Multi-scale physics-informed neural networks for solving high Reynolds number boundary layer flows based on matched asymptotic expansions

Multi-scale system remains a classical scientific problem in fluid dynamics,biology,etc.In the present study,a scheme of multi-scale Physics-informed neural networks is proposed to solve the boundary layer flow at high Reynolds numbers without any data.The flow is divided into several regions with different scales based on Prandtl's boundary theory.Different regions are solved with governing equations in different scales.The method of matched asymptotic expansions is used to make the flow field continuously.A flow on a semi infinite flat plate at a high Reynolds number is considered a multi-scale problem because the boundary layer scale is much smaller than the outer flow scale.The results are compared with the reference numerical solutions,which show that the msPINNs can solve the multi-scale problem of the boundary layer in high Reynolds number flows.This scheme can be developed for more multi-scale problems in the future.

Cascading Delays for the High-Speed Rail Network Under Different Emergencies:A Double Layer Network Approach

High-speed rail(HSR) has formed a networked operational scale in China. Any internal or external disturbance may deviate trains’ operation from the planned schedules, resulting in primary delays or even cascading delays on a network scale. Studying the delay propagation mechanism could help to improve the timetable resilience in the planning stage and realize cooperative rescheduling for dispatchers. To quickly and effectively predict the spatial-temporal range of cascading delays, this paper proposes a max-plus algebra based delay propagation model considering trains’ operation strategy and the systems’ constraints. A double-layer network based breadth-first search algorithm based on the constraint network and the timetable network is further proposed to solve the delay propagation process for different kinds of emergencies. The proposed model could deal with the delay propagation problem when emergencies occur in sections or stations and is suitable for static emergencies and dynamic emergencies. Case studies show that the proposed algorithm can significantly improve the computational efficiency of the large-scale HSR network. Moreover, the real operational data of China HSR is adopted to verify the proposed model, and the results show that the cascading delays can be timely and accurately inferred, and the delay propagation characteristics under three kinds of emergencies are unfolded.

Research on Physical Layer Security in Cognitive Wireless Networks with Multiple Eavesdroppers Based on Resource Allocation Algorithm

With the rapid development of the Internet of Things (IoT), non-Orthogonal Multiple Access (NOMA) technology and cognitive wireless network are two promising technologies to improve the spectral efficiency of the system, which have been widely concerned in the field of wireless communication. However, due to the importance of ownership and privacy protection, the IoT system must provide corresponding security mechanisms. From the perspective of improving the transmission security of CR-NOMA system based on cognitive wireless network, and considering the shortcomings of traditional relay cooperative NOMA system, this paper mainly analyzes the eavesdropping channel model of multi-user CR-NOMA system and derives the expressions of system security and rate to improve the security performance of CR-NOMA system. The basic idea of DC planning algorithm and the scheme of sub-carrier power allocation to improve the transmission security of the system were introduced. An algorithm for DC-CR-NOMA was proposed to maximize the SSR of the system and minimize the energy loss. The simulation results show that under the same complexity, the security and speed of the system can be greatly improved compared with the traditional scheme.

Feed-Forward Neural Network Based Petroleum Wells Equipment Failure Prediction

In the oil industry, the productivity of oil wells depends on the performance of the sub-surface equipment system. These systems often have problems stemming from sand, corrosion, internal pressure variation, or other factors. In order to ensure high equipment performance and avoid high-cost losses, it is essential to identify the source of possible failures in the early stage. However, this requires additional maintenance fees and human power. Moreover, the losses caused by these problems may lead to interruptions in the whole production process. In order to minimize maintenance costs, in this paper, we introduce a model for predicting equipment failure based on processing the historical data collected from multiple sensors. The state of the system is predicted by a Feed-Forward Neural Network (FFNN) with an SGD and Backpropagation algorithm is applied in the training process. Our model’s primary goal is to identify potential malfunctions at an early stage to ensure the production process’ continued high performance. We also evaluated the effectiveness of our model against other solutions currently available in the industry. The results of our study show that the FFNN can attain an accuracy score of 97% on the given dataset, which exceeds the performance of the models provided.

Multi-Scale-Matching neural networks for thin plate bending problem

Physics-informed neural networks are a useful machine learning method for solving differential equations,but encounter challenges in effectively learning thin boundary layers within singular perturbation problems.To resolve this issue,multi-scale-matching neural networks are proposed to solve the singular perturbation problems.Inspired by matched asymptotic expansions,the solution is decomposed into inner solutions for small scales and outer solutions for large scales,corresponding to boundary layers and outer regions,respectively.Moreover,to conform neural networks,we introduce exponential stretched variables in the boundary layers to avoid semiinfinite region problems.Numerical results for the thin plate problem validate the proposed method.

