657 results about "Artificial neural network model" patented technology

The present invention discloses a radar scattering field simulation method for a three-dimensional moving target group combined with a neural network, and belongs to the field of electromagnetic simulation technology. The method of the present invention converts the electromagnetic response of a three-dimensional moving target group into the response of the target when it is distributed at different positions, and extracts the three-dimensional numerical Green's function of the randomly distributed target group in the solution domain by changing the spatial distribution of the target group; the three-dimensional numerical Green's function is split into scattering terms and coupling terms, and the artificial neural network is trained separately; when training the artificial neural network, according to the characteristics of the moving target group, its position distribution and scattering terms / coupling terms are input into the artificial neural network as the main features; the scattering terms and coupling terms output by the artificial neural network model are combined into numerical Green's function terms, which can be applied to integral equations to achieve fast and accurate solution of the scattering field of the three-dimensional moving target group.

The invention provides a composite material performance prediction and process optimization method based on a neural network, and the method comprises the steps: firstly collecting multi-source data in the preparation and test process of a composite material, carrying out the preprocessing of the data, screening key feature variables as input variables, constructing a feedforward artificial neural network model, and predicting and outputting the performance indexes of the composite material. And training the model, performing iterative optimization on model parameters, and optimizing composite material process parameters by using the optimized model based on a reverse optimization strategy of a genetic algorithm to obtain an optimal process parameter combination. The invention provides a scientific, efficient and reliable tool for design and optimization of composite materials, and particularly has wide application prospects in high-requirement industries such as aerospace and the like.

The embodiment of the present invention discloses a method for predicting corneal reshaping parameters based on an artificial neural network, and relates to the technical field of orthokeratology lenses. The method comprises: obtaining an initial sample data set, deleting target samples that do not contain necessary parameters, and obtaining a basic sample data set; preprocessing the basic sample data set to obtain a final sample data set; training the initial artificial neural network model to obtain a preliminary model; using the preliminary model to forward propagate the input layer data of the basic sample data set to obtain a calculation result and compare it with the result of the basic sample to obtain a final sample data set, and using the final sample data set to train the preliminary model to obtain a prediction model; inputting actual eye detection data into the prediction model to obtain corresponding corneal reshaping parameters. The present invention splits the data to be predicted into multiple dimensions, constructs different prediction models according to each data to be predicted, and then performs distributed training and uses them in combination.

The invention discloses an electric vehicle charging load prediction method, which comprises the following steps: firstly, screening variables with high correlation through a maximum information coefficient, selecting features with high correlation as model input, and taking a charging load as output; reconstructing the data training set by adopting cross validation; a charging load integration model is constructed, the integration model is divided into a base learning device and a meta learning device, the base learning device comprises a DCformer model, a BiTCN model and an artificial neural network ANN model, and a lightweight gradient elevator LightGBM is used as the meta learning device; the parameters of the LightGBM model are optimized by adopting an improved star graffiti optimization algorithm INOA; an error sequence is obtained after a preliminary prediction value is obtained through model prediction, and a final prediction value is obtained by taking LightGBM as an error prediction model. According to the method, data features are effectively extracted by integrating a plurality of single models, the prediction accuracy is improved, and compared with a traditional model, the stability is better, and the generalization ability is higher.

A processor-implemented method of pre-processing for deep neural network compilation comprising receiving a representation of an artificial neural network (ANN) model. An operator embedding is generated to represent operators of the ANN model in an embedding space. A graph neural network (GNN) processes the operator embedding to generate a graph embedding corresponding to the ANN model according to a learned distance metric. The GNN determines a set of hyperparameters for the ANN model based on the graph embedding.

The invention belongs to the technical field of deep learning, and discloses a cervical intraepithelial neoplasia risk prediction model and a construction method thereof, electronic equipment and a storage medium. A data analysis step: screening out original significant features; constructing a decision tree model by taking the original significant features as input data to obtain a prediction result of each sample; extracting a prediction result as a new significant feature; and combining the original significant features and the new significant features into an extended feature set, and training an artificial neural network model by using the extended feature set to obtain a cervical intraepithelial neoplasia risk prediction model. According to the method, the decision tree and the artificial neural network are integrated, and clinical screening indexes are combined, so that the cervical intraepithelial neoplasia risk assessment is realized, the prediction accuracy and robustness are improved, and the clinical diagnosis efficiency is greatly improved.

