Structure Of Neural Network Model Research Articles

Deep-Learning-Based Information Fusion Methodology for Oil Film Coefficient Identification of Squeeze Film Dampers

Squeeze film damper (SFD) is one of the key components in the sophisticated rotating machines in the high-end aerospace industry. Rolling bearings in the aeroengine spindle system are often used in combination with elastic supports and SFDs to reduce rotor vibrations. Accurate coefficient identification of oil films in SFDs is critical to ensure safe and reliable operations of the complex equipment. However, this task remains a challenging issue in the past decades due to the difficulties in real-time measurement and computation during operations. The highly nonlinear relationship between the measurement and oil film coefficient cannot be effectively captured using traditional methods. This article proposes a novel deep-learning-based information fusion methodology for oil film coefficient identification of SFDs. An end-to-end deep neural network model structure is established, which explores heterogeneous SFD data simultaneously with fusion architecture and directly outputs the identification results. The high-dimensional data can be automatically processed, and high identification accuracy can be achieved. Experiments on the SFD test rig are carried out for validation. The experimental results demonstrate that the proposed method can well identify the oil film coefficients in an intelligent manner and is robust against environmental noise and different parameter settings. The proposed methodology is, thus, promising for applications in real industries.

Brain Age Prediction/Classification through Recurrent Deep Learning with Electroencephalogram Recordings of Seizure Subjects.

With modern population growth and an increase in the average lifespan, more patients are becoming afflicted with neurodegenerative diseases such as dementia and Alzheimer's. Patients with a history of epilepsy, drug abuse, and mental health disorders such as depression have a larger risk of developing Alzheimer's and other neurodegenerative diseases later in life. Utilizing recordings of patients' brain waves obtained from the Temple University abnormal electroencephalogram (EEG) corpus, deep leaning long short-term memory neural networks are utilized to classify and predict patients' brain ages. The proposed deep learning neural network model structure and brain wave-processing methodology leads to an accuracy of 90% in patients' brain age classification across six age groups, with a mean absolute error value of 7 years for the brain age regression analysis. The achieved results demonstrate that the use of raw patient-sourced brain wave information leads to higher performance metrics than methods utilizing other brain wave-preprocessing methods and outperforms other deep learning models such as convolutional neural networks.

Medical Data Classification Assisted by Machine Learning Strategy.

With the development of science and technology, data plays an increasingly important role in our daily life. Therefore, much attention has been paid to the field of data mining. Data classification is the premise of data mining, and how well the data is classified directly affects the performance of subsequent models. In particular, in the medical field, data classification can help accurately determine the location of patients' lesions and reduce the workload of doctors in the treatment process. However, medical data has the characteristics of high noise, strong correlation, and high data dimension, which brings great challenges to the traditional classification model. Therefore, it is very important to design an advanced model to improve the effect of medical data classification. In this context, this paper first introduces the structure and characteristics of the convolutional neural network (CNN) model and then demonstrates its unique advantages in medical data processing, especially in data classification. Secondly, we design a new kind of medical data classification model based on the CNN model. Finally, the simulation results show that the proposed method achieves higher classification accuracy with faster model convergence speed and the lower training error when compared with conventional machine leaning methods, which has demonstrated the effectiveness of the new method in respect to medical data classification.

Chinese News Text Classification and Its Application Based on Combined-Convolutional Neural Network

<p>A method based on combined-convolutional neural network (Combined-CNN) for Chinese news text classification is proposed. First of all, in order to solve the problem of a lack of special term set for Chi-nese news classification, a vocabulary suitable for Chinese long text classification is made by construct-ing a data index method. The Word2Vec pre-trained model was used to embed the text features word vectors. Second, by optimizing the structure of the classical convolutional neural network (CNN) model, a new idea of Combined-CNN model is proposed, which solves the problem of incomplete feature ex-traction of local text blocks and improves the accuracy rate of Chinese news text classification. Effective model regularization and RAdam optimization algorithm are designed in the model to enhance the model training effect. The experimental results show that the precision of the Combined-CNN model for Chi-nese news text classification reaches 93.69%. Compared with traditional machine learning methods and deep learning algorithms, the accuracy rate is improved by a maximum of 11.82% and 1.9%, respectively, and it is better than the comparison model in Recall and F-Measure. Finally, the Chinese news classifica-tion algorithm of the Combined-CNN is applied to realize a personalized recommendation system.</p> <p> </p>

Fast Multi-Physics Simulation of Microwave Filters via Deep Hybrid Neural Network

