Automated Aircraft Visual Inspection With Artificial Data Generation Enabled Deep Learning,
2022
Air Force Institute of Technology
Automated Aircraft Visual Inspection With Artificial Data Generation Enabled Deep Learning, Nathan J. Gaul
Theses and Dissertations
Aircraft visual inspection, which is essential to daily maintenance of an aircraft, is expensive and time-consuming to perform. Augmenting trained maintenance technicians with automated UAVs to collect and analyze images for aircraft inspection is an active research topic and a potential application of CNNs. Training datasets for niche research topics such as aircraft visual inspection are small and challenging to produce, and the manual process of labeling these datasets often produces subjective annotations. Recently, researchers have produced several successful applications of artificially generated datasets with domain randomization for training CNNs for real-world computer vision problems. The research outlined herein builds …
Application Of Machine Learning Models With Numerical Simulations Of An Experimental Microwave Induced Plasma Gasification Reactor,
2022
Air Force Institute of Technology
Application Of Machine Learning Models With Numerical Simulations Of An Experimental Microwave Induced Plasma Gasification Reactor, Owen D. Sedej
Theses and Dissertations
This thesis aims to contribute to the future development of this technology by providing an in-depth literature review of how this technology physically operates and can be numerically modeled. Additionally, this thesis reviews literature of machine learning models that have been applied to gasification to make accurate predictions regarding the system. Finally, this thesis provides a framework of how to numerically model an experimental plasma gasification reactor in order to inform a variety of machine learning models.
Analysis Of Generalized Artificial Intelligence Potential Through Reinforcement And Deep Reinforcement Learning Approaches,
2022
Air Force Institute of Technology
Analysis Of Generalized Artificial Intelligence Potential Through Reinforcement And Deep Reinforcement Learning Approaches, Jonathan Turner
Theses and Dissertations
Artificial Intelligence is the next competitive domain; the first nation to develop human level artificial intelligence will have an impact similar to the development of the atomic bomb. To maintain the security of the United States and her people, the Department of Defense has funded research into the development of artificial intelligence and its applications. This research uses reinforcement learning and deep reinforcement learning methods as proxies for current and future artificial intelligence agents and to assess potential issues in development. Agent performance were compared across two games and one excursion: Cargo Loading, Tower of Hanoi, and Knapsack Problem, respectively. …
Design Demand Trend Acquisition Method Based On Short Text Mining Of User Comments In Shopping Websites,
2022
School of Design, South China University of Technology, Guangzhou, 510006, China
Design Demand Trend Acquisition Method Based On Short Text Mining Of User Comments In Shopping Websites, Zhiyong Xiong, Zhaoxiong Yan, Huanan Yao, Shangsong Liang
Machine Learning Faculty Publications
In order to facilitate designers to explore the market demand trend of laptops and to establish a better “network users-market feedback mechanism”, we propose a design and research method of a short text mining tool based on the K-means clustering algorithm and Kano mode. An improved short text clustering algorithm is used to extract the design elements of laptops. Based on the traditional questionnaire, we extract the user’s attention factors, score the emotional tendency, and analyze the user’s needs based on the Kano model. Then, we select 10 laptops, process them by the improved algorithm, cluster the evaluation words and …
An Ensemble Approach For Patient Prognosis Of Head And Neck Tumor Using Multimodal Data,
2022
Mohamed Bin Zayed University of Artificial Intelligence
An Ensemble Approach For Patient Prognosis Of Head And Neck Tumor Using Multimodal Data, Numan Saeed, Roba Al Majzoub, Ikboljon Sobirov, Mohammad Yaqub
Computer Vision Faculty Publications
Accurate prognosis of a tumor can help doctors provide a proper course of treatment and, therefore, save the lives of many. Tradi-tional machine learning algorithms have been eminently useful in crafting prognostic models in the last few decades. Recently, deep learning algorithms have shown significant improvement when developing diag-nosis and prognosis solutions to different healthcare problems. However, most of these solutions rely solely on either imaging or clinical data. Utilizing patient tabular data such as demographics and patient med-ical history alongside imaging data in a multimodal approach to solve a prognosis task has started to gain more interest recently and …
Modeling & Simulation Based System Of Systems Engineering,
2022
1.Beihang University, Beijing, 100191, China;2.Engineering Research Center of Complex Product Advanced Manufacturing System, Ministry of Education, Beijing, 100191, China;