Application of Convolutional Neural Networks in Classification of GBM for Enhanced Prognosis

The lethal brain tumor “Glioblastoma” has the propensity to grow over time. To improve patient outcomes, it is essential to classify GBM accurately and promptly in order to provide a focused and individualized treatment plan. Despite this, deep learning methods, particularly Convolutional Neural Networks (CNNs), have demonstrated a high level of accuracy in a myriad of medical image analysis applications as a result of recent technical breakthroughs. The overall aim of the research is to investigate how CNNs can be used to classify GBMs using data from medical imaging, to improve prognosis precision and effectiveness. This research study will demonstrate a suggested methodology that makes use of the CNN architecture and is trained using a database of MRI pictures with this tumor. The constructed model will be assessed based on its overall performance. Extensive experiments and comparisons with conventional machine learning techniques and existing classification methods will also be made. It will be crucial to emphasize the possibility of early and accurate prediction in a clinical workflow because it can have a big impact on treatment planning and patient outcomes. The paramount objective is to not only address the classification challenge but also to outline a clear pathway towards enhancing prognosis precision and treatment effectiveness.

A multi-source information fusion layer counting method for penetration fuze based on TCN-LSTM

When employing penetration ammunition to strike multi-story buildings,the detection methods using acceleration sensors suffer from signal aliasing,while magnetic detection methods are susceptible to interference from ferromagnetic materials,thereby posing challenges in accurately determining the number of layers.To address this issue,this research proposes a layer counting method for penetration fuze that incorporates multi-source information fusion,utilizing both the temporal convolutional network(TCN)and the long short-term memory(LSTM)recurrent network.By leveraging the strengths of these two network structures,the method extracts temporal and high-dimensional features from the multi-source physical field during the penetration process,establishing a relationship between the multi-source physical field and the distance between the fuze and the target plate.A simulation model is developed to simulate the overload and magnetic field of a projectile penetrating multiple layers of target plates,capturing the multi-source physical field signals and their patterns during the penetration process.The analysis reveals that the proposed multi-source fusion layer counting method reduces errors by 60% and 50% compared to single overload layer counting and single magnetic anomaly signal layer counting,respectively.The model's predictive performance is evaluated under various operating conditions,including different ratios of added noise to random sample positions,penetration speeds,and spacing between target plates.The maximum errors in fuze penetration time predicted by the three modes are 0.08 ms,0.12 ms,and 0.16 ms,respectively,confirming the robustness of the proposed model.Moreover,the model's predictions indicate that the fitting degree for large interlayer spacings is superior to that for small interlayer spacings due to the influence of stress waves.

A Kind of Second-Order Learning Algorithm Based on Generalized Cost Criteria in Multi-Layer Feed-Forward Neural Networks

A kind of second order algorithm--recursive approximate Newton algorithm was given by Karayiannis. The algorithm was simplified when it was formulated. Especially, the simplification to matrix Hessian was very reluctant, which led to the loss of valuable information and affected performance of the algorithm to certain extent. For multi layer feed forward neural networks, the second order back propagation recursive algorithm based generalized cost criteria was proposed. It is proved that it is equivalent to Newton recursive algorithm and has a second order convergent rate. The performance and application prospect are analyzed. Lots of simulation experiments indicate that the calculation of the new algorithm is almost equivalent to the recursive least square multiple algorithm. The algorithm and selection of networks parameters are significant and the performance is more excellent than BP algorithm and the second order learning algorithm that was given by Karayiannis.

Exploring the Multi-Layer Structural Properties of the Bus-Subway Transportation Network of Shanghai

Buses and subways are essential to urban public transportation systems and an important engine for activating high-quality urban development. Traditional multi-modal transportation networks focus on the structural feature mining of single-layer networks or each layer, ignoring the structural association of multi-layer networks. In this paper, we examined the multi-layer structural property of the bus-subway network of Shanghai at both global and nodal scales. A dual-layer model of the city’s bus and subway system was built. Single-layer complex network indicators were also extended. The paper also explored the spatial coupling properties of the city’s bus and subway system and identified its primary traffic nodes. It was found that 1) the dual-layer network increased the network’s connectivity to a certain extent and broke through the spatial limitation in terms of physical structure, making the connection between any two locations more direct. 2) The dual-layer network changed the topological characteristics of the transit network, increasing the centrality value and bit order in degree centrality, betweenness centrality, and closeness centrality to different degrees, and making each centrality tend to converge to the city center in spatial distribution. Enhancing the management of critical network nodes would help the integrated public transportation system operate more effectively and provide higher-quality services.