One fundamental difficulty in multiphysics numerical simulation is the complex interactions between different physics domains leading to plenty of computational costs. Although neural networks have recently been introduced in multiphysics simulations, the modeling complexity and the enormous amount of training data required may still pose significant challenges to researchers. In this work, we introduce a low-cost, electromagnetic-centric, multiphysics modeling approach to simulate microwave filters. With ground-truth datasets being generated from the finite element method, a novel deep hybrid neural network (DHNN) model structure is introduced, which uses the sigmoid and the ReLU functions as activators to mimic the diversity of biological neurons. A new, more feasible training algorithm is proposed for the efficient development of the DHNN model. The algorithm adopts the design-of-experiment (DOE) sampling technique and is specifically designed for the simulation of multiphysics responses. The strong approximation ability of the DHNN can lead to high-accuracy modeling with fewer training data and less resource consumption. Another advantage of this approach is that the modeling process is more concise and easier to apply compared with other modeling technologies. Numerical examples show that the DHNN can achieve higher accurate results with much less training data compared to traditional ANNs. The advantages of the proposed method in computational efficiency are more pronounced, especially when the amount of input data increases.

Analysis of Legal Attributes and Rights Attributes of Personal Information from the Perspective of Big Data.

With the popularization and rapid development of Internet technology (IT), social activities based on the exchange of personal information are rapidly popularizing. Although the free flow of a huge volume of personal data meets the needs of information exchange, its illicit use poses legal risks to rights holders and infringes on the applicable rights and interests of the information subject. The ambiguity of the right attribute and legal attribute of personal information leads to the scattered protection of personal information in the legislative system in different legal norms. Based on this background, this paper constructs an evaluation index based on the legal attributes and rights attributes of personal information rights and proposes a rationality evaluation method for the attributes of personal information rights based on neural networks. The completed work is as follows: (1) focusing on the legal attributes and the right attribute is discussed in detail, and the dissimilar views of scholars at home and abroad are presented. (2) The back propagation neural network (BPNN) model structure required in this study is defined, and the evaluation index based on the legal attribute and right attribute of personal information right is produced. (3) Use big data technology to collect relevant data and compare the results obtained by the BPNN model with the results obtained by the improved BPNN model. The results show that the improved BPNN evaluate model has smaller error and higher accuracy.

Economic Forecasting Model Based on Chaos Simulated Annealing Neural Network

In view of the prototype of the traditional chaotic network model, a neural network model with continuously updated chaotic noise is innovatively improved. This model has the advantages of two network models. On this basis, a logic graph is proposed, which can replace the chaotic noise generated by the running process of the model function. Models with these advantages can innovatively solve problems such as constrained optimization with dimensions larger than three, high discreteness, and weak linear relationship between convex and concave. Simulation results can be confirmed that the algorithm of this model is very close to the predicted value. This neural network model is an improved model with strong applicability and can be applied to optimization problems of economic systems or other industrial systems. In order to effectively alleviate the predictive control problem of nonlinear research objects, we propose a control method based on chaotic neural network in this study. Taking the economic model as the nonlinear object, establishing the basic structure of the chaotic neural network model, reconstructing the time and space structure, and obtaining the optimal solution of time continuation and embedding dimension through information entropy and pseudonearest neighbor, then the chaotic properties of nonlinear objects and the topology of chaotic neural networks are determined. In the simulation, test samples and experimental samples are established, and the predicted and true values are compared. The prediction results of the model established by this method in this study prove the effectiveness of the method. The training time of the chaotic neural network will not fall into a local minimum, thereby reducing the training time and ensuring the accuracy of the prediction.

Artificial Neural Network-Based Planting Arrangement of Smart City in Green Ecological Environment

The traditional urban planting arrangement is largely limited by the designer’s idea and has a high repetition rate and a low reference reuse rate. Therefore, a scientific and reasonable planting arrangement of the urban environment is necessary. In this work, research on planting arrangements in a smart city is carried out under green ecology and environment. Firstly, the planting arrangement is analyzed based on the structure, characteristics, and basic principles of the artificial neural network (ANN) model. ANN is frequently applied in pattern recognition, signal processing, system identification, and optimization. In the field of control, neural networks are used to deal with the nonlinearity and uncertainty of the control system and to approximate the identification function of the system. Secondly, the output value of the planting arrangement in the smart city is calculated according to the error backpropagation algorithm. During this period, the weight is adjusted according to the Hebb criterion, and the relevant statistical model of planting arrangement in the smart city is analyzed by ANN. Finally, suggestions on planting arrangements are given. The research shows that steamed bun-shaped plants have the largest total number in smart cities, followed by spherical and bush-like plants. Planting arrangement for spherical and palm or coconut-form plants is more frequent while planting arrangements for wind-shaped plants have a lower frequency. In terms of the importance of the planting arrangement, these 18 types of plants are very important for the green ecological environment in the smart city. Finally, suggestions on planting arrangements are given according to the research.

