to obtain space-time fusion features of the state data; the method comprises the following steps: extracting time sequence features of state data space-time fusion features by adopting a convolutional neural network-long short-term memory network, measuring the importance of the extracted time sequence features based on an attention mechanism to obtain a hidden layer state vector weight, and predicting a voltage transformer error according to the obtained time sequence feature weight and the predicted voltage transformer error. And completing error prediction of the voltage transformer based on spatio-temporal feature fusion.

The present invention discloses a lithium battery health diagnosis method and system based on an extensible recurrent neural network, comprising the following steps: performing wavelet decomposition on the voltage segment data near the switching segment of the fast charging stage to obtain the aging characteristics of the battery to be selected; performing correlation analysis on the aging characteristics of the battery to be selected to obtain an aging characteristic sequence; combining a long short-term memory neural network with a Gaussian process regression layer to form an extensible deep recurrent neural network model; inputting the aging characteristic sequence into the trained extensible deep recurrent neural network model to obtain the health state estimation and remaining service life estimation results of the lithium battery. The present invention includes two parts: the construction of a neural network structure and the implementation of a data-driven method, which can effectively extract features from battery data under fast charging conditions, and obtain high-precision health state estimation and remaining service life estimation results.

The application discloses a method for predicting rectangular narrow-gap re-submerged liquid droplet entrainment behavior of a plate-shaped element, which comprises the following steps: in an offline stage, constructing and initializing a BP neural network, training and optimizing weight values and threshold values by taking experimental boundary conditions as input variables and liquid entrainment rate as an output variable, and predicting in an online stage. The application can rapidly and accurately calculate the liquid entrainment rate value suitable for the re-submerged working condition of the bottom of the plate-shaped element by training the input-output relationship through a large amount of data by using the artificial neural network prediction method.

The invention relates to a common-mode current prediction method, a detection system, computing equipment and a medium. The common-mode current prediction method comprises the following steps: sleeving a magnetic ring on an electrically controlled direct current end and a three-phase end of an electric drive system to obtain a maximum external impedance, and then obtaining an equivalent common-mode current of the electric drive system when the maximum external impedance is set; enabling the electric control to be equivalent to a two-port model, and calculating a first equivalent noise source according to the maximum external impedance and an equivalent common-mode current of the electric control when the maximum external impedance is set; and under the condition that the magnetic ring is taken down and the electric drive system is provided with any external filter, predicting the common-mode current when any external filter is arranged according to the first equivalent noise source. By implementing the common-mode current prediction method provided by the invention, the problems of complex modeling and long modeling and prediction period in the prior art can be improved.

The present invention proposes a fire smoke temperature measurement method based on Kalman filtering and PID control to solve the technical problems that the existing smoke temperature measurement methods have hysteresis and sometimes no effective measurement. The present invention uses a temperature-sensitive fire detector to measure the fire ignition area in real time to obtain fire smoke temperature data; establishes a soft measurement mathematical model based on the obtained fire smoke temperature data; uses a PID controller to process the soft measurement model to control the smoke temperature signal; and predicts the temperature of the smoke temperature signal through Kalman filtering. The present invention uses the PID control technology based on the Kalman filter to optimally estimate the smoke temperature when a fire occurs, has a good control function on the smoke temperature signal, and can make the fire smoke temperature monitoring system reach a stable state relatively quickly. The present invention can quickly and accurately measure the fire smoke temperature, shorten the prediction time of the fire, and has good accuracy and timeliness.

The application provides a tool state detection method based on a main shaft current signal, comprising the following steps: designing an orthogonal experiment related to a cutting speed, a feed rate and a cutting depth; collecting a single-phase current of a main shaft motor driver by using a current sensor; collecting a wear amount of a rear tool face of a milling cutter after each experiment, fitting a tool wear curve according to the collected wear amount; performing denoising on the collected original signal of the single-phase current of the main shaft motor driver based on a third-fourth quantile method of a sliding window; extracting time-frequency domain features and time-frequency joint domain features of the single-phase current signal of the main shaft motor driver based on the preprocessed data; selecting n features from the time-frequency domain features and the time-frequency joint domain features as input of a prediction model by using grey correlation degree analysis; establishing a prediction model based on a least square support vector machine regression model, optimizing the prediction model by using a genetic algorithm, and predicting the tool wear amount based on the optimized prediction model.