A Tractable Approach to Analyzing the Physical-Layer Security in K-Tier Heterogeneous Cellular Networks

This study proposes a tractable approach to analyze the physical-layer security in the downlink of a multi-tier heterogeneous cellular network. This method is based on stochastic geometry, has low computational complexity, and uses the two-dimensional Poisson point process to model the locations of K-tier base stations and receivers, including those of legitimate users and eavesdroppers. Then, the achievable secrecy rates for an arbitrary user are determined and the upper and lower bounds of secrecy coverage probability derived on the condition that cross-tier interference is the main contributor to aggregate interference. Finally, our analysis results reveal the innate connections between information-theoretic security and the spatial densities of legitimate and malicious nodes.

Cooperative Jamming for Physical Layer Security in Hybrid Satellite Terrestrial Relay Networks

To integrate the satellite communications with the LTE/5G services, the concept of Hybrid Satellite Terrestrial Relay Networks(HSTRNs) has been proposed. In this paper, we investigate the secure transmission in a HSTRN where the eavesdropper can wiretap the transmitted messages from both the satellite and the intermediate relays. To effectively protect the message from wiretapping in these two phases, we consider cooperative jamming by the relays, where the jamming signals are optimized to maximize the secrecy rate under the total power constraint of relays. In the first phase, the Maximal Ratio Transmission(MRT) scheme is used to maximize the secrecy rate, while in the second phase, by interpolating between the sub-optimal MRT scheme and the null-space projection scheme, the optimal scheme can be obtained via an efficient one-dimensional searching method. Simulation results show that when the number of cooperative relays is small, the performance of the optimal scheme significantly outperforms that of MRT and null-space projection scheme. When the number of relays increases, the performance of the null-space projection approaches that of the optimal one.

THE APPLICATION OF ARTIFICIAL NEURAL NETWORKS TO INVESTIGATION ON THE THICKNESS OF INTERMETALLIC LAYER UNDER SOLID-LIQUID PRESSURE BONDING OF STEEL AND ALUMINIUM

Artificial neural networks (ANN), being a sophisticated type of information processing system by imitating the neural system of human brain, can be used to investigate the effects of concentration of flux solution, temperature of liquid aluminium, temperture of tools and pressure on thickness of the intermetallic layer at the interface between steel and aluminium under solid-liquid pressure bonding of steel and aluminium perfectly. The optimum thickness has been determined according to the value of the optimum shearing strength.

Deep Learning Based Physical Layer Security of D2D Underlay Cellular Network

In order to improve the physical layer security of the device-to-device(D2D)cellular network,we propose a collaborative scheme for the transmit antenna selection and the optimal D2D pair establishment based on deep learning.Due to the mobility of users,using the current channel state information to select a transmit antenna or establish a D2D pair for the next time slot cannot ensure secure communication.Therefore,in this paper,we utilize the Echo State Network(ESN)to select the transmit antenna and the Long Short-Term Memory(LSTM)to establish the D2D pair.The simulation results show that the LSTMbased and ESN-based collaboration scheme can effectively improve the security capacity of the cellular network with D2D and increase the life of the base station.

Layered learning of soccer robot based on artificial neural network

Discusses the application of artificial neural network for MIROSOT, introduces a layered model of BP network of soccer robot for learning basic behavior and cooperative behavior, and concludes from experimental results that the model is effective.

An Interpretable Denoising Layer for Neural Networks Based on Reproducing Kernel Hilbert Space and its Application in Machine Fault Diagnosis

Deep learning algorithms based on neural networks make remarkable achievements in machine fault diagnosis,while the noise mixed in measured signals harms the prediction accuracy of networks.Existing denoising methods in neural networks,such as using complex network architectures and introducing sparse techniques,always suffer from the difficulty of estimating hyperparameters and the lack of physical interpretability.To address this issue,this paper proposes a novel interpretable denoising layer based on reproducing kernel Hilbert space(RKHS)as the first layer for standard neural networks,with the aim to combine the advantages of both traditional signal processing technology with physical interpretation and network modeling strategy with parameter adaption.By investigating the influencing mechanism of parameters on the regularization procedure in RKHS,the key parameter that dynamically controls the signal smoothness with low computational cost is selected as the only trainable parameter of the proposed layer.Besides,the forward and backward propagation algorithms of the designed layer are formulated to ensure that the selected parameter can be automatically updated together with other parameters in the neural network.Moreover,exponential and piecewise functions are introduced in the weight updating process to keep the trainable weight within a reasonable range and avoid the ill-conditioned problem.Experiment studies verify the effectiveness and compatibility of the proposed layer design method in intelligent fault diagnosis of machinery in noisy environments.