Designing efficient convolutional neural network structure: A survey

As a powerful machine learning method, deep learning has attracted the attention of numerous researchers. While exploring a high-performance neural network model, the floating-point operations of a neural network model are also increasing. In recent years, many researchers have noticed that efficiency is also one of important indicators to measure the property of neural network models. Obviously, the efficient neural network model is more helpful to deploy on mobile and embedded devices. Therefore, the efficient neural network model becomes a hot research spot. In this paper, we review the methods related to the structural design of efficient convolution neural networks in recent years. According to the characteristics of these methods, we divide them into three kinds of methods: model pruning, efficient architecture, and neural architecture search. Detailed analyses of each method are presented to demonstrate their advantages and disadvantages. Then, we comprehensively compare them in detail and propose many suggestions about the design of the efficient convolution neural network model structure. Inspired by these suggestions, we built a new efficient neural network model, SharedNet. And the SharedNet obtains the best accuracy of manually-designed efficient CNN models on the ImageNet dataset.

A Hybrid Physics-Based and Stochastic Neural Network Model Structure for Diesel Engine Combustion Events

Estimation of combustion phasing and power production is essential to ensuring proper combustion and load control. However, archetypal control-oriented physics-based combustion models can become computationally expensive if highly accurate predictive capabilities are achieved. Artificial neural network (ANN) models, on the other hand, may provide superior predictive and computational capabilities. However, using classical ANNs for model-based prediction and control can be challenging, since their heuristic and deterministic black-box nature may make them intractable or create instabilities. In this paper, a hybridized modeling framework that leverages the advantages of both physics-based and stochastic neural network modeling approaches is utilized to capture CA50 (the timing when 50% of the fuel energy has been released) along with indicated mean effective pressure (IMEP). The performance of the hybridized framework is compared to a classical ANN and a physics-based-only framework in a stochastic environment. To ensure high robustness and low computational burden in the hybrid framework, the CA50 input parameters along with IMEP are captured with a Bayesian regularized ANN (BRANN) and then integrated into an overall physics-based 0D Wiebe model. The outputs of the hybridized CA50 and IMEP models are then successively fine-tuned with BRANN transfer learning models (TLMs). The study shows that in the presence of a Gaussian-distributed model uncertainty, the proposed hybridized model framework can achieve an RMSE of 1.3 × 10−5 CAD and 4.37 kPa with a 45.4 and 3.6 s total model runtime for CA50 and IMEP, respectively, for over 200 steady-state engine operating conditions. As such, this model framework may be a useful tool for real-time combustion control where in-cylinder feedback is limited.

The Design of a Dual Servo Drive and Control Integrated Platform Based on VNet Neural Network

A dual-servo drive and control integrated platform based on VNet neural network development is aimed to address the concerns of poor realization function, operational stability, and lengthy reaction time in the presently built dual-servo drive and control integrated platform. The VNet neural network model structure is established, the VNet neural network model is trained, and the VNet neural network adaptive control method is introduced. Based on VNet neural network development, the hardware design of the dual servo drive control integrated platform has been completed through the overall hardware architecture of the platform, power drive circuit design, and control unit circuit design. The control program was developed using the C language based on a combination of STM32F4 library functions. Through the overall structure of the platform software, the design of functional modules and the design of the servo drive controller, the software design of the dual-servo drive and control integrated platform is completed, and the dual-servo drive and control integrated platform design is realized. The experimental results show that the proposed method design platform’s implementation function and operational stability are good, and the platform response time can be effectively shortened.