The invention provides a rock-soil characteristic parameter prediction method based on a non-uniform normalization and spectrum strengthening depth fusion network, which comprises the following steps: step 1, performing data processing on rock-soil characteristic parameters by using a non-uniform normalization NUN method, the rock-soil characteristic parameters comprising a foundation bearing capacity characteristic value, a compression modulus and a rock-soil body state; changing the geotechnical characteristic parameters from a text type to an image type to obtain image data of the geotechnical characteristic parameters; step 2, training a spectrum enhanced deep fusion network SEDFN by using the image data obtained in the step 1; and step 3, predicting an actual geological profile map of a certain project by using the SEDFN trained in the step 2, and evaluating a prediction result by using a similarity index SI. According to the invention, the accuracy, reliability and precision of investigation work can be improved.

The application discloses a kind of reparameterization and stereoscopic coding brain disease classification method and system, it is related to computer medical image analysis technical field, including: S1.data acquisition step;S2.data processing step;S3.Establish RepBoTNet network model;S4.Establish RepBoTNet-CESA network model;S5.model training step;S6.classification step.The RepBoTNet-CESA network model constructed in the application combines the advantages that CNN network is good at capturing local information and transformer network is good at integrating global information, while obtaining excellent performance indicators, reduces the calculation cost;By constructing RepBoTNet-CESA network model, more rich and effective features can be extracted, and the region of interest is less likely to be lost, and when analyzing the whole brain structure, the accuracy of disease classification can be effectively improved.

The invention discloses a device and a method for identifying male and female information of ducklings based on a spectral analysis technology, and relates to a poultry male and female identification technology in the field of agricultural machinery. The device is characterized in that a duckling is placed on a sample collection platform (3), and an output end probe of a Y-shaped optical fiber (5) is aligned with a cloaca of the duckling; the spectrograph (6) and the halogen tungsten lamp light source (7) are respectively fixed at the lower part of the inner cavity of the box body (1) through a spectrograph fixing frame (9) and a light source fixing frame (10); the input end of the Y-shaped optical fiber (5) is respectively connected with a spectrograph (6) and a halogen tungsten lamp light source (7); and the spectrograph (6) is connected with the computer (8). The method comprises the following steps: (1) sample preparation; (2) spectrum acquisition; (3) data processing; (4) sample analysis; and (5) an identification result display interface. The method has the advantages of high precision, fast speed and no harm to ducklings; (2) an identification result display interface is visual and clear;

The invention discloses a cable displacement prediction system of a self-adaptive code spraying control system, and relates to the technical field of machining and automatic production control. The method comprises the following steps of: analyzing a cable cutting environment and a working condition of an ink-jet printer, distinguishing input characteristic parameters which influence a character identification jet printing effect of the ink-jet printer, training identification of the input characteristic parameters, and screening an influence factor screening model for cable displacement prediction from previous character displacement data; the filtered data are input into an ANFIS model, and the model is trained; the ANFIS model structure comprises an input layer, a fuzzification layer, a rule layer, a normalization layer and an output layer. According to the invention, the prediction precision of the cable displacement prediction model is improved, the code changing efficiency is improved, the deformation of the character identification after switching is reduced, and the system presents a better control effect.

The invention provides a power flow prediction method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring actually-measured meteorological data and actually-measured tidal current data of a to-be-measured station in at least one time period before a prediction time period; determining first astronomical tide data of the to-be-measured station in at least one previous time period based on the actually-measured meteorological data and the actually-measured tide data; according to the first astronomical tide data, determining second astronomical tide data of the to-be-measured station in the prediction time period; and obtaining predicted meteorological data of the to-be-measured station in the prediction time period, and determining predicted tidal current data of the to-be-measured station in the prediction time period based on the predicted meteorological data and the second astronomical tide data. According to the invention, the timeliness of tide forecasting can be improved.

