Derivation and validation of wind tunnel free-flight similarity law for store separation from aircraft
اشتقاق و اعتبار قانون تشابه پرواز آزاد تونل باد برای جداسازی فروشگاه از هواپیما-2020
This paper describes a design method for a similarity law for free-flight tests of aircraft load separation. The effect of the initial separation velocity on the motion similarity is considered. For the first time, the initial separation velocity is introduced into the equation of motion to identify similar trajectories. Finally, the model mass parameter characteristics and separation velocity equation are solved to determine similarity laws for wind tunnel tests, greatly improving the accuracy and applicability of test results from wind tunnels. The proposed derivation overcomes the problems faced by the traditional light model method and the traditional heavy model method, namely that they are limited in terms of ejection separation and cannot be realized in wind tunnel tests. The typical separation state under wind load scenarios is simulated using computational fluid dynamics (CFD). Separation data from real aircraft and previous test methods are compared with the simulation data obtained by the new similarity law design method. The improvement of the new similarity law in terms of trajectory simulation is verified through a comprehensive data comparison. The data show that the new similarity law greatly improves the accuracy of wind tunnel tests.
Keywords: Similarity law derivation | High-speed weapon delivery | Carrier and missile interference | Multi-body separation | Free-flight wind tunnel test
Wake modeling of wind turbines using machine learning
مدل سازی توربین های بادی با استفاده از یادگیری ماشین-2020
In the paper, a novel framework that employs the machine learning and CFD (computational fluid dynamics) simulation to develop new wake velocity and turbulence models with high accuracy and good efficiency is proposed to improve the turbine wake predictions. An ANN (artificial neural network) model based on the backpropagation (BP) algorithm is designed to build the underlying spatial relationship between the inflow conditions and the three-dimensional wake flows. To save the computational cost, a reduced-order turbine model ADM-R (actuator disk model with rotation), is incorporated into RANS (Reynolds-averaged Navier-Stokes equations) simulations coupled with a modified k − ε turbulence model to provide big datasets of wake flow for training, testing, and validation of the ANN model. The numerical framework of RANS/ADM-R simulations is validated by a standalone Vestas V80 2MW wind turbine and NTNU wind tunnel test of double aligned turbines. In the ANN-based wake model, the inflow wind speed and turbulence intensity at hub height are selected as input variables, while the spatial velocity deficit and added turbulence kinetic energy (TKE) in wake field are taken as output variables. The ANN-based wake model is first deployed to a standalone turbine, and then the spatial wake characteristics and power generation of an aligned 8-turbine row as representation of Horns Rev wind farm are also validated against Large Eddy Simulations (LES) and field measurement. The results of ANNbased wake model show good agreement with the numerical simulations and measurement data, indicating that the ANN is capable of establishing the complex spatial relationship between inflow conditions and the wake flows. The machine learning techniques can remarkably improve the accuracy and efficiency of wake predictions.
Keywords: Wind turbine wake | Wake model | Artificial neural network (ANN) | Machine learning | ADM-R (actuator-disk model with rotation) | model | Computational fluid dynamics (CFD)
AI Aided Noise Processing of Spintronic Based IoT Sensor for Magnetocardiography Application
پردازش نویز به کمک هوش مصنوعی مبتنی بر حسگر اینترنت اشیا بر Spintronic برای کاربرد مغناطیسی قلب-2020
As we are about to embark upon the highly hyped “Society 5.0”, powered by the Internet of Things (IoT), traditional ways to monitor human heart signals for tracking cardio-vascular conditions are challenging, particularly in remote healthcare settings. On the merits of low power consumption, portability, and non-intrusiveness, there are no suitable IoT solutions that can provide information comparable to the conventional Electrocardiography (ECG). In this paper, we propose an IoT device utilizing a spintronic-technology-based ultra-sensitive Magnetic Tunnel Junction (MTJ) sensor that measures the magnetic fields produced by cardio-vascular electromagnetic activity, i.e. Magentocardiography (MCG). We treat the low-frequency noise generated by the sensor, which is also a challenge for most other sensors dealing with low-frequency bio-magnetic signals. Instead of relying on generic signal processing techniques such as moving average, we employ deep-learning training on biomagnetic signals. Using an existing dataset of ECG records, MCG signals are synthesized. A unique deep learning model, composed of a one-dimensional convolution layer, Gated Recurrent Unit (GRU) layer, and a fully-connected neural layer, is trained using the labeled data moving through a striding window, which is able to smartly capture and eliminate the noise features. Simulation results are reported to evaluate the effectiveness of the proposed method that demonstrates encouraging performance.