Modifying Hidden Layer in Neural Network Models to Improve Prediction Accuracy: A Combined Model for Estimating Stock Price

Investment experts, who deal with stock price estimation, commonly look for the most accurate and appropriate statistical techniques to make decisions on investment. The aim of this study is to improve the accuracy of stock price prediction models through modifying the structure of a combined neural network model with time-series data, in which the main contribution is to insert the time-series analysis prediction into the hidden layer of the neural network. The proposed structure is made up of neural networks and time-series analysis, with variable reduction used to remove attributes with inter-correlations. Data has been collected over six years (72 months) from the Iranian stock market, including the number of trades, new-coin price, gold-18 price, US Dollar and Euro equivalent currencies, oil-index price, Brent-oil price, industry index, and balanced stock index, followed by developing the prediction models. Comparing the performance criteria of the proposed structure to the traditional ones in terms of the mean square and mean absolute errors revealed that inserting time-series estimated variables into hidden layers would improve the performance of neural network models to estimate stock prices for making investment decisions. Doi: 10.28991/HIJ-2022-03-01-05 Full Text: PDF

Target recognition of heterogeneous fusion images based on improved YOLO model deep learning algorithm

Aiming at the problem of target recognition of heterogeneous fusion images, the algorithm flow and network structure of the YOLO convolutional neural network model are analyzed and explained based on the deep learning of the convolutional neural network. Combining the characteristics of the fusion image, the interface, structure and parameters of the YOLOv3 convolutional neural network are improved and adjusted while the recognition effect is guaranteed, the YOLO model is streamlined, and the training is successfully implemented and the training weights are output. The algorithm is effective and stable, and the recognition speed is faster.

Learning time-dependent PDEs with a linear and nonlinear separate convolutional neural network

Partial differential equations (PDEs) that describe many physical phenomena are discovered or derived based on professional knowledge or empirical observations. However, the rapid development of machine learning technology interests us to discover the PDEs with a data-driven approach based on neural networks. We introduce the convolution neural network (CNN) to learn the time-dependent PDEs. Instead of customizing a complex regularization term in loss function, we introduce a series of constraints on the structure of neural network model to avoid the overfitting problems. Firstly, since the linear PDEs can be accurately approximated by a linear CNN, we construct a network consisting of a linear CNN and a nonlinear CNN. The linear CNN helps to mitigate the overfitting problems of learning nonlinear PDEs. Secondly, the boundary conditions are hard encoded into the neural networks by custom padding operations. Finally, the time-series data are learned by an auto-regressive framework corresponding with the Euler scheme. We test the proposed framework with a series of PDEs, including heat equation, Burgers equation, reaction-diffusion equation and Kuramoto–Sivashinsky equation. Moreover, we also test the two-dimensional equations, and it shows similar accuracy as one-dimensional case. These numerical results demonstrate that the proposed framework is able to learn the PDEs from few data more accurately and stably.

In-Firm Planning and Business Processes Management Using Deep Neural Networks

Objective - The objective of this paper is to consider using machine learning approaches for in-firm processes prediction and to give an estimation of such values as effective production quantities. Methodology - The research methodology used is a synthesis of a deep-learning model, which is used to predict half of real business data for comparison with the remaining half. The structure of the convolutional neural network (CNN) model is provided, as well as the results of experiments with real orders, procurements, and income data. The key findings in this paper are that convolutional with a long-short-memory approach is better than a single convolutional method of prediction. Findings - This research also considers useof such technologies on business digital platforms. According to the results, there are guidelines formulated for the implementation in the particular ERP systems or web business platforms. Novelty - This paper describes the practical usage of 1-dimensional(1D) convolutional neural networks and a mixed approach with convolutional and long-short memory networks for in-firm planning tasks such as income prediction, procurements, and order demand analysis. Type of Paper - Empirical. Keywords: Business; Neural, Networks; CNN; Platform JEL Classification: C45

Prediction Method of College Students’ Psychological Pressure Based on Deep Neural Network

Aiming at the problems of low prediction accuracy and efficiency and poor prediction effect in the current psychological pressure prediction methods, a psychological pressure prediction method for college students based on deep neural network is proposed. The structure and algorithm of depth neural network and gray theory model are analyzed. Using the deep neural network, this paper establishes the sample set data of college students’ psychological pressure prediction and constructs the college students’ psychological pressure prediction model combined with the deep neural network algorithm of gray theory. The physical network information model is formed through the relationship between neurons. According to the dynamic changes of college students’ psychological pressure in each neuron of the physical network, the prediction of college students’ psychological pressure is completed. The experimental results show that the proposed method is effective in predicting college students’ psychological pressure and can effectively improve the accuracy and efficiency of college students’ psychological pressure prediction.