The present application relates to a kind of Helmholtz resonance material sound absorption performance prediction method, comprising: based on the sound absorption performance database between the structure parameters of different dimensions of Helmholtz resonance material and sound absorption coefficient, obtain training set and verification set;Using the data in training set trains the Helmholtz resonance material sound absorption performance prediction network based on machine learning, test verification is carried out using verification set, obtain Helmholtz resonance material sound absorption performance prediction model;The structure parameters corresponding to different dimensions of the Helmholtz resonance material to be measured are input into Helmholtz resonance material sound absorption performance prediction model, obtain the predicted value of the sound absorption coefficient of different dimensions of Helmholtz resonance material.Compared with prior art, the present application solves the problem that it is difficult to predict sound absorption performance by simulation technology due to too many parameters of acoustic Helmholtz resonance material, solves the problem of long simulation prediction time and low efficiency, and can efficiently and accurately predict the sound absorption coefficient of different dimensions of Helmholtz resonance material.

The present application relates to a kind of short-term wind speed prediction method based on secondary decomposition and hybrid network, first using the integrated mode decomposition of adaptive noise technique, wind speed historical data is decomposed into multiple components, using sample entropy and heat map for analysis to different components, it is divided into high-frequency component, mid-frequency component, low-frequency component, high correlation component.And then using singular spectrum analysis, high-frequency component is further decomposed into subcomponent.Prediction of subcomponent, mid-frequency component, key component uses the hybrid network proposed, is predicted, to ensure overall prediction accuracy;Low-frequency component is predicted using extreme learning machine, to further improve operation efficiency, finally, all prediction results are superimposed to obtain wind speed prediction result.

The invention relates to a battery processing method and device, electronic equipment, a storage medium and a program product, and the method comprises the steps: obtaining a target internal resistance based on a target capacity retention ratio of a battery and a first mapping relation; wherein the first mapping relation is used for indicating a corresponding relation between the capacity retention ratio and the internal resistance; based on the target internal resistance and a second mapping relation, obtaining the number of available charge and discharge cycles of the battery; wherein the second mapping relation is used for indicating the corresponding relation between the number of charge and discharge cycles and the internal resistance, so that the cycle life of the battery is accurately predicted, and meanwhile, the prediction time of the cycle life of the battery is shortened.

The invention provides a target detection method, device and equipment, a computer readable storage medium and a computer program product. The method comprises the steps that three-dimensional data to be detected and a trained three-dimensional detection model are acquired, and the trained three-dimensional detection model is obtained by training the three-dimensional detection model to be trained through three-dimensional sample data; the first model parameter of the to-be-trained three-dimensional detection model is obtained by fusing the second model parameter of the trained two-dimensional detection model; wherein the first model parameter is a model parameter of a to-be-trained three-dimensional detection model, the trained two-dimensional detection model can perform target detection on a two-dimensional image, and the second model parameter is a model parameter of the trained two-dimensional detection model; performing feature extraction on to-be-detected three-dimensional data by using the trained three-dimensional detection model to obtain feature information; and performing prediction processing on the to-be-detected three-dimensional data according to the feature information to obtain a target detection result. According to the invention, the target detection efficiency can be improved.

The invention discloses an efficient detection system and method based on S-D network full-level perception, the system comprises a backbone network, a neck network and a head network, the neck network comprises a full-level feature sharing module (FLIS) and a detailed information extraction module (DFE); the FLIS module designs branches of a first hierarchy and a second hierarchy according to different features, information of all the hierarchies is aligned, fused and then injected by using the thought of structural reparameterization and Transformer to enable all the hierarchies to obtain global information, and cross-layer information flow is enhanced. The DFE module extracts detailed information by using three-feature fusion mapping, and then fuses and outputs the multi-scale information and the detailed information by using a channel and position attention mechanism; according to the S-D neck network, the multi-scale feature fusion capability is improved, and the ideal balance between the rate and the accuracy is realized on all model scales; the precision of an experimental result in a VOC data set is improved by 3.1% compared with that of an original SOTA model, the calculated amount of the model is reduced by 7%, and meanwhile the optimal characteristic is achieved in a generalization verification vidrone data set.