A Physical Layer Network Coding Based Tag Anti-Collision Algorithm for RFID System

In RFID(Radio Frequency IDentification)system,when multiple tags are in the operating range of one reader and send their information to the reader simultaneously,the signals of these tags are superimposed in the air,which results in a collision and leads to the degrading of tags identifying efficiency.To improve the multiple tags’identifying efficiency due to collision,a physical layer network coding based binary search tree algorithm(PNBA)is proposed in this paper.PNBA pushes the conflicting signal information of multiple tags into a stack,which is discarded by the traditional anti-collision algorithm.In addition,physical layer network coding is exploited by PNBA to obtain unread tag information through the decoding operation of physical layer network coding using the conflicting information in the stack.Therefore,PNBA reduces the number of interactions between reader and tags,and improves the tags identification efficiency.Theoretical analysis and simulation results using MATLAB demonstrate that PNBA reduces the number of readings,and improve RFID identification efficiency.Especially,when the number of tags to be identified is 100,the average needed reading number of PNBA is 83%lower than the basic binary search tree algorithm,43%lower than reverse binary search tree algorithm,and its reading efficiency reaches 0.93.

SLCRM： Subjective Logic-Based Cross-Layer Reputation Mechanism for Wireless Mesh Networks

Wireless Mesh Networks (WMNs) are vulnerable to various security threats because of their special infrastructure and communication mode, wherein insider attacks are the most challenging issue. To address this problem and protect innocent users from malicious attacks, it is important to encourage cooperation and deter malicious behaviors. Reputation systems constitute a major category of techniques used for managing trust in distributed networks, and they are effective in characterizing and quantifying a node's behavior for WMNs. However, conventional layered reputation mechanisms ignore several key factors of reputation in other layers; therefore, they cannot provide optimal performance and accurate malicious node identification and isolation for WMNs. In this paper, we propose a novel dynamic reputation mechanism, SLCRM, which couples reputation systems with a cross-layer design and node-security-rating classification techniques to dynamically detect and restrict insider attacks. Simulation results show that in terms of network throughput, packet delivery ratio, malicious nodes identification, and success rates, SLCRM implements security protection against insider attacks in a more dynamic, effective, and efficient manner than the subjective logic and uncertainty-based reputation model and the familiarity-based reputation model.

Feed-Forward Artificial Neural Network Model for Air Pollutant Index Prediction in the Southern Region of Peninsular Malaysia

This paper describes the application of principal component analysis (PCA) and artificial neural network (ANN) to predict the air pollutant index (API) within the seven selected Malaysian air monitoring stations in the southern region of Peninsular Malaysia based on seven years database (2005-2011). Feed-forward ANN was used as a prediction method. The feed-forward ANN analysis demonstrated that the rotated principal component scores (RPCs) were the best input parameters to predict API. From the 4 RPCs, only 10 (CO, O3, PM10, NO2, CH4, NmHC, THC, wind direction, humidity and ambient temp) out of 12 prediction variables were the most significant parameters to predict API. The results proved that the ANN method can be applied successfully as tools for decision making and problem solving for better atmospheric management.

Laplacian energy maximizationfor multi-layer air transportation networks

To increase airspace capacity, alleviate flight delay,and improve network robustness, an optimization method ofmulti-layer air transportation networks is put forward based onLaplacian energy maximization. The effectiveness of takingLaplacian energy as a measure of network robustness isvalidated through numerical experiments. The flight routesaddificm optimization model is proposed with the principle ofmaximizing Laplacian energy. Three methods including thedepth-first search （DFS） algorithm, greedy algorithm andMonte-Carlo tree search （MCTS） algorithm are applied tosolve the proposed problem. The trade-off between systemperformance and computational efficiency is compared throughsimulation experiments. Finally, a case study on Chineseairport network （CAN） is conducted using the proposedmodel. Through encapsulating it into multi-layer infrastructurevia k-core decomposition algorithm, Laplacian energymaximization for the sub-networks is discussed which canprovide a useful tool for the decision-makers to optimize therobustness of the air transoortation network on different scales.

Distributed Contact Plan Design for Multi-Layer Satellite-Terrestrial Network

In multi-layer satellite-terrestrial network, Contact Graph Routing(CGR) uses the contact information among satellites to compute routes. However, due to the resource constraints in satellites, it is extravagant to configure lots of the potential contacts into contact plans. What's more, a huge contact plan makes the computing more complex, which further increases computing time. As a result, how to design an efficient contact plan becomes crucial for multi-layer satellite network, which usually has a large scaled topology. In this paper, we propose a distributed contact plan design scheme for multi-layer satellite network by dividing a large contact plan into several partial parts. Meanwhile, a duration based inter-layer contact selection algorithm is proposed to handle contacts disruption problem. The performance of the proposed design was evaluated on our Identifier/Locator split based satellite-terrestrial network testbed with 79 simulation nodes. Experiments showed that the proposed design is able to reduce the data delivery delay.