127 results about "Artificial neural network model" patented technology

The invention discloses a rapid progression type nasopharyngeal carcinoma risk prediction method based on an artificial neural network, and relates to the field of medical informatics crossing. The invention provides a rapid progressive nasopharyngeal carcinoma risk prediction method based on an artificial neural network, and aims to solve the problem that a rapid progressive nasopharyngeal carcinoma patient is difficult to recognize in time by depending on TNM staging and experience judgment in the prior art. According to the method, historical case data collection, missing value filling and standardization preprocessing, core feature determination through feature screening, class imbalance correction, feature coding and feature matrix construction are sequentially carried out, an artificial neural network model is trained and optimized under a cross validation framework, and performance and threshold values are determined on a validation set. During clinical application, patient features are input, and the model outputs a rapid progress risk probability and a risk level. Compared with a conventional staging or linear model, the method can improve the prediction accuracy, and achieves the early recognition and individualized treatment of a high-risk patient.

To provide a balance function management system for balance function state information generation and a balance function rehabilitation program.SOLUTION: A memory configured to store at least one processor and at least one artificial neural network model that stores instructions executable by the processor and is executed on a computing device, wherein the at least one processor is configured to input frame images of n (where n is a natural number) videos obtained by photographing a subject through n cameras to the at least one artificial neural network model, At least one of information related to the coordinates of the head, the coordinates of the center of the pupil, and the phase change of the eyeball of the subject according to the order of the frame images of the m-th (m is a natural number from 1 to n) moving image may be acquired, and information related to the movement of the head and the movement of the eyeball for generating the balance state information or performing the balance rehabilitation program may be generated using the acquired information.SELECTED DRAWING: Figure 1

The invention relates to the technical field of coating preparation, in particular to a high-entropy alloy corrosion-resistant coating based on machine learning optimization and a preparation method thereof.The preparation method comprises the steps that a preparation data set of the high-entropy alloy coating is constructed, and coating element components, technological parameters and corrosion environment parameters in the preparation data set serve as input characteristics; taking the thickness, the hardness and the elastic modulus of the corrosion layer as output characteristics, training an artificial neural network model, and establishing a performance prediction model; a multi-objective particle swarm optimization algorithm is adopted, and magnetron sputtering process parameters are optimized and solved by taking the minimum thickness of a corrosion layer and the maximum hardness and elasticity modulus as optimization objectives; according to the optimal process parameter combination, preparing an AlCrFeMoTi high-entropy alloy coating on a ferrite / martensitic steel matrix by adopting a magnetron sputtering five-target co-deposition technology; and carrying out corrosion resistance test on the prepared coating, and comparing a test result with a predicted value of the model to verify the accuracy of the model.

The present invention relates to a method for detecting and removing a motion artifact of a functional near-infrared spectroscopy signal. Disclosed are a method and a device for detecting and removing a motion artifact of a near-infrared spectroscopy signal in real time on the basis of an artificial neural network model, the method comprising: a conversion step of converting time series data of functional near-infrared spectroscopy signals measured through a plurality of channels from a target into image data; a detection step of detecting whether noise is present in the converted image data using a pretrained detection model; and a removal step of removing noise from image data where noise has been detected using a pretrained noise removal model.

This disclosure provides a message delivery method, etc., that is executed by at least one processor. [Solution] In one embodiment of the present disclosure, a message delivery method performed by at least one processor includes: identifying an event relating to a message request in an instant messaging application; using an artificial neural network model to obtain second information relating to at least one of a context relating to the time of event identification or a prompt relating to the event from first information relating to a user account connected to the instant messaging application; using an artificial neural network model to generate a first message based on the second information; and providing the first message through a first message room of the instant messaging application relating to the artificial neural network model.

