92 results about "Hybrid neural network" patented technology

The application provides an uplink sensing integrated system signal processing method and device and a storage medium, comprising: performing coarse synchronization on a first signal received by a plurality of antennas to obtain a first time offset, a first carrier frequency offset and a second signal after coarse synchronization; performing preliminary channel estimation according to the second signal to obtain an estimated first channel state information tensor; performing deep learning based on a hybrid neural network according to the first channel state information tensor to obtain an estimated second channel state information tensor, a second time offset and a second carrier frequency offset; performing tensor decomposition according to the second channel state information tensor, and acquiring sensing information according to the first time offset, the first carrier frequency offset, the second time offset and / or the second carrier frequency offset. The application realizes more accurate synchronization error estimation and channel state information denoising, guarantees more reliable communication performance, and realizes high-precision positioning and speed measurement of sensing.

This invention belongs to the field of power equipment monitoring technology, specifically relating to a non-intrusive monitoring system and method for switchgear fault indicators. The system includes: a non-contact optical sensing unit, installed on the outside of the switchgear panel via a magnetic structure, for acquiring the brightness timing signal of the fault indicator light; an electrical parameter acquisition unit, for acquiring the electrical parameter timing signal of the circuit associated with the fault indicator light; and an edge computing unit, deploying a fault diagnosis neural network model with a hybrid neural network architecture, for performing feature fusion processing on the two time-aligned signals, outputting a status recognition result to distinguish between correct alarms caused by line faults and false alarms caused by abnormalities in the indicator light itself. This invention achieves non-intrusive retrofitting of fault indicators in old switchgear, converting local light signals into remotely transmittable digital quantities without dismantling or modifying the cabinet or electrical circuits, and effectively identifying true and false alarms through multi-source signal fusion analysis.

This invention provides an EEG emotion recognition method based on gating spatiotemporal mechanisms and a bidirectional state-space model, relating to the fields of biomedical engineering and artificial intelligence. The EEG emotion recognition method includes: acquiring shallow features of an EEG signal; projecting the shallow features to obtain corresponding high-dimensional features; performing hierarchical gating processing on the high-dimensional features in both temporal and spatial dimensions to obtain spatiotemporal aggregated features; performing bidirectional state-space encoding on the spatiotemporal aggregated features to obtain fusion features; and performing weighted processing on the fusion features to obtain the emotion classification result of the EEG signal. The EEG emotion recognition method provided by this invention can effectively filter out local noise and efficiently model long-term and long-distance dependencies, constructing a computationally lightweight hybrid neural network architecture to meet the requirements of real-time and robust EEG emotion recognition.

The intelligent home of the present application relates to the technical field of intelligent home and medical care, in particular to an intelligent anti-falling mattress device with active buffering function, which solves the problems of passive, low precision and the like of the existing bed falling protection. The core innovation is the deep integration of lightweight end-side intelligent algorithm and mechanical execution structure, data is collected through a non-uniform layout flexible pressure sensor array, and a CNN-LSTM-AM hybrid neural network model is relied on to realize real-time inference and decision of the bed falling risk end-side, trigger the electromagnetic buckle to release the spring to pop out the telescopic arm to form a protective barrier, synchronously upload the alarm information, realize the upgrade from passive monitoring to active prediction-immediate protection-synchronous alarm, the end-side algorithm response reaches the level of hundreds of milliseconds, and it is suitable for the elderly and people with difficulty in movement.

This invention discloses a cloud removal method based on hybrid neural networks fusing optical and SAR images, belonging to the field of image processing technology. The method involves acquiring a dataset and preprocessing it, then dividing it into training and validation sets according to a predetermined ratio. A cloud removal model based on hybrid neural networks fusing optical and SAR images is constructed, mainly composed of a lightweight fusion residual block, an asymmetric feature pyramid module, a parallel gated attention module, and a cloud mask-guided attention fusion module. The preprocessed training and validation datasets are input into the cloud removal model for training and validation. The loss function is calculated, backpropagation is performed, network parameters are updated, and the optimal parameter model is obtained. The preprocessed test set is input into the trained optimal parameter model, and the cloud-removed image is output. This invention achieves high-quality restoration of cloud-covered areas and preservation of ground feature characteristics by jointly utilizing optical image information and SAR image characteristics.