The invention discloses a construction method of a TiH binary system machine learning potential. The construction method comprises the following steps: constructing a reference electronic structure data set by adopting first principle calculation software VASP; training and testing an artificial neural network potential-ANN potential model on the reference data set by using the artificial neural network model; vASP and ANN potential obtained through training are used for calculating and comparing a Dimer curve, a lattice constant, an elastic modulus and the like for performance testing; and performing targeted improvement on data with large comparison error between the ANN model and the DFT result in model test and performance test results by increasing related configuration weights until an ANN potential model conforming to the precision is obtained. The machine learning potential model constructed by the invention can better fit the interatomic interaction potential energy surface of the TiH binary system, solves the problems of small time-space scale and low precision of the traditional empirical potential in the calculation of the first principle, and provides a technical basis for the high-precision and large-size simulation and research of the TiH binary system.

An electronic device according to an embodiment in this document includes a memory configured to store an artificial neural network model and a processor functionally connected to the memory, wherein the processor obtains a user's current utterance, generates a plurality of response candidates according to the current utterance using the artificial neural network model, and performs reinforcement learning on the artificial neural network model by selecting a response according to the current utterance that best matches a specified criterion including a performance indicator from among the plurality of response candidates, using a large pre-trained model.

The invention discloses a three-dimensional moving target group radar scattering field simulation method combined with a neural network, and belongs to the technical field of electromagnetic simulation. The method comprises the following steps of: converting electromagnetic response of a three-dimensional moving target group into response of targets distributed at different positions, and extracting a three-dimensional numerical green function of a randomly distributed target group in a solution domain by changing spatial distribution of the target group; splitting the three-dimensional numerical green function into a scattering term and a coupling term, and training an artificial neural network respectively; when the artificial neural network is trained, position distribution and scattering terms / coupling terms of the moving target group are used as main characteristics to be input into the artificial neural network according to the characteristics of the moving target group; scattering terms and coupling terms output by the artificial neural network model are combined into numerical Green function terms which can be suitable for integral equations, and rapid and accurate solving of a three-dimensional moving target group scattering field is achieved.

The invention discloses a modeling method for an IGBT switch transient state of an electric drive system of a new energy vehicle. The modeling method comprises the steps that S1, original waveform data and simulation waveform data are collected; s2, calculating the original waveform data and the simulation waveform data by using a linear interpolation method to obtain pre-processed original waveform data and pre-processed simulation waveform data; s3, determining a training data set of a neural network; S4, establishing an artificial neural network model of a multi-layer perceptron structure, and determining the number of neurons of each layer according to the original waveform data and the simulation waveform data; s5, dividing the training data set into a training data set and a verification data set, performing training, and determining model parameters of optimal verification loss according to the verification data set; and S6, carrying out cycle training on each training time point, storing the weight and bias of the neural network corresponding to each training time point, and forming an IGBT switch transient model of the electric drive system. The transient waveform simulation device can accurately simulate the transient waveform when the IGBT is switched on and switched off, and is suitable for simulation of a large amount of data.

The invention belongs to the technical field of big data information processing, and provides a micro-service system root cause positioning method and system based on a space-time diagram neural network, and the method comprises the steps: obtaining calling data information and monitoring index data of a micro-service system, constructing a micro-service time sequence calling topological graph stack, and combining a topological graph and monitoring index information to obtain a micro-service time sequence calling topological graph stack; learning time sequence features of the topological graph by using a long-short-term memory artificial neural network model, and learning space structure features of the topological graph by using a graph convolutional neural network model; performing spatio-temporal feature fusion on the learned time sequence feature information and spatial structure feature information, and introducing an attention mechanism during spatio-temporal feature fusion; and based on the fused space-time fusion features of each topological graph node, determining the probability that each node is a root cause by using a root cause positioning model, and obtaining a final result of root cause positioning. According to the invention, spatial-temporal characteristic information in the topological graph can be fully mined, effective root cause positioning is carried out, and the accuracy of root cause positioning is improved.

A brake motor abnormality detection device may detect abnormality of a brake motor generating a driving force to displace a friction member at a side of a wheel of a vehicle to generate a braking force caused by friction due to the displacement of the friction member. The brake motor abnormality detection device includes a memory in which one or more instructions are stored and a processor configured to execute the one or more instructions to input input data to an artificial neural network model configured to receive the input data and output an estimation value of an output variable related to an output of the brake motor, obtain the estimation value of the output variable, and detect whether the brake motor is abnormal by using a difference between the estimation value of the output variable and a measurement value of the output variable.