The application discloses a design method and device of a metasurface optical neural network, and the design method comprises the following steps: obtaining an artificial neural network model and a metasurface optical neural network model, wherein the artificial neural network model is a trained model; selecting n intermediate layers in the artificial neural network model in the order from shallow to deep, one-to-one corresponding the n intermediate layers and n metasurface layers in sequence, and calculating an intermediate feature loss according to the similarity between the features output by each layer in the n intermediate layers and the features output by the corresponding metasurface layer; training the metasurface optical neural network model; and determining the process parameters of the metasurface according to the trained metasurface optical neural network model. The design method completely transplants the calculation capacity of the artificial neural network into the metasurface optical neural network, solves the problems of low design efficiency and low precision of the metasurface optical neural network, and is beneficial to the development of the metasurface optical neural network and the deployment in an environment with limited computing power and storage.

A method of determining deformation of a battery includes obtaining a first image by scanning a cross-section of a battery cell in one direction, inputting the first image into an artificial neural network model trained to distinguish a plurality of parts of the battery cell in the first image and obtaining coordinates corresponding to each of the parts, and generating a second image in which at least some of the coordinates are aligned according to a winding sequence related to the battery cell. Deformation of the battery cell is determined based on the second image.

The invention discloses a torque prediction method suitable for high-viscosity material mixing, and belongs to the field of hydraulic equipment. Rotation speed and pressure drop, temperature and flow in hydraulic power transmission are used as input, torque actually measured by a torque sensor is used as output, accurate prediction of the torque of the hydraulic kneading machine is achieved by training an artificial neural network model, the artificial neural network model is continuously optimized in a machine learning mode, and the prediction precision of the torque of the hydraulic kneading machine is improved. The prediction accuracy is improved, the artificial neural network model has good mobility, and an accurate torque prediction result is provided for the working condition that the torque sensor cannot be installed. The method is not limited by installation conditions and environmental factors of the torque sensor; high-precision torque prediction is realized through multi-working-condition testing, comprehensive data acquisition, redundant information removal, data normalization processing, and establishment and optimization of a feedforward neural network. The constructed artificial neural network model has good mobility and can adapt to different hydraulic mixing devices.

Techniques of nonlinear quantization of an artificial neural network model having first weights. For example, a predetermined number of unique, second weights having a nonlinear distribution in a weight space of the first weights can be identified to generate a quantized model based on replacing, in the artificial neural network model, the first weights with closest ones from the second weights. A linear mapping between the second weights and values of conductance of memristors of an accelerator configured to perform operations of multiplication and accumulation can be used to determine the same predetermined number of programming voltages. Conductance of the memristors can be programmed using the programming voltages in preparation of the accelerator to perform an operation of multiplication and accumulation in the quantized model. The nonlinear distribution and the linear mapping can be adjusted to increase or optimize the accuracy of the quantized model.

This invention discloses a method for accurate lithology identification and correction of logging data based on artificial neural networks, comprising the following steps: collecting logging data, laboratory test data, and regional geological background data of the target area from buried hills; preprocessing the collected logging data and laboratory test data to establish a correspondence between the logging data and laboratory test data, forming a dataset; constructing an artificial neural network model; training and optimizing the constructed artificial neural network model based on the dataset; inputting the logging data of the target well to be corrected into the trained artificial neural network model, outputting corrected mineral element content data; and, based on the corrected mineral element content data and combined with the regional geological background, achieving lithology classification and accurate identification of the target well. This invention uses the above method to convert logging data into laboratory-precision data, improving the accuracy of lithology identification under complex geological conditions such as Archean metamorphic buried hills.

The present disclosure relates to a thermal analysis system and method for a battery system. The thermal analysis system includes a learning data generation device configured to generate second thermal analysis data using first thermal analysis data and a first artificial neural network model. The first thermal analysis data is obtained through numerical thermal analysis of a battery system. The thermal analysis system also includes a model construction device configured to construct a thermal analysis model by using a second artificial neural network model with the first thermal analysis data and the second thermal analysis data as learning data.