The present application relates to the technical field of flexible touch screen detection, and specifically discloses a flexible touch screen polarization characteristic detection method based on multi-modal data fusion. The method comprises the following steps: building a multi-parameter measurement optical path system, collecting speckle characteristic data, interferometric polarization phase information data and spectral characteristic data after the flexible touch screen is transmitted; generating a multi-modal fusion feature vector through time-space alignment and feature extraction; constructing a CNN-LSTM hybrid neural network model; dynamically adjusting the parameters of the CNN-LSTM hybrid neural network model by using an adaptive algorithm; inputting the multi-modal fusion feature vector into the adjusted CNN-LSTM hybrid neural network model, and outputting the polarization degree, the extinction ratio and the polarization axis direction parameters. Through the multi-parameter measurement optical path system cooperative detection, the multi-modal data fusion and the adaptive algorithm, the present application solves the polarization characteristic drift problem caused by the deformation of the flexible screen, and significantly improves the detection accuracy and stability.

This disclosure relates to a device, method, equipment, storage medium, and program product for identifying abnormal development of the maxillofacial region in children. The device includes: a three-dimensional image registration module for performing global rigid registration and local non-rigid registration of three-dimensional images of facial soft tissue and jawbone; a geometric and visual feature extraction module for inputting spatially aligned three-dimensional images of facial soft tissue and jawbone into a hybrid neural network to obtain geometric and visual features; a geometric and visual feature fusion module for inputting geometric and visual features into an attention-based feature fusion network to obtain maxillofacial fusion features; and a child maxillofacial abnormal development identification module for inputting the maxillofacial fusion features into a child maxillofacial abnormal development identification model to obtain the identification result of child maxillofacial abnormal development. This reduces the user experience requirement, avoids the influence of subjective factors, and identifies early and subtle abnormalities.

This application provides a method and system for visual defect detection of capsule materials based on a hybrid neural network. The method includes simultaneously inputting the acquired visual image of the capsule material into parallel local feature extraction branches and global feature extraction branches; inputting the local feature map and global feature vector into a gated cross-attention interactive fusion module for bidirectional interaction and fusion to generate a fused feature vector; and inputting the fused feature vector into a classification head for classification and recognition, outputting the defect category of the capsule material. This application addresses the fundamental differences in feature levels between "damage" and "wrinkles" on the surface of capsule materials, designing a parallel hybrid model that can consider both local and global features and perform feature complementarity and intelligent fusion. This solves the core problems of low accuracy, poor generalization ability, and difficulty in real-time deployment faced by surface defect detection in the automated production process of capsule materials.

A lithium ion battery state of health estimation method based on domain generalization of adversarial learning belongs to the technical field of lithium ion battery health management. First, incremental capacity curve features are extracted from a multi-source heterogeneous data set. Structural adaptive alignment processing is performed on the incremental capacity curve, including amplitude normalization and voltage axis normalization, to map the heterogeneous features to a unified representation space. A hybrid neural network of a residual convolutional network and a Transformer encoder is constructed as a feature extractor to jointly capture the local morphological features and global degradation dependencies of the incremental capacity curve. A domain generalization method based on an adversarial mechanism is used to learn the degradation representation that is invariant to the domain through adversarial training of the multi-domain discriminator and the feature extractor. The SOH regression loss and the adversarial domain classification loss are jointly optimized, and the end-to-end adversarial training is realized through the gradient reversal layer. Finally, zero-shot SOH estimation is realized for unobserved heterogeneous battery data sets. The present application realizes high-precision zero-shot state of health estimation.

The application discloses a CNN-LSTM-based phase calibration method for mode division multiplexing optical fiber communication and belongs to the technical field of optical fiber communication. By constructing a double-input CNN-LSTM hybrid neural network architecture, MIMO input signals and original target signals of mode division multiplexing optical fiber communication are acquired, data preprocessing is carried out, the spatial feature extraction capability of a convolutional neural network and the time series modeling capability of a long short-term memory network are combined, effective compensation of phase noise in long-distance strong coupling mode division multiplexing optical fiber communication is realized, complex MIMO signals are separated into real and imaginary parts and sliding window training samples are generated, and a double-branch network containing an image input layer, a convolutional layer, an LSTM layer, a deep connection layer and a self-defined phase calibration layer is constructed. The problems that the phase noise compensation capability of a traditional algorithm is limited and the convergence performance is poor under a strong coupling condition are effectively solved, and the equalization performance and transmission quality of a mode division multiplexing optical fiber communication system are significantly improved.