The invention relates to the technical field of forest fire spreading prediction, in particular to a forest fire spreading prediction method and system based on a physical information neural network. The method comprises the following steps: acquiring forest fire sample data, and preprocessing the forest fire sample data to obtain a sample set; constructing an artificial neural network model; defining a physical constraint error term according to mass conservation and momentum conservation, defining a data error term, and constructing a loss function according to the physical constraint error term and the data error term; training and testing the model by using the sample set and the loss function; and acquiring real-time data of the forest fire, preprocessing the real-time data of the forest fire, and inputting the data into the model to predict the spreading of the forest fire. According to the method, physical knowledge can be fused into deep learning driven based on historical fire data, and the model is guided by using a physical rule in the learning process, so that the generalization ability of the forest fire spreading prediction model is improved, and the forest fire spreading situation is more reliably and accurately predicted.

A processor-implemented method for converting pre-trained models from a first floating point format to a second floating point format includes receiving an artificial neural network (ANN) model as an input. The ANN model is trained to operate on values in the first floating point format. A layer of the ANN model in which to apply a scaling parameter is selected based on a layer type and a topology of the ANN model. A maximum intermediate value is determined according to a distribution of weights in the selected layer of the ANN model. The scaling parameter is computed based on a ratio of the maximum intermediate value and a maximum representable value. The scaling parameter is applied to weights of the selected layer of the ANN model.

A method for generating a personalized artificial neural network (ANN) model receives an input at a first artificial neural network. The input is processed to extract a set of intermediate features. The method determines if the input is out-of-distribution relative to a dataset used to train the first artificial neural network. The intermediate features corresponding to the input are provided to a second artificial neural network bases on the out-of-distribution determination. Additionally, the system resources for performing the training and inference tasks of the first artificial neural network and the second, personalized artificial neural network are allocated according to a computational complexity of the training and inference tasks and a power consumption of the resources.

The invention discloses a design method of a special-shaped underwater pressure-resistant shell, which comprises the following steps: S1, according to the use environment of a to-be-designed special-shaped underwater pressure-resistant shell, determining various types of structure parameters which are required by the to-be-designed special-shaped underwater pressure-resistant shell and have association relationship and the initial value range of the structure parameters; s2, collecting a plurality of groups of data samples of the structure parameters conforming to the incidence relation in an initial value range; s3, m evaluation indexes used for evaluating the performance of the to-be-designed special-shaped underwater pressure-resistant shell are determined according to the use environment of the to-be-designed special-shaped underwater pressure-resistant shell, original evaluation values corresponding to the m evaluation indexes of each group of data samples are obtained, and the multiple data samples and the corresponding m original evaluation values form a data set; and S4, training an artificial neural network model by using the data set to obtain an initial prediction model for predicting an original evaluation value of the to-be-designed special-shaped underwater pressure-resistant shell. According to the method, multi-dimensional design of the special-shaped underwater pressure-resistant shell can be achieved, the multi-performance requirement is met, the design process is simple, and the design efficiency is high.

The invention discloses a highway reconstruction and extension project asphalt pavement compaction quality accurate prediction algorithm, and the algorithm comprises the steps: collecting the rolling times, speed, track, temperature and vibration data of a road roller in real time according to an on-site intelligent device and a sensor, and carrying out the correlation analysis of the compaction degree data and corresponding information supervision key parameters; and an artificial neural network model is established, real-time prediction of the compaction degree by informatization supervision key parameters is realized, quality problems are found in time, and the construction quality is mastered in real time, so that shoveling and re-paving of completed parts in the later period are avoided, the pavement quality is ensured, and the construction period is shortened.

The invention discloses a concrete durability prediction method based on data expansion and machine learning, and the method comprises the steps: 1, collecting concrete mix proportion data and environmental parameters in an engineering environment, and building a standard concrete durability database; 2, performing data expansion on the standard concrete durability database based on a Markov Monte Carlo algorithm; 3, constructing an artificial neural network model, taking the expanded database as training sample input, and optimizing and selecting model hyper-parameters; and 4, selecting the artificial neural network model of the hyper-parameter combination with the optimal performance for training, and realizing accurate prediction of the concrete durability under different mix proportions and environmental influences. According to the method, the Markov Monte Carlo algorithm is adopted for data expansion, the loss of the parameter relationship is avoided by adopting an internal preset model method, the defect that huge data is needed to support the establishment of an accurate machine learning model is overcome, and accurate and efficient prediction of the concrete durability chloride ion diffusion coefficient in an engineering environment is realized.