The invention belongs to the field of environmental protection and micro-plastic identification, and discloses a micro-plastic infrared characteristic spectrum extraction and efficient and accurate identification method, which comprises the following steps of: 1, acquiring infrared spectrum data of a micro-plastic sample, and constructing a micro-plastic spectrum database; 2, extracting a characteristic spectrum from the infrared spectrum data in the step 1 by adopting a progressive two-step method combining equal-interval sampling and a competitive self-adaptive reweighted sampling algorithm; step 3, performing standard normal transformation on the transmittance of the characteristic spectrum in the step 2; and 4, constructing a feature training set and a feature test set by taking the transmittance of the infrared spectrum transformed in the step 3 as input and the micro-plastic type as output. Training the artificial neural network model by adopting the feature training set, and optimizing hyper-parameters in the model through the feature test set by applying a genetic algorithm to form a final micro-plastic identification artificial neural network model; and step 5, for micro-plastics to be identified, obtaining the characteristic transmittance of the micro-plastics to be identified through the spectral data obtained in the step 1 and the characteristic wave number extracted in the step 2, performing standard normal transformation on the micro-plastics to be identified through the step 3, inputting the transformed infrared spectral transmittance into the micro-plastic identification artificial neural network model, and giving the type of the micro-plastics. And identification of the micro-plastic is realized. According to the method, traversal sampling of the whole spectrum data set is effectively avoided, the extraction efficiency of the characteristic spectrum is improved, hyper-parameter optimization is performed on the artificial neural network model by adopting the genetic algorithm, overfitting of the model is effectively prevented, and the generalization ability of the model is improved.

The application discloses a turbine blade loss model construction method based on deep learning and loss weight analysis, in order to explore how to correct the existing turbine blade loss model, firstly, each loss in the blade loss needs to be split, and the prediction size of the existing model is compared, the coefficients and terms that need to be corrected and the correction terms that need to be added are analyzed, and then a loss prediction model with a correction form is formed. By comparing the turbine blade loss with different blade parameters, the blade parameter variables that need to be considered are found. The function relationship between the blade parameter variables that need to be considered and the coefficients that need to be corrected (or the terms and the correction terms that need to be added) is established by using an artificial neural network model, and is brought into the loss prediction model with the correction form, and then a turbine blade loss prediction model is constructed. The method can accurately predict the turbine blade loss in a large attack angle working range.

The present specification provides a training method of an artificial neural network model for correcting image color and an image correction method using a trained artificial neural network model. The training method of the artificial neural network model according to the present specification uses, as training data, a pair of images before and after color correction. In the training method, the image before color correction is input into the artificial neural network model, N parameter values (N is a natural number) are obtained from the artificial neural network model, and at least one color correction program is executed according to the obtained parameter values, whereby a converted image, the color of which has been converted from the image before color correction contained in the training data, may be obtained. In addition, the artificial neural network model may be trained such that a difference value between the image after color correction contained in the training data and the converted image is reduced.

The invention relates to the technical field of power device transient junction temperature evaluation, and discloses a neural network modeling method for IGBT transient junction temperature evaluation based on historical temperature rise characteristics, which comprises the following steps: collecting chip loss, thermal resistance aging grade, chip junction temperature and environment temperature data of an IGBT device under different working conditions, and preprocessing the collected data; constructing a transient junction temperature evaluation model by using an artificial neural network, introducing a Bayesian optimization algorithm to optimize hyper-parameters trained by a neural network model, training the model by using preprocessed data, and establishing a mapping relationship between historical temperature rise characteristic related data and future junction temperature; performing transient junction temperature evaluation on the IGBT device by using the trained artificial neural network model, evaluating the performance of the model, and performing optimization and improvement on the model according to an evaluation result; the technical problem that an existing transient junction temperature evaluation method cannot deal with a multi-chip packaging scene is solved.