Research on the impact of trade uncertainty on national grain supply and risk cost control

ABSTRACT In order to explore the impact of trade uncertainty on the national grain supply, based on the BP neural network algorithm, this paper combines the characteristics of grain supply and trade data to construct an intelligent analysis model. Moreover, this paper uses historical data as the basic data to study the trade uncertainty and the relevant characteristics of grain supply, and combine the actual demand to construct the structure of the BP neural network model. After constructing the model, this paper analyzes the function of the model, combines the comparative analysis method to classify the data and studies the influence mechanism of trade uncertainty on the national grain supply. In addition, this paper combines the actual data to analyze the performance of the algorithm model of this paper, and the analysis results are consistent with historical data. Finally, this paper analyzes the cost control of national grain supply based on the results of the model analysis, and proposes several countermeasures.

Research on the Psychological Distribution Delay of Artificial Neural Network Based on the Analysis of Differential Equation by Inequality Expansion and Contraction Method

Abstract This article first introduces neural networks and their characteristics. Based on a comparison of the structure and function of biological neurons and artificial neurons, it focuses on the structure, classification, activation rules, and learning rules of neural network models. Based on the existing literature, this article adds a distributed time lag term of the neural network system. In the actual problem, history has a very important influence on the current change situation, and it is not only at a specific time in the past. It has an impact on the current state change rate. Therefore, based on the existing literature that only has discrete time lags, this paper adds distributed time lags. Such neural network systems can better reflect real-world problems. In this paper, we use three different inequality scaling methods to study the existence, uniqueness, and global asymptotic stability of a class of neural network systems with mixed delays and uncertain parameters. First, using the principle of homeomorphism, a new upper-norm norm is introduced for the correlation matrix of the neural network, and enough conditions for the existence of unique equilibrium points in several neural network systems are given. Under these conditions, the appropriate Lyapunov is used. Krasovskii functional, we prove that the equilibrium point of the neural network system is globally robust and stable. Numerical experiments show that the stability conditions of the neural network system we obtained are feasible, and the conservativeness of the stability conditions of the neural network system is reduced. Finally, some applications and problems of neural network models in psychology are briefly discussed.

Artificial Intelligence Analysis of EEG Amplitude in Intensive Heart Care.

This article first studied the morphological characteristics of the EEG for intensive cardiac care; that is, based on the analysis of the mechanism of disease diagnosis and treatment, a signal processing and machine learning model was constructed. Then, the methods of signal preprocessing, signal feature extraction, new neural network model structure, training mechanism, optimization algorithm, and efficiency are studied, and experimental verification is carried out for public data sets and clinical big data. Then, the principle of intensive cardiac monitoring, the mechanism of disease diagnosis, the types of arrhythmia, and the characteristics of the typical signal are studied, and the rhythm performance, individual variability, and neurophysiological basis of electrical signals in intensive cardiac monitoring are researched. Finally, the automatic signal recognition technology is studied. In order to improve the training speed and generalization ability, a multiclassification model based on Least Squares Twin Support Vector Machine (LS-TWIN-SVM) is proposed. The computational complexity of the classification model algorithm is compared, and intelligence is adopted. The optimization algorithm selects the parameters of the classifier and uses the EEG signal to simulate the model. Support Vector Machines and their improved algorithms have achieved the ultimum in shallow neural networks and have achieved good results in the classification and recognition of bioelectric signals. The LS-TWIN-SVM algorithm proposed in this paper has achieved good results in the classification and recognition of bioelectric signals. It can perform bioinformatics processing on intensive cardiac care EEG signals, systematically biometric information, diagnose diseases, the real-time detection, auxiliary diagnosis, and rehabilitation of patients.

COVSeg-NET: A deep convolution neural network for COVID-19 lung CT image segmentation.

COVID‐19 is a new type of respiratory infectious disease that poses a serious threat to the survival of human beings all over the world. Using artificial intelligence technology to analyze lung images of COVID‐19 patients can achieve rapid and effective detection. This study proposes a COVSeg‐NET model that can accurately segment ground glass opaque lesions in COVID‐19 lung CT images. The COVSeg‐NET model is based on the fully convolutional neural network model structure, which mainly includes convolutional layer, nonlinear unit activation function, maximum pooling layer, batch normalization layer, merge layer, flattening layer, sigmoid layer, and so forth. Through experiments and evaluation results, it can be seen that the dice coefficient, sensitivity, and specificity of the COVSeg‐NET model are 0.561, 0.447, and 0.996 respectively, which are more advanced than other deep learning methods. The COVSeg‐NET model can use a smaller training set and shorter test time to obtain better segmentation results.