The application discloses a kind of based on deep neural network's root canal therapy postoperative tooth mechanics performance fast, accurate prediction method.The method is first based on microcomputer tomography image, comprehensively utilizes Mimics Research 21.0, Geomagic Studio 2013 and SolidWorks 2017 software, reconstructs the high-precision, multi-component three-dimensional entity model of root canal therapy postoperative tooth.On this basis, using Python script in Abaqus 2022 software carries out parameterized finite element simulation, batch calculates the mechanical response of postoperative tooth under different pulp cavity filling materials.Subsequently, using the standardized data set containing integral point space coordinates, stress and pulp cavity filling material attributes, three deep fully connected neural networks with independent architecture design are trained.The method shortens the prediction time of single stress distribution from tens of seconds of finite element analysis to tens of milliseconds while ensuring high consistency between the prediction results and the finite element analysis results, realizes thousands of times of calculation acceleration, and lays a technical foundation for scientific and personalized diagnosis and treatment plan.

The present invention provides a high-precision trajectory extrapolation method and system based on an instantaneous inertial coordinate system, the method comprising: according to the prior information conditions obtained by passive radar detection, accurately considering the influence of earth gravity, earth J2 gravity, atmospheric damping and wind, and modeling the ballistic motion equation based on the geocentric instantaneous inertial coordinate system; during active radar detection, converting the radar measurement coordinate system to the geocentric instantaneous inertial coordinate system to obtain projectile measurement data; performing polynomial fitting on the projectile measurement data and the ballistic motion model, deleting points with large fitting errors in the projectile measurement data, performing reverse processing on time and speed, and using the reversely processed data to perform filtered trajectory extrapolation to reduce the trajectory extrapolation error. The present invention accurately models multiple forces, uses an inertial coordinate system to remove the influence of Coriolis inertial force on the projectile motion trajectory; deletes abnormal data through polynomial fitting, uses reverse filtering to improve reconnaissance positioning accuracy, and improves trajectory extrapolation positioning accuracy.

This application discloses a liveness detection method, a training method for a liveness detection model, and a corresponding apparatus. The main technical solution includes: acquiring a first image and a second image of a target object using a first camera and a second camera; obtaining depth map information based on the disparity between the first image and the second image; extracting image features and depth map features from the first image and the depth map information respectively using a liveness detection model; fusing the image features and depth map features to obtain multimodal fusion features; and using the multimodal fusion features to obtain a detection result indicating whether the target object is live. This application can effectively improve the accuracy of liveness detection.

The present invention provides a method for rapid tumor localization based on full-field pathology slice images with multiple resolutions, including: constructing an image block training set; using the image block training set to obtain two classification models; and using the two classification models in succession to predict full-field pathology slices to obtain positioning results. The present invention trains two models of preset resolutions respectively while ensuring that image detail information is not sufficiently lost, one classification model uses a faster full convolutional network, and the other classification model adds an efficient feature extraction module. Compared with the traditional framework that uses a single resolution for positioning, it achieves rapid positioning of tumor areas on full-field pathology slices while ensuring positioning accuracy.

The invention belongs to the technical field of multi-source power prediction, and discloses a multi-source power prediction comparative analysis method based on artificial intelligence, which realizes accurate prediction and comparative analysis of photovoltaic power generation power by fusing multiple data sources and an artificial intelligence technology. The technical scheme has the advantages of comprehensive data, high prediction precision, strong model generalization ability and the like, and can be widely applied to the fields of photovoltaic power stations, power grid dispatching centers, energy management companies and the like. And multi-source data fusion and intelligent analysis: carrying out fusion and intelligent analysis on multiple data sources to realize comprehensive prediction and comparative analysis on the photovoltaic power generation power. And multiple artificial intelligence algorithms are fused: the multiple artificial intelligence algorithms are adopted to construct a prediction model, an optimal prediction model is selected through comparative analysis, and prediction precision and generalization ability are improved. Model fusion and result optimization: carrying out fusion and optimization processing on prediction results of a plurality of prediction models, and further improving the accuracy and reliability of the prediction results.