The invention discloses a steel rail abrasion intelligent prediction method and system considering railway space line shape, and the method comprises the steps: S1, building a vehicle-track coupling dynamic model based on vehicle structure characteristic parameters and track structure characteristic parameters; s2, based on the vehicle-track coupling dynamic model, establishing a steel rail abrasion calculation model, and calculating inner and outer side steel rail abrasion values under different track space linear parameters to construct a steel rail abrasion sample data set; s3, constructing a steel rail abrasion prediction artificial neural network model, and training the steel rail abrasion prediction artificial neural network model by using the steel rail abrasion sample data set; s4, optimizing the trained steel rail abrasion prediction artificial neural network model to obtain a final prediction model; and S5, predicting the abrasion of the railway steel rail by using the final prediction model. According to the method, the railway space linear parameter-steel rail abrasion prediction model is established, so that the steel rail abrasion under different plane and vertical section linear parameters and combinations is efficiently and accurately obtained.

The invention discloses an intelligent rehabilitation training method, system and device based on a wearable hemiplegic patient, and relates to the technical field of rehabilitation training, and the method comprises the steps: collecting multi-dimensional data of the patient, carrying out the preprocessing, obtaining an evaluation value of the multi-dimensional data through an artificial neural network model, carrying out the splicing, generating an evaluation vector, and carrying out the recognition of the evaluation vector; performing optimization as an individual of a bald eagle search optimization algorithm to obtain an optimal evaluation value vector; defining an evaluation value in the optimal evaluation value vector as an input variable of fuzzy logic to perform fuzzy reasoning, forming a setting vector and a training target vector, constructing Bayesian prior distribution, a likelihood function and Bayesian posteriori distribution to perform maximization solution, obtaining an optimal training target vector to perform linear mapping, generating a specific value vector, and obtaining the optimal training target vector. A rehabilitation training task and dynamic adjustment and feedback of a specific value vector are executed; according to the invention, effective optimization and personalized customization of the rehabilitation training process are realized, and the efficiency and effect of rehabilitation training are improved.

A device and a method for detecting an anomaly in a rear-wheel steering motor and a non-transient, computer-readable storage medium that stores a program for carrying out the method are disclosed. The device comprises a memory configured to store one or more instructions and a processor configured to execute one or more instructions.The processor executes one or more instructions to input data relating to the vehicle's rear-wheel steering into an artificial neural network model, to obtain estimated data relating to the vehicle's rear-wheel steering as output data from the artificial neural network model, to obtain measured data relating to the vehicle's rear-wheel steering from a sensor, to compare the estimated data and the measured data, and to detect the rear-wheel steering motor anomaly based on the result of comparing the estimated data and the measured data.

The invention relates to a medical image report generation method and device based on multi-modal data. The method comprises the following steps: acquiring original multi-modal data and preprocessing the original multi-modal data, wherein the multi-modal data comprises image data, complaint text data and clinical examination text data; predicting and generating a corresponding medical image report by using the trained multi-modal deep learning model; the prediction process specifically comprises the following steps: based on preprocessed image data, extracting image features by using a trained convolutional neural network model; based on the preprocessed chief complaint text data and clinical examination text data, using the trained artificial neural network model to extract semantic features; fusing the image features and the corresponding semantic features to obtain fused features; and inputting the fusion features into a full connection layer, and mapping to obtain a medical image report. Compared with the prior art, the method has the advantages of improving the reliability of medical image report generation, reducing the manual load and the like.

Methods and systems for predicting and monitoring gas emissions from a cogeneration unit is developed by use of the stacked multivariate deep learning algorithms. Deep learning-powered predictive emission monitoring system (PEMS), which utilizes the operational parameters of relevant instruments to predict and monitor the emitted NOx concentration, mass flow rate, and temperature of the flue gas at the exhaust stack of a cogeneration unit. The established system comprises a data pipeline module, a deep learning module, and a user interface module. The data pipeline module acquires the datasets of the measured gas emission variables and the representative settings of the operational process. The deep learning module stacks a hybrid convolutional neural network and recurrent neural network model with an artificial neural network model to process sequential and tabular datasets simultaneously. The user interface module demonstrates the results of predicted gas emissions and provides monitoring and analysis of gas emission processes.

The present invention relates to methods using artificial intelligence, AI, techniques to predict the berthing capacity of ships in onshore and offshore structures and comprises embodiments of a method for training an artificial neural network, ANN, model to predict the berthing capacity of ships, a method for predicting the berthing capacity of ships, and a non-transitory computer-readable medium.

An object of initial unknown position on a map may be determined by traversing through moving and turning to establish motion trajectory to reduce its spatial uncertainty to a single location that would fit only to a certain map trajectory. An artificial neural network model learns from object motion on different map topologies may establish the object's end-to-end positioning from embedding map topologies and object motion. The proposed method includes learning potential motion patterns from the map and perform trajectory classification in the map's edge-space. Two different trajectory representations, namely angle representation and augmented angle representation (incorporates distance traversed) are considered and both a Graph Neural Network and an RNN are trained from the map for each representation to compare their performances. The results from the actual visual-inertial odometry have shown that the proposed approach is able to learn the map and localize the object based on its motion trajectories.

