11results about How to "Fast prediction" patented technology

The application relates to the technical field of artificial intelligence optical character recognition, in particular to an end-to-end communication big data journey card recognition method, and the steps of the method comprise the following steps: adjusting and correcting a text box according to size information of the text box, so that the size of the text box meets the required angle and proportion of adjustment; performing separated feature extraction on the corrected text box through a preset neural network to obtain feature data of a font; and obtaining text data corresponding to the feature data through a Bi-GRU recurrent neural network combined with an activation function.

The invention discloses a method for predicting an impact displacement time history curve of a concrete filled steel tubular column under the influence of defect coupling, comprising the following steps: S1, acquiring a data set: acquiring test data and verified finite element numerical simulation data, the data set sample comprising structural characteristic parameters of a component and a corresponding displacement time history curve under transverse impact, s2, data preprocessing: extracting geometric, material, defect and impact working condition parameters of the component to form an input feature vector, standardizing input features, performing time alignment and resampling on a displacement time history, and constructing a time sequence input tensor of a unified dimension; s3, a CNN-LSTM-Attention model is established, and the CNN-LSTM- S4, training and optimizing the model by using sample data, and taking a mean square error as a loss function; s5, target component parameters are input, and an impact displacement time history prediction result is output. According to the method, the impact displacement time history of the concrete filled steel tubular column containing multiple defects can be accurately and efficiently predicted, and the impact resistance evaluation efficiency of the structure is improved.

This invention discloses a rapid and refined forecasting method for marine environmental parameters based on graph neural networks. Using data from Argo historical observations and the Dynamic Analysis of the Marine Environment (MODAS) system as datasets, a graph neural network is employed to predict future marine environmental parameters (underwater 3D temperature and salinity, 3D acoustic field). Through online learning and numerical assimilation, the forecasting system is corrected based on environmental information from a limited dataset without affecting the normal operation of the network model. A ray casting algorithm from volume rendering techniques is used to perform 3D visualization mapping of the marine environmental parameters. This invention introduces graph neural networks for marine environmental parameter forecasting, improving the timeliness and accuracy of forecasts and enabling rapid and refined forecasting of marine environmental parameters, thus supporting their application in marine environmental protection.

This invention relates to the field of automated detection technology, and more particularly to a method and system for predicting ferrous oxide content during sintering. The invention first acquires original images, ferrous oxide content data, and environmental parameters during the fracture process of the sintered product. The best frame, best representing the cross-sectional characteristics, is selected as the original image. The original image is preprocessed, including grayscale processing and contrast enhancement. An SVM model and an improved BP neural network model are constructed and trained respectively to obtain a hierarchical model and multiple ferrous oxide content prediction models. In the testing phase, images from one cycle are acquired, and the best frame image is obtained through optimal frame processing. The hierarchical model is used to classify the images, and the corresponding ferrous oxide content prediction model is selected based on the classification results to predict the ferrous oxide content in the sintered product. Through multi-level training and model construction, and by fully utilizing image features and environmental parameters, the prediction accuracy and precision are improved.

This invention discloses a method, system, and medium for predicting and controlling the normalized microstructure of oriented electrical steel based on a multilayer perceptron deep learning network and cellular automata. It takes two-dimensional normalized process parameters as input and ODF slice images as output. A cellular automata model driven by physical mechanisms generates virtual labeled data of "(t,T)-ODF" in batches within a preset process design space, constructing a dataset. Then, a deep learning surrogate model is used to achieve an end-to-end nonlinear mapping from the two-dimensional normalized process parameters to the ODF slice images, thereby achieving rapid prediction and visualization of the normalized microstructure without repeatedly calling the cellular automata model. This invention solves the key challenge of balancing prediction speed, structural accuracy, and inversion design capabilities within an extremely short process window.