A hybrid neural network based adaptive prediction method of rate of penetration (ROP) is proposed. Firstly, the data of multiple wells during drilling operation are collected, and then the data are preprocessed and the data samples are constructed by sliding window. On this basis, a hybrid neural network composed of LSTM branch and MLP branch is constructed, in which the LSTM branch is used to extract the time series information of historical multi-step drilling condition features, and the MLP branch is used to model the future multi-step controllable parameter features, and the multi-step prediction of ROP is realized through feature fusion. The model is pre-trained by using the data of multiple training wells, and the drift detection and adaptive fine-tuning mechanism based on prediction error change is introduced on the data of target well to realize the online update of model parameters. This method can effectively improve the accuracy and adaptability of ROP prediction under cross-well conditions.

This invention relates to the field of temperature field prediction for vehicle transmission components, and discloses a method, system, medium, and device for predicting the temperature field of wet friction components based on a cross-domain transfer learning model. The method includes: acquiring temperature field data and interface morphology data of the friction components, constructing a Transformer-LSTM-AdaBoost hybrid neural network model using heterogeneous physical quantity corresponding sample inputs, and pre-training the model as a feature extractor; constructing a DGDAN cross-domain transfer model, using interface morphology data as the source domain and temperature field data as the target domain, and inputting both into the DGDAN cross-domain transfer model for training, initially extracting common features from the two sets of data through the feature extractor; conducting adversarial training between the feature extractor and the domain discriminator through a gradient inversion layer, explicitly aligning the feature distributions of the two domains using the maximum mean difference metric, and obtaining the final common features; and inferring the full surface temperature field distribution online through the input interface morphology data, achieving dynamic state synchronization between the physical entity and the virtual model.

The application discloses a recommendation model training and application method and device based on multi-platform behaviors, and a medium, relates to the technical field of user recommendation, and the method comprises the steps of constructing a mixed neural network model, the mixed neural network model comprising a feature embedding module, a multiple-hole convolution module, a multi-level attention mechanism module and a prediction module; the feature embedding module processes a multi-platform behavior sequence to obtain a feature embedding vector; the multiple-hole convolution module processes the feature embedding vector to obtain a multi-scale feature vector; the multi-level attention mechanism module processes the multi-scale feature vector and a commodity feature vector to obtain a user interest vector; the prediction module predicts a predicted click rate of a user on a candidate commodity based on the user interest vector and the commodity feature vector; the mixed neural network model is trained to obtain a recommendation model; and the recommendation model is used for recommendation. The application can train a recommendation model with better performance and improve recommendation performance.

The application discloses an Internet of Things intelligent road network safety monitoring system based on distributed optical fiber sensing, which comprises a sensing module, a transmission module, a processing module and a disease development prediction module connected in sequence. The sensing module comprises a sensing optical fiber network, a light source submodule and a photoelectric detection submodule. The transmission module is used for uploading the electrical signal to the processing module. The processing module is used for processing the electrical signal, and performing disease classification identification and health state evaluation. The disease development prediction module is used for predicting the disease development and providing a maintenance scheme. The application fuses a lightweight deep learning model, a hybrid neural network algorithm and a Monte Carlo simulation technology to construct the Internet of Things intelligent road network safety monitoring system based on distributed optical fiber sensing, realizes efficient processing of sensing data, accurate identification and prediction of diseases and cost benefit optimization of the maintenance scheme, and solves the pain points of the prior art.

This application relates to the field of big data technology, and provides a method and apparatus for auditing shared data. The method includes: obtaining an audit model based on a trained neural network model; and performing audit processing on the shared data based on the audit model. The trained neural network model is obtained through the following steps: establishing a first hybrid neural network model based on shared data samples; training the semantic annotations in the first hybrid neural network model based on manual annotation results to obtain a second hybrid neural network model; and updating the network weights of the second hybrid neural network model based on a keyword sample set to obtain the trained neural network model. This application improves the efficiency and accuracy of auditing shared data by combining a neural network model with auditing technology.