A training data generation method for training an artificial neural network model including inputting a first prompt related to at least one first object in a specific context into a language model, acquiring first text data related to the at least one first object output from the language model, acquiring a first image related to the specific context, generating, based on at least one of the first text data or the first image, arrangement information related to an arrangement of the at least one first object for the first image, generating, based on at least one of the first text data, the first image, or the arrangement information, a second image in which the at least one first object is arranged in the first image, and outputting the second image.

The present application relates to a kind of three valve combined intelligent control method based on fuel cell membrane moisture content, it relates to fuel cell control technical field, for the unstable operation phenomenon caused by hydrogen pressure fluctuation, flow mismatch and membrane moisture content imbalance of fuel cell, specifically provide a kind of three valve combined intelligent control method based on fuel cell membrane moisture content, method includes: based on artificial neural network model, membrane moisture content is predicted and valve opening is corrected;Current hydrogen pressure fluctuation value in hydrogen discharge process of hydrogen system is obtained, and valve is again fine-tuned using PID control mode, realizes the self-adaptive regulation of system under all operating conditions and decoupling control under multiple outputs, finally achieves the purpose of hydrogen accurate supply and stack stable operation.

A neural processing unit includes an internal memory including a plurality of memory units; a controller configured to control read and write operations of data in at least one of an input feature map domain, a weight domain, and an output feature map domain with respect to each of the plurality of memory units based on an operation schedule in a machine code in which a plurality of operation steps of an artificial neural network model are set.

An electrode design system based on artificial intelligence includes: an electrode design factor prediction system to predict an electrode design factor, the electrode design factor prediction system including: a model execution unit to execute a first artificial neural network model trained to predict an electrode design factor of a target battery based on a condition of the target battery and random data; a storage unit to store data associated with the first artificial neural network model; and a communication unit to transmit, to an electrode process facility system, the predicted electrode design factor of the target battery. The electrode process facility system is to adjust process equipment based on the predicted electrode design factor.

A radio frequency circuit optimization design method based on high-performance computing, comprising the following steps: sub-modularizing the radio frequency circuit; constructing a circuit sub-module neural network; using GPU for parallel computing based on the circuit sub-module neural network; cascading the scattering parameter results of each level of sub-module circuit to obtain the scattering parameter results of the overall circuit; searching for an optimization scheme that meets the performance index condition from the scattering parameter results of the overall circuit; finding the Pareto optimal solution for the pros and cons of the search results, and screening out the circuit that meets the actual engineering index. Compared with the traditional artificial neural network used for modeling the relationship between the design size and electrical behavior of the full-stage radio frequency circuit, the method is more flexible, can improve the reusability and accuracy of the artificial neural network model, and can accelerate the calculation by using the high parallel computing characteristics of the neural network model.

This invention discloses a method and system for assessing the service status and diagnosing faults of cables, belonging to the field of cable aging and fault diagnosis technology. The method first constructs a simulation model of an equivalent cable and verifies the model. Based on the verified simulation model, it simulates the cable materials to obtain simulation data. Using the aging characteristic data and state characteristic data of the cable materials as training data, it trains an artificial neural network model, resulting in a cable prediction model. This model includes a fault prediction unit and an aging prediction unit, which are correlated probabilistically. The aging prediction unit outputs the aging state of the cable, and the fault prediction unit diagnoses faults based on probabilities. This method, through neural network model analysis, constructs a cable material identification and classification model under the influence of multiple interference factors, thereby achieving cable service status assessment and fault diagnosis.

This invention discloses a pipeline leak detection method and apparatus based on a distributed optical fiber sensing signal and feature joint input FC-ANN network, comprising the following steps: acquiring optical fiber signals including pipeline perimeter disturbances and leak conditions, and performing filtering preprocessing; extracting features including time-domain and frequency-domain features from the filtered and preprocessed optical fiber signals, and constructing a joint sequence from the filtered and preprocessed optical fiber signals and the extracted features; constructing a fully connected artificial neural network model, inputting the joint sequence into the fully connected artificial neural network model for model training, and optimizing the model training effect using evaluation metrics; and using the trained fully connected artificial neural network model to detect pipeline leaks from newly acquired optical fiber signals. This invention can more accurately classify disturbance events with fuzzy features and leakage events with stable features, and is suitable for efficient and accurate identification and location of pipeline leak events.