The invention provides a soil nutrient detection method and device based on a vehicle-mounted double-channel soil spectrum, and relates to the technical field of soil detection.The method comprises the steps that first soil spectrum data of a first position and second soil spectrum data of a second position in a farmland area to be detected are collected at the same time in a vehicle-mounted double-channel mode; then obtaining a first feature vector corresponding to the first soil spectrum data and a second feature vector corresponding to the second soil spectrum data through an artificial neural network model, determining a target feature vector by calculating an average value of the first feature vector and the second feature vector, and then obtaining a target feature vector through a trained target neural network model. And obtaining soil nutrient detection results of the first position and the second position. According to the soil nutrient detection method based on the vehicle-mounted double-channel soil spectrum, detection can be carried out in a farmland through the vehicle-mounted equipment, operation is simple and rapid, and the accuracy and efficiency of soil nutrient detection are improved.

The invention discloses a geological monitoring and early warning method and system for geological exploration, and relates to the technical field of land exploration, and the method comprises the steps: monitoring the displacement at different depths in a sliding rock-soil body; carrying out real-time data acquisition on factors influencing the displacement of the rock-soil body; preprocessing the deep displacement time sequence data of the unstable slope slip plane; decomposing the displacement time sequence into a trend sample and a fluctuation sample by using a function fitting method; predicting a displacement time sequence trend sample by using an artificial neural network model; predicting a displacement time sequence fluctuation sample by using the long-short time memory network model; obtaining a displacement time sequence predicted value through a superposition summation method; and judging the deformation stage of the unstable slope according to the displacement time sequence predicted value, and outputting a corresponding early warning signal. By arranging the data monitoring module and the data prediction module, advanced perception of slope deformation trend development is realized in real time.

The invention provides a bay water health status evaluation method, device and product, and the method comprises the following steps: S1, determining a sampling site and collecting a zooplankton eDNA sample to obtain amplicon sequence variant information; s2, annotating amplicon sequence variant information to obtain species composition information and abundance data; s3, obtaining a first evaluation candidate index according to the species composition information and the abundance data; constructing a co-occurrence network and determining a second evaluation candidate index according to the network topology parameters; combining the first evaluation candidate index with the second evaluation candidate index to obtain a third evaluation candidate index; s4, selecting a reference point and a damaged point, and obtaining key evaluation candidate indexes through an artificial neural network model; and S5, constructing a zooplankton integrity index based on the key evaluation candidate indexes, and evaluating the water quality health grade according to the zooplankton integrity index. By utilizing the technical scheme, the precision of evaluating the ecological health condition of the severely polluted bay water body can be improved, and meanwhile, the complexity of an evaluation index system is reduced.

Disclosed are a device and method for detecting an abnormality of a solenoid valve of an electronically controlled suspension (ECS) system, and a non-transitory computer-readable storage medium storing a program for performing the method. The device for detecting the abnormality of the solenoid valve of the ECS system is a device for detecting an abnormality of a solenoid valve of an ECS system, which detects an abnormality of a solenoid valve disposed in an ECS system of a vehicle, and includes a memory configured to store one or more instructions, and a processor configured to execute the one or more instructions, wherein the processor executes the one or more instructions to input input data representing a state of the ECS system into an artificial neural network model, obtain an estimated value of a physical quantity representing an output of the ECS system that is output by the artificial neural network model, compare the estimated value with a measurement value of the physical quantity, and detect the abnormality of the solenoid.

The invention discloses a method for mix proportion optimization and performance prediction of canal lining concrete, which comprises the following steps: 1) collecting compressive strength and durability data of concrete under different mix proportion conditions through laboratory testing, and establishing an early-stage data set; 2) constructing a distribution model, and simulating and expanding the data set by adopting a Markov chain-Monte Carlo method; 3) training a Kriging agent model by using the expanded data set to obtain a high-precision virtual sample set; 4) standardizing a virtual sample output by the KS model, inputting the standardized virtual sample into the artificial neural network model as a training set, and determining an ANN model structure; 5) creating different types of test data sets, and evaluating the precision and robustness of the KS-ANN model; and 6) proposing an optimization method for concrete. According to the invention, by expanding and training limited test data, a high-precision concrete performance index prediction model is constructed, an optimization method is established, and the trial and error cost of the test is effectively reduced.