The application belongs to the technical field of fault monitoring of feature equipment, and relates to an elevator video retrieval and identification method and system based on an improved neural network, and the technical points are as follows: real-time original video data is collected through a monitoring device, preprocessed, and video frame data is obtained; in combination with physical constraint conditions, feature screening is performed on the video frame data, the screened effective feature area is input into a twin liquid mixed neural network, feature extraction, similarity matching and time sequence optimization are completed, and the final retrieval and identification result is obtained. The application designs a device-specific physical constraint feature screening mechanism for the feature equipment such as the elevator, and eliminates interference features; a mixed neural network model is used to realize accurate extraction of device features, similarity matching and time sequence correlation capture; a complete video retrieval and identification system is constructed to form a complete process technical scheme from data collection to result application, and high-precision and high-efficiency retrieval and identification of elevator videos are realized.

This invention relates to the field of privacy-preserving machine learning, and particularly to a privacy-preserving data processing method and prediction system. The method includes: obtaining an encryption priority value obtained by weighted fusion of at least two complementary encryption priority quantification indices; dividing the original feature vector into a first feature subset and a second feature subset according to the encryption priority value; receiving plaintext data of the first feature subset and homomorphically encrypted data of the second feature subset from a data provider; inputting the plaintext data into a first neural network subnet to extract a first hidden layer representation; inputting the homomorphically encrypted data into a second neural network subnet; fusing the first and second hidden layer representations in the ciphertext domain, and outputting an encryption prediction result. This invention reduces computational overhead while protecting the privacy of sensitive features by quantifying encryption requirements in multiple dimensions and dynamically dividing features, combined with a hybrid neural network of plaintext and homomorphic encryption.

This invention discloses a dynamic gain-controlled optoelectronic hybrid neural network computing system for low-altitude UAV rubble search and rescue, comprising a UAV camera, a convolutional neural network preprocessing module, a dynamic range adaptive matching module, a matrix calculation mapping control module, a weighted DAC module, an optical matrix calculation chip, a scaling restoration module, and a convolutional neural network post-processing module. The system identifies weak feature regions and performs adaptive gain amplification through the dynamic range adaptive matching module. The matrix calculation mapping control module adapts the optical calculation format, the optical matrix calculation chip performs high-speed, low-power matrix operations, and the scaling restoration module restores the true signal amplitude, ultimately outputting the target recognition result. This invention improves the accuracy of weak vital sign detection, reduces end-to-end latency and power consumption, adapts to UAV SWaP constraints, and is suitable for low-altitude UAV rubble search and rescue scenarios.

This invention provides a radar signal modulation recognition method based on temporal modeling and cross-variable fusion, comprising: preprocessing the radar IQ signal to be identified into blocks to obtain block tensors; constructing a hybrid neural network model, which includes a block transform temporal encoder, a modern temporal convolutional network cross-variable fusion module, and a classification head. The block transform temporal encoder is used to capture long-range temporal dependencies, and the modern temporal convolutional network cross-variable fusion module is used to perform cross-variable fusion on the extracted temporal feature tensors; inputting the block tensors into the trained hybrid neural network model for processing using the block transform temporal encoder, the modern temporal convolutional network cross-variable fusion module, and the classification head, and outputting radar signal modulation recognition results. This improves the model's adaptability and generalization ability; compensates for the shortcomings of local segmentation methods in global modeling capabilities; and achieves joint feature extraction of multi-channel radar signals.

The application discloses a complex terrain wind field simulation method and system based on physical mechanism empowerment, and the method comprises the following steps: obtaining target area digital elevation model data and WRF-CFD coupling simulation wind field data; screening high-quality training labels through terrain-wind field kinetic energy matching degree and other physical indexes; extracting and utilizing static terrain features weighted and optimized by terrain influence potential indexes, and dynamic inflow features from boundary conditions; constructing a hybrid neural network model, and training by using a loss function containing a physical constraint regularization term; and finally inputting a feature vector to output a high-precision wind field. The application deeply integrates physical mechanisms into data, features, models and loss functions, realizes the collaborative improvement of the simulation accuracy, efficiency and physical credibility of the complex terrain wind field, and is suitable for wind energy evaluation, power grid wind resistance design and other scenes.