The invention provides a tomato shelf life prediction method based on a hyperspectral and RGB imaging technology. The tomato shelf life prediction method comprises the following steps: S1, obtaining a hyperspectral image and an RGB image of a tomato sample; s2, preprocessing the hyperspectral image to obtain hyperspectral feature data; s3, extracting color feature data and texture feature data of the RGB image; s4, splicing the data obtained in the steps S2 and S3, performing standardization processing, and dividing the data into a training set, a test set and a verification set; s5, constructing an MLP model; screening the target feature set; s6, training the MLP model to obtain a tomato shelf life identification model; s6, obtaining a hyperspectral image and an RGB image of the detected tomato, performing preprocessing, and screening data according to the target feature set; and S7, splicing the screened color feature data, texture feature data and hyperspectral feature data of the detected tomatoes, inputting the spliced data into the tomato shelf life recognition model for recognition, and outputting the shelf life of the detected tomatoes. According to the invention, the accuracy of tomato shelf life identification is improved.

The application discloses a kind of comprehensive energy system multivariate load prediction method, comprising: obtaining comprehensive energy system data, according to the correlation analysis of each meteorological factor and load to comprehensive energy system data, and strong correlation history meteorological factor data is screened out;Build multivariate load coupling feature matrix;According to meteorological factor data, multivariate load coupling feature matrix is clustered using adaptive K-means clustering algorithm;The multivariate load coupling feature matrix after clustering is input into the Attention-BiGRU prediction model pre-trained to predict, and the load prediction result is obtained.Considering meteorological factors and the cross-coupling characteristics of multivariate load, the model complexity and prediction time are reduced, so that the final prediction result takes into account stability and accuracy.

The invention provides a three-vector model prediction direct power control method and system for a doubly-fed asynchronous motor and a computer storage medium, and the method comprises the steps: calculating the negative conjugate of the complex power of a stator side according to the voltage and flux linkage equation of the doubly-fed asynchronous motor, and selecting three optimal voltage vectors on this basis to eliminate the power error; by adopting the constructed voltage vector relation graph, the difference value between the negative conjugate of the complex power at the next moment and the reference value of the complex power is calculated, a zero voltage vector correlation component and an effective voltage vector correlation component are defined, and then the action time and the duty ratio of each vector are calculated. And transmitting a switching signal corresponding to the action time and the duty ratio of each vector to a switching tube in the converter so as to control the power of the doubly-fed asynchronous motor. According to the method, the reasonable selection of the negative conjugate error component of the stator side complex power and the simplification of the calculation process of different voltage vector action time are realized, so that the prediction time is shortened.

This invention provides a short-term power grid load forecasting method based on the PSO-LSSVM algorithm, comprising: performing modal decomposition on the load data sequence to obtain multiple intrinsic mode function components; removing noise components from the intrinsic mode function components to obtain effective components; constructing a nonlinear prediction model for the effective components based on least squares support vector machine; processing the nonlinear prediction model through structure minimization to obtain a load forecasting model; optimizing the load forecasting model based on particle swarm optimization algorithm to obtain a hybrid prediction model; and predicting the power grid load characteristic data at the time of prediction based on the hybrid prediction model to obtain the predicted power grid load. By removing high-frequency noise from the power grid load data, the nonlinear variation trend of the short-term power grid load can be accurately determined. The optimization of the load forecasting model based on particle swarm optimization algorithm improves prediction accuracy and shortens prediction time.

The invention relates to the technical field of composite material damage evaluation and performance prediction, and discloses a method and a system for rapidly predicting thermal-mechanical-electrical response of a quartz / epoxy resin composite material under laser irradiation based on a vector quantization variational auto-encoder (VQ-VAE). According to the method, a cascade deep learning neural network architecture of a DNN prediction network and a VQ-VAE decoder is constructed, laser working condition parameters are mapped to a discrete potential space, and complete temperature field, stress field and dielectric constant parameters are rapidly reconstructed. Wherein the VQ-VAE adopts a vector quantization mechanism to extract material ablation phase change and damage features, the visual field resolution is adjusted by improving the size of a convolutional layer in an encoder, and extraction of small-scale damage features is achieved. According to the method, second-level forecasting from parameter input to full-field output is achieved, the calculation efficiency and the forecasting precision are remarkably improved, and the method is suitable for rapid performance evaluation and design optimization of the composite material under laser irradiation.