Index Terms: Smart health | IoT | ECG | MCG | deep learning | noise | spintronic sensor | convolution | GRU | medical analytics
Derivation and verification of a similarity law for wind-tunnel free-flight tests of heavy-store separation
استخراج و تأیید قانون تشابه برای آزمایش های پرواز آزاد تونل بادی از تفکیک فروشگاه های سنگین-2020
Test-method research was carried out to consider the separation of a heavy store, and a similarity law for unsteady wind-tunnel free-flight tests of air-launch rockets was derived. The derivation of this similarity law considers the particular characteristics of a heavy store and aerodynamic interference with the carrier and focuses on solving the following problems: the separation of a heavy store causes a real carrier to have an acceleration and a velocity that cannot be ignored; the carrier in a wind-tunnel test is vertically fixed; and a wind-tunnel test cannot meet the Froude number (Fr) similarity condition. According to the special characteristics of heavy-store separation, a similarity law for wind-tunnel free-flight tests of heavy-store separation is derived. Computational fluid dynamics simulations are used to verify the new similarity law. The results show that the new similarity law can simulate the separation trajectory more realistically than existing methods, and the linear and angular displacement errors are decreased by an order of magnitude. The experimental accuracy of the new similarity law is even higher than that of a separation trajectory satisfying Fr matching. It is demonstrated that the new similarity law can be used to carry out unsteady experimental research on the separation of a heavy store such as an air-launch rocket, and this new law provides strong support for establishing the safety boundaries of heavy-store separation.
Keywords: Air-launch rocket | Similarity law derivation | Carrier and store interference | Multi-body separation | Wind-tunnel free-drop testing | Heavy-store airdrop
Risk assessment and management via multi-source information fusion for undersea tunnel construction
ارزیابی و مدیریت ریسک از طریق تلفیق اطلاعات چند منبع برای ساخت تونل زیر زمینی -2020
The construction of undersea tunnels is an extremely risky endeavor that is vulnerable to water seepage and gushing due to the high water pressure, complex geological conditions, and pore water trapped in unstable rocks. This risk can lead to the collapse of tunnels under construction and disastrous consequences of fatalities and injuries as well as project delays and financial losses. The current risk management practices for tunnel construction projects in China are static and rely on the subjective judgement of experts and practitioners and do not incorporate real-time monitoring data during the construction process at this time. This paper presents a new method and system to assess and manage the risks during the construction process by coupling the risk management system and the quality management system and integrating jobsite monitoring data, design data, and environmental data. In this new method and system, the risk factors are categorized into (hu)man, material, machine, method, and environment, or 4M1E, and are quantitatively measured. The Dempster-Shaffer (D-S) theory was adopted in this method to both fuse the 4M1E data and to compute the aggregate risk index. This new method and system was tested during the Xiamen Metro Line No. 3 project when a shield machine cutter accident occurred. The results show that, before the accident, the individual risk measures in all five dimensions (4M1E) and the aggregate risk index were extremely high, which clearly illustrated the feasibility and capability of the newly developed method and system.
Keywords: Undersea tunnel construction | Multi-source information fusion | Construction risk | D-S evidence theory | Fuzzy matter element
Distribution laws of freeze-thaw cycles and unsaturated concrete experiments in cold-region tunnels
قوانین توزیع چرخه انجماد و ذوب و آزمایش های بتن اشباع نشده در تونل های منطقه سرد-2020
Studying tunnel temperature fields could prevent frost damage. However, few studies have revealed the distribution laws of freeze-thaw cycles in cold-region tunnels. In this study, the distribution laws of freeze-thaw cycles in a tunnel were carried out, and the concrete deterioration induced by freeze-thaw cycling was explored. First, in situ monitoring equipment was used to collect the temperature in the longitudinal direction of the tunnels. A one-dimensional heat transfer model was used to analyze the temperature distribution in the circumferential direction of the tunnels, where a function was proposed to describe both the annual and diurnal temperature fluctuations. After that, the distribution laws of freeze-thaw cycles inside a tunnel were investigated. Specifically, in the tunnel longitudinal direction, the number of freeze-thaw cycles decreased from the entrance to the middle and then increased while approaching the exit, thereby exhibiting a V-shaped distribution. In the circumferential direction, the intrados lining nearly always exhibited freeze-thaw cycles. The number of freeze-thaw cycles and the temperature amplitude decreased rapidly with increasing depth. Furthermore, a series of unsaturated concrete experiments were performed to explore the concrete deterioration under different numbers of freeze-thaw cycles. The results showed that after being subjected to 100 freeze-thaw cycles, the concrete specimens displayed visible damage, whereas the concrete compressive strength was not affected. These findings can enrich the research of freeze-thaw cycles for cold-region tunnels and are significant for guiding tunnel maintenance.