Modeling & Simulation Based System Of Systems Engineering, Lin Zhang, Kunyu Wang, Yuanjun Laili, Lei Ren
Journal of System Simulation
Abstract: To accommodate the dark and unstructured underwater working environment, the near-body pressure distribution characteristics of a bionic robot fish undulating in near wall region is studied. The feasibility of using artificial lateral line(ALL) to estimate the wall effect and flow field parameters is analyzed theoretically. A CFD(computational fluid dynamics) coupled solution model for near-body pressure simulation of a bionic robotic fish swimming near the wall is established and a near-body pressure data extraction and processing method is proposed. The effect of the wall clearance, inlet flow velocity and strouhal number (St) on the fish near-body pressure distribution …
Transmission Line Operation And Inspection Training Simulation Based On Multiple Time Scales And Vr,
2022
1.State Grid Hebei Electric Power Co. Ltd. Training Center, Shijiazhuang, 050023, China;
Transmission Line Operation And Inspection Training Simulation Based On Multiple Time Scales And Vr, Jiawen Yan, Jijie Huang, Lie Zhou, Changjin Chen, Qiang Wu, Jintao Zhao
Journal of System Simulation
Abstract: Social learning is defined as the process that consumers use online reviews to fetch more precise information about product quality.Consequently, consumers would be more likely to purchase the product if the product quality learned was higher than their expectation, reference point effect named.To understand the impact of this effect in social learning on a firm's product decisions, we built a multi-agent model to solve the problem through simulation. According to the results, the reference point effect has a negative influence on the firm. The firm has to higher the product quality and price and therefore loses some profits. …
Product Decisions In Presence Of Social Learning And Reference Point Effect,
2022
1.School of Business Administration, South China University of Technology, Guangzhou 510640, China;
Product Decisions In Presence Of Social Learning And Reference Point Effect, Feng Li, Ying Wei
Journal of System Simulation
Abstract: In order to find out the source of VOCs(volatile organic compounds) emission and diffusion to the target area, and prevent the target area from further pollution, this paper propose an analytical method of VOCs hazard causes in related areas based on object function Petri net. The net structure describes the relationship between the potential pollution sources and the target area, and the operation of the net system reflects the change of VOCs hazard degree in the target area, and the calculation of hazard degree is integrated into the operation of the Petri net system. Through the actual case study, …
Lod Modeling Method For Three-Dimensional Objects With Energy Operator,
2022
1.School of Civil and Surveying & Mapping Engineering, Jiangxi University of Science and Technology, Ganzhou 341000, China;
Lod Modeling Method For Three-Dimensional Objects With Energy Operator, Yongzhi Wang, Zhenchao Li, Pengyu Liu, Hui Wang
Journal of System Simulation
Abstract: Current level of detail (LOD) modeling methods do not combine simplification and subdivision, which results in models not being rich in detail level. Therefore, the energy operator was used to combine the simplification algorithm of a three-dimensional (3D) object model with the subdivision algorithm, and a LOD modeling method based on an energy operator was proposed for 3D objects. The method involves three major steps: calculation of energy operators for 3D models, model simplification and model subdivision based on an energy operator. Experiments show that this method can generate models with rich levels of detail and has good visualization …
Job Scheduling And Simulation In Cloud Based On Deep Reinforcement Learning,
2022
1.College of Computer Science, Guangdong University of Petrochemical Technology, Maoming 525000, China;
Job Scheduling And Simulation In Cloud Based On Deep Reinforcement Learning, Qirui Li, Xinyi Peng
Journal of System Simulation
Abstract: To solve the difficulty in job scheduling in the complex and transient multi-user, multi-queue, and multi-data-center cloud computing environment, this paper proposed a job scheduling method based on deep reinforcement learning. A system model of cloud job scheduling and its mathematical model were built, and an optimization goal consisting of transmission time, waiting time, and execution time was obtained. A job scheduling algorithm based on deep reinforcement learning was designed, and its state space, action space, and reward function were given. A simulated cloud job scheduler was designed and developed, and simulated scheduling experiments were conducted on it. The …
Research On Nonlinear Evaluation Method Of Situational Hot Spots,
2022
College of Command and Control Engineering, Army Engineering University of PLA, Nanjing 210007, China;
Research On Nonlinear Evaluation Method Of Situational Hot Spots, Tengjiao Mao, Dongge Zhang, Xuefeng Liang, Yanjie Niu, Minggang Yu, Ming He
Journal of System Simulation