This invention belongs to the interdisciplinary field of artificial intelligence and visual inspection, specifically disclosing a deep learning-based intelligent method for detecting concrete slump. The method includes: acquiring a thermal infrared video sequence of the concrete slump process; extracting three-dimensional spatiotemporal temperature field data; generating a multi-channel dynamic feature map containing temperature change rate, gradient magnitude, and entropy value; inputting the map into a convolutional-temporal hybrid neural network model for slump prediction; outputting the results and evaluating the confidence level, triggering retesting if necessary. By integrating thermodynamic evolution features with deep learning modeling, this invention can predict slump with high accuracy and automation, improving detection robustness and engineering applicability, and meeting the real-time quality control needs of construction sites.

The building cold load prediction method based on physical simulation and a hybrid neural network of the present application designs a building cold load prediction system, trains the system by combining simulation data obtained through the building cold load prediction system with OpenStudio model simulation, and constructs a basic model.Meanwhile, a model fusion method is designed: GRU and Transformer models are trained in parallel, and the results of the two are fused through an attention mechanism, thereby improving the prediction accuracy.After the project goes online and actual weather and equipment operation data are obtained, actual project data will be used to fine-tune the basic model based on transfer learning until the model accuracy meets the standard and the model is deployed online for application.The traditional method usually uses the prediction model of similar buildings for transfer learning to solve the cold start problem, while the present application is based on actual physical building simulation, training and transfer, ensuring higher accuracy.

The application discloses a kind of numerical control processing tool wear online monitoring and life prediction method, belong to intelligent manufacturing equipment technical field.The application is realized by FPGA multi-source sensor signal hardware level synchronous acquisition, and adaptive time-frequency feature extraction is completed by Bayesian optimization VMD and multi-head self-attention, and hybrid neural network embedding wear physical mechanism is constructed;Concept drift detection is realized by combining KS test, model incremental learning is completed using EWC algorithm, and remaining life uncertainty quantitative prediction is realized by Monte Carlo and particle filtering;Finally, through EtherCAT bus and numerical control system closed-loop linkage, process adaptive compensation and automatic tool changing are completed.The application solves the defects of traditional technology signal synchronization difference, weak model generalization, no closed-loop control, high monitoring accuracy, strong real-time, can effectively reduce tool cost, reduce unplanned downtime, suitable for multi-axis numerical control processing tool intelligent operation and maintenance.

The invention discloses a power system cascading failure simulation method and device considering a new energy off-network risk, a medium and a product, and relates to the technical field of power system risk assessment, and the method comprises the steps: obtaining training data; according to the training data, training a new energy off-network simulation hybrid neural network by using an active learning method to obtain a new energy off-network simulation hybrid neural model; constructing a cascading failure dynamic evolution environment; the cascading failure dynamic evolution environment is integrated with a line overload factor model, a frequency deviation factor model, an emergency control model and a new energy off-network simulation hybrid neural model; in a cascading failure dynamic evolution environment, a reinforcement learning algorithm is adopted to train an intelligent agent, so that the intelligent agent learns a failure evolution law and is used for deducing and simulating a cascading failure development path. According to the method, the active learning neural network replaces the new energy plant station to quit simulation, and a complex and time-consuming simulation process does not need to be carried out when cascading failure simulation is carried out.

This invention proposes a method for denoising vibration signals from structural health monitoring based on a hybrid neural network. This method, implemented according to embodiments of the invention, eliminates the need for prior signal knowledge and manual parameter settings. It effectively removes various noise types from structural health monitoring vibration signals, significantly improves the signal-to-noise ratio, and reduces the root mean square error, achieving efficient and automated noise reduction of vibration signals.

This invention relates to the field of deep learning technology, specifically to a power generation equipment anomaly prediction and adaptive maintenance system based on fused deep learning. The system first acquires multimodal local and global data through a multimodal acquisition unit; then, a first neural network processing unit inputs the multimodal local data into a knowledge-embedded hybrid neural network model to obtain local anomaly predictions; a second neural network processing unit inputs the multimodal global data into an encoder-decoder network based on cross-modal attention to obtain global anomaly predictions; a reinforcement learning processing unit optimizes the maintenance strategy using a reinforcement learning model and performs hyperparameter optimization on the reinforcement learning model using a genetic algorithm, thus achieving model interpretability; finally, an output unit provides a comprehensive anomaly prediction result and an updated maintenance strategy optimized by reinforcement learning and genetic algorithms to improve the effectiveness of anomaly prediction and maintenance.