This invention discloses a method for constructing and predicting a relay health index based on Mahalanobis distance and multi-channel information fusion. The method includes: first, sampling the coil drive current signal sequence and contact voltage signal sequence during a single relay cycle, extracting the peak and trough time points and amplitudes of each sequence; then defining the first 10% of the relay's cycles as the healthy period, extracting the relative deviation between the current cycle feature value and the mean feature value of the healthy period, further calculating the Mahalanobis distance of the feature vector relative to the health status benchmark, and sequentially performing Box-Cox transformation, normalization, and monotonicity mapping to obtain the final health index. Using [variable name] as input and [variable name] as output, a random forest is selected as the regression prediction model for model training, and online prediction is achieved. The HI curve output by this invention avoids severe oscillations, greatly improving the stability of the prediction results and its engineering application value. Furthermore, the model is lightweight and suitable for edge deployment.

The application discloses a physical information multi-fidelity neural network three-dimensional cascade flow field reconstruction method based on Fourier characteristics, and relates to the field of aero-engine turbine machinery aerodynamic design. The method comprises the following steps: generating high / low fidelity data sets with boundary consistent and gradient perception sampling; constructing a multi-fidelity neural network containing Fourier characteristic embedding, multi-fidelity collaborative network and RANS equation residual constraint; adopting a two-stage training strategy, first optimizing data fitting loss to learn the flow field mapping relationship, and then jointly improving the conservation with physical loss; finally, realizing fast, high-precision and physically consistent reconstruction of three-dimensional cascade space coordinates to key physical quantities such as velocity, static pressure and temperature. The method significantly reduces the dependence on high-fidelity data, improves the prediction accuracy in high gradient areas, and is suitable for rapid iterative design of aero-engine cascades.

The application discloses a power station fan state prediction method based on clustering dynamic division of time series, and belongs to the technical field of power station fan state prediction. In an offline state, monitoring parameter data capable of representing a fan operation state is utilized, a k-means clustering algorithm is used to realize fan state classification and mark original data, and a relevant state prediction model is trained according to the marked data. In online prediction, data sequences are simultaneously input into each state prediction model, initial values of weight coefficients are set according to the data classification in the input sequences, the deviation between final calculation values and actual observation values is taken as an index, and a search algorithm is used to optimize the weight coefficients, so that accurate power station fan state prediction is realized.

This invention relates to the field of tire performance prediction technology based on artificial intelligence, and more particularly to a method and system for predicting tire inflation profiles based on machine learning. The method acquires the structural parameters, material parameters, and actual inflation profile data of a sample tire, extracting the partitioning mechanism parameters of the crown, shoulder, sidewall, and heel regions; unifies the profiles to the rim reference coordinate system and standardizes the partitions, constructing local reference profiles and local basis vectors for each partition; establishes a multi-output machine learning model with partitioning mechanism parameters as input and local basis vector coefficients and key geometric quantities as outputs to complete the prediction of the target tire inflation profile, and outputs the final profile through geometric feasibility verification and local projection correction. This invention can significantly shorten the tire inflation profile prediction time, improve prediction accuracy and engineering applicability, and is suitable for rapid tire design and optimization.

The application relates to a classification method, device and equipment for pathological images and a storage medium. The method comprises the following steps: processing a full-view digital pathological image to be detected to obtain a down-sampling image block; constructing a distillation and adversarial network, training the distillation and adversarial network, and extracting a classification sub-network from the trained distillation and adversarial network; inputting the down-sampling image block into the classification sub-network for classification to obtain an initial classification result; and obtaining a classification result corresponding to the full-view digital pathological image to be detected based on the initial classification result. According to the method, small data amount down-sampling image blocks can be input into a light classification sub-network to quickly and efficiently classify corresponding pathological images.

This invention relates to the field of natural language processing technology, providing an end-to-end multi-turn dialogue rewriting method, system, and storage medium that integrates dialogue detection. In the proposed end-to-end multi-turn dialogue rewriting method, a joint learning approach is used to train the detector and generator, solving the problems of cascading errors and incoherent sentences after matching and filling in the Pipeline method, thus improving the detector's detection performance. By having the detector ignore sentences that do not need rewriting and integrating dialogue detection information into the generator, the slow speed and repetitive encoding problems of the Seq2Seq method are solved. The dialogue rewriting process of this invention does not require the generator to repeatedly encode the text, improving the model's prediction speed. Simultaneously, integrating dialogue detection information into the generator improves the generation quality of the rewritten sentences.