Abstract: The excessive amount of data and information in the situation map is likely to produce a huge cognitive load that exceeds the physiological limit. This can lead to delays in the perception, judgment, and decision-making of the commander, or even failure of decision-making activities in severe cases. For this reason, situation information needs to be processed to assess and highlight situational hotspots so that cognitive overload can be tackled. Based on the aggregation factors of collaborative targets, this paper takes the effective impact possibility of situational targets and the collaboration as the indicators, derives the attention function, and designs …
Research On Semantic Segmentation Of Natural Landform Based On Edge Detection Module,
2022
School of Mechanical Engineering, Hebei University of Technology, Tianjin 300130, China;
Research On Semantic Segmentation Of Natural Landform Based On Edge Detection Module, Qizong Shen, Chunyan Gao
Journal of System Simulation
Abstract: To classify pixels of natural landform edges in remote sensing images, this paper proposes a multi-channel fusion model and a decoder-side module model both integrating an edge detection module. The edge detection module takes the Canny operator as the base to perform closed operations and mean filtering, as a result of which accurate image edges can be achieved. Based on DeepLabV3+, the semantic segmentation network is connected with an edge planning module in parallel at encoder and decoder sides respectively. The experimental results show that the two improved networks can achieve a better segmentation effect on a high-resolution natural …
Research On Collaborative Computing Offloading Model For Base Station Groups Based On Fireworks Algorithm,
2022
1.Nanjing University of Posts and Telecommunications, Nanjing 210003, China;2.Nanjing Pharmaceutical Co. , Ltd. , Nanjing 210012, China;3.Jiangsu Key Laboratory of Data Science and Smart Software, Jinling Institute of Technology, Nanjing 211169, China;
Research On Collaborative Computing Offloading Model For Base Station Groups Based On Fireworks Algorithm, Bin Xu, Wenqing Yan, Zhuofan Han, Guangshen He, Tao Deng, Yunkai Zhao, Jin Qi
Journal of System Simulation
Abstract: Internet of Vehicles (IoV), AR, AI, and other computing-intensive, time-delay-sensitive applications are developing rapidly. However, due to the relatively insufficient computing capacity of mobile devices, such application tasks face serious latency, which seriously affects user experience and even fails to meet the needs of users. To solve this problem, by comprehensively considering delays and costs, we propose a cooperative computing offloading model based on a multi-user and multi-mobile edge computing (multi-MEC) server for base station groups. In addition, an improved fireworks algorithm based on convex optimization (CVX-FWA) is presented to solve the model and perform reasonable offloading and resource …
Energy Consumption Prediction For Air-Conditioning System Based On Dynamic Temperature Control,
2022
School of Science, Xi'an University of Architecture and Technology, Xi'an, 710055, China;
Energy Consumption Prediction For Air-Conditioning System Based On Dynamic Temperature Control, Yan Bai, Lulu Wu, Yin'e He, Yuying Wang
Journal of System Simulation
Abstract: To solve the problem of energy consumption prediction for air-conditioning systems implementing dynamic temperature control, we designed a dynamic temperature control strategy and obtained a dataset on the hourly energy consumption of the air-conditioning system through EnergyPlus simulation. An improved particle swarm optimization-back propagation neural network (IPSO-BPNN) prediction model was built on the basis of energy consumption analysis by an integrated method. Clustering, classification, and correlation analysis methods were integrated to mine the energy consumption pattern of the air-conditioning system and determine the input variables for the prediction model. A nonlinear change strategy was designed to adjust the inertia …
Optimal Path Planning For Multi-Stage Automatic Parking And Simulation Analysis,
2022
School of Mechanical Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China;
Optimal Path Planning For Multi-Stage Automatic Parking And Simulation Analysis, Qiming Wang, Gaoqiang Zong, Jinming Xu
Journal of System Simulation
Abstract: To resolve the path planning for narrow parallel parking spaces and the discontinuous curvature of the parking trajectory, this paper proposes a method of optimal multi-stage parking path planning considering collision avoidance constraints. A trajectory equation for the center of the vehicle rear axle is derived for the case when the steering wheel speed is constant. A function of collision avoidance constraints is developed to ensure the safe parking of the vehicle. With the center of the rear axle of the parking path as the control point, the optimal path is solved according to parking indicators such as the …
Multi-Floor Evacuation Model Based On Wavelet Neural Network,
2022