A neural processing unit (NPU), a method for driving an artificial neural network (ANN) model, and an ANN driving apparatus are provided. The NPU includes a semiconductor circuit that includes at least one processing element (PE) configured to process an operation of an artificial neural network (ANN) model; and at least one memory configurable to store a first kernel and a first kernel filter. The NPU is configured to generate a first modulation kernel based on the first kernel and the first kernel filter and to generate second modulation kernel based on the first kernel and a second kernel filter generated by applying a mathematical function to the first kernel filter. Power consumption and memory read time are both reduced by decreasing the data size of a kernel read from a separate memory to an artificial neural network processor and / or by decreasing the number of memory read requests.

The invention discloses an artificial intelligence-based electrode design system and method. The system comprises: an electrode design factor prediction system for predicting an electrode design factor, the electrode design factor prediction system comprising: a model execution unit for executing a first artificial neural network model, the first artificial neural network model being trained to predict the electrode design factor of a target battery based on a condition of the target battery and random data; the storage unit is used for storing data associated with the first artificial neural network model; and a communication unit for transmitting the predicted electrode design factor of the target battery to an electrode process facility system. The electrode process facility system adjusts the process equipment based on the predicted electrode design factors.

The application relates to a scalar engine processing method and device for an artificial intelligence chip. The method comprises the following steps: an upper module in the chip acquires an artificial neural network model to be deployed in the chip; the upper module performs conversion processing on the artificial neural network model based on an instruction set built in a scalar engine in the chip, obtains a plurality of target instructions corresponding to the artificial neural network model, and sends the plurality of target instructions to the scalar engine; and the scalar engine executes the plurality of target instructions to realize the compilation processing corresponding to the artificial neural network model in the chip. The method can improve the flexibility of the in-chip compilation processing of the artificial intelligence chip on the artificial neural network model.

A method and device with motion control are provided. The method includes determining reachability constraint information of the electronic device based on current pose information of the electronic device, determining collision avoidance constraint information of the electronic device based on the current pose information and sensor data of the electronic device, determining feasibility constraint information of the electronic device based on the reachability constraint information and the collision avoidance constraint information, and infer next pose information of the electronic device by inputting the feasibility constraint information and the sensor data into an artificial neural network model.

A method of estimating depth information includes generating a first simulated image using a simulator provided with a first depth map, training an artificial neural network model based on the first depth map and the first simulated image, generating a second depth map by inputting an actual image into the trained artificial neural network model, and generating a second simulated image using the simulator provided with the second depth map.

A method of determining deformation of a battery includes: obtaining a first image by scanning a cross-section of a battery cell in one direction; inputting the first image into an artificial neural network model trained to distinguish a plurality of parts of the battery cells in the first image and obtaining coordinates corresponding to each of the plurality of parts; and generating a second image in which at least some of the coordinates are aligned according to a winding order associated with the battery cell and determining a deformation of the battery cell based on the second image.

A control method for an endoscopic device includes obtaining, from an image sensor, an image of an inside of a body, inputting the image to an artificial neural network model trained to classify the obtained image, classifying and outputting, by the artificial neural network model, the image as an air situation image, a water situation image, and / or a suction situation image, and controlling the endoscopic device to drive an air unit, a water unit, and / or a suction unit according to an output classification result.

A method for training a medical image performed by a medical image processing apparatus for processing a medical image for a body is provided. The method for training a medical image includes preparing a first input data set including a training chest X-ray image, and a bone enhancement image or a bone extraction image acquired from the training chest X-ray image; preparing a label data for the first input data set including osteoporosis information or bone mineral density information corresponding to the training chest X-ray image; and training an artificial neural network model using the first input data set and the label data.