1.School of Computer Science, Chengdu Normal University, Chengdu 611130, China;2.Key Laboratory of Interior Layout Optimization and Security, Institutions of Higher Education of Sichuan Province, Chengdu Normal University, Chengdu 611130, China;
Multi-Floor Evacuation Model Based On Wavelet Neural Network, Juan Wei, Lei You, Yangyong Guo, Zhihai Tang
Journal of System Simulation
Abstract: Crowd evacuation in a multi-floor environment is a popular social concern, while the stagnation phenomenon easily occurs when simulating a multi-floor complex environment with the traditional social force model. Therefore, An improved social force model is proposed by a wavelet neural network, and a new multi-floor evacuation model is built. In the model, a pedestrian's direction of movement is obtained by the field model, which is used as the self-driving direction of the social force model. Meanwhile, the evaluation indexes of the exit congestion degree, path congestion degree, and average velocity in a multi-floor environment are given, and a …
Zoomfft-Based Demodulation Algorithm For Underwater Acoustic Ofdm Signals,
2022
1.School of Electronic Engineering and Automation, Guilin University of Electronic Technology, Guilin 541004, China;2.Guangxi Key Laboratory of Automatic Detection Technology and Instruments, Guilin 541004, China;
Zoomfft-Based Demodulation Algorithm For Underwater Acoustic Ofdm Signals, Qing Guo, Angdi Li, Jing Wu, Haitao Su
Journal of System Simulation
Abstract: The picket fence effect of fast Fourier transform (FFT) restricts the demodulation performance of the underwater acoustic(UWA) communication systemusing orthogonal frequency division multiplexing (OFDM). To solve this problem, we propose a demodulation algorithm based on ZoomFFT. Specifically, the received signal is processed by frequency shifting and downsampling forarefined spectrum, which improves the spectralresolution and weakens the picketfence effect.Meanwhile, the channel response is refined, and the channel equalization algorithm is constructed on the basis of the minimum mean square error (MMSE) principle to eliminate the channel influence. Simulations show that the performance of underwater acoustic OFDM demodulation algorithm based on …
Design Of Fatigue Driving Detection System Based On Cascaded Neural Network,
2022
1.College of Engineering, Zunyi Normal University, Zunyi 563006, China;
Design Of Fatigue Driving Detection System Based On Cascaded Neural Network, Bangqian Ao, Sha Yang, Jinqing Linghu, Zhenhuan Ye
Journal of System Simulation
Abstract: An algorithm is proposed to greatly improves the face detection rate and ensures the accuracy by adjusting the size of input images, expanding the minimum face size, and reducing the scaling ratio between layers of the detection window. The detection efficiency of this algorithm is 18 times higher than that of the original MTCNN. By building a new CNN structure model for the detection of eyes and mouths, we can achieve network detection accuracy of 95.6%. The proposed network is cascaded with the original MTCNN to continue classifying and locating the eyes and mouth in the formerly …
Segmentation Line Detection In Dental Model Based On Target Region Constraint,
2022
College of Computer Science & Technology, Xi'an University of Science and Technology, Xi'an 710054, China;
Segmentation Line Detection In Dental Model Based On Target Region Constraint, Tian Ma, Yun Li, Jiaojiao Li, Yuancheng Li
Journal of System Simulation
Abstract: It is an important pretreatment of a virtual orthodontic system to accurately segment teeth from a dental model. In the present methods, all patches are computed directly. To solve this problem, this paper proposes a segmentation line detection method based on target region constraint, which narrows down the detection range to the area around the actual segmentation line. In this method, the cutting plane and the cutting line are automatically formed according to the positions of seed points. The detection range is determined by the search for the position with the greatest negative curvature on the cutting line. The …
Constructing The Agent Discrete Simulation Based On Devs Atomic Model,
2022
1.Beihang University, Beijing 100191, China;2.Engineering Research Center of Complex Product Advanced Manufacturing Systems, Ministry of Education, Beijing 100191, China;
Constructing The Agent Discrete Simulation Based On Devs Atomic Model, Xiaohan Wang, Lin Zhang, Yuanjun Laili, Kunyu Xie, Tingchun Hu
Journal of System Simulation
Abstract: In order to resolve the nonlinear and ill-posed inverse problem of the image reconstruction of electrical capacitance tomography (ECT), an image reconstruction algorithm based on one-dimensional convolutional neural network (1D CNN) is presented. The nonlinear mapping relationship between the independent measurement value of ECT system and the gray value of reconstructed image is established by 1D CNN. Six typical flow regimes with random distribution are obtained by the finite element simulation software and a 1D CNN is successfully trained. Simulation and static experiments are carried out and the reconstructed images using linear back projection, Landweber iterative algorithm and 1D …