Keywords: Cold-region tunnels | Temperature field | Freeze-thaw cycles | Unsaturated concrete | Numerical simulation
Integrating domain knowledge with deep learning models: An interpretable AI system for automatic work progress identification of NATM tunnels
ادغام دانش دامنه با مدل های یادگیری عمیق: یک سیستم هوش مصنوعی قابل تفسیر برای شناسایی پیشرفت کار خودکار تونل های NATM-2020
Finding a reliable and cost-effective approach to monitor the activities of the New Austrian Tunneling Method (NATM) tunnel construction automatically is a challenging yet important task. This study presents an interpretable artificial intelligence (AI) framework that automatically identifies NATM construction works using lowcost site surveillance images. The framework adopts the Bayesian statistics to combine the prior NATM construction knowledge with the visual evidence extracted by deep learning (DL) based computer vision models. The analysis results of Site CCTV surveillance videos of four NATM tunneling projects are presented to demonstrate its ability (i) to label NATM work cycles from the work timeline, (ii) to identify NATM work categories inside each work cycle, and (iii) to estimate the degree of plan-work deviation at the construction cycle level. The proposed framework yields promising results on a real NATM tunneling project.
Keywords: Artificial intelligence | NATM project monitoring | Deep learning | Computer vision
Data-driven safety enhancing strategies for risk networks in construction engineering
راهبردهای افزایش ایمنی داده محور برای شبکه های ریسک در مهندسی ساخت -2020
Risk management is crucial and indispensable to the success of projects, while identifying critical risks is the fundamental step in devising the corresponding safety measures. To fully exploit the value of richly accumulated accidental cases, this paper presents a data-driven research framework for proposing effective safety enhancing strategies based on risk networks in construction engineering, spanning the whole process from extracting accident chains from accidents to construct a risk network to devising safety measures. Aiming at the weighted heterogeneity of the risk network, both the performance metrics at network level and critical-risk identification metrics at node level are deliberately designed. These metrics then enable the proposing of a series of safetyenhancing strategies. In the case study, based on the accident-related data in China’s bridge-and-tunnel hybrid projects, different safety-enhancing strategies are compared through simulation experiments and analyzed to verify their effectiveness on optimizing costs and improving safety. Finally, based on results from simulations, relevant managerial suggestions are proposed.
Keywords: Safety enhancing strategies | Risk network | Data-driven | Construction engineering
Reinforcement learning based optimizer for improvement of predicting tunneling-induced ground responses
بهینه ساز مبتنی بر یادگیری تقویتی برای بهبود پیش بینی پاسخ های ناشی از tunneling-2020
Prediction of ground responses is important for improving performance of tunneling. This study proposes a novel reinforcement learning (RL) based optimizer with the integration of deep-Q network (DQN) and particle swarm optimization (PSO). Such optimizer is used to improve the extreme learning machine (ELM) based tunnelinginduced settlement prediction model. Herein, DQN-PSO optimizer is used to optimize the weights and biases of ELM. Based on the prescribed states, actions, rewards, rules and objective functions, DQN-PSO optimizer evaluates the rewards of actions at each step, thereby guides particles which action should be conducted and when should take this action. Such hybrid model is applied in a practical tunnel project. Regarding the search of global best weights and biases of ELM, the results indicate the DQN-PSO optimizer obviously outperforms conventional metaheuristic optimization algorithms with higher accuracy and lower computational cost. Meanwhile, this model can identify relationships among influential factors and ground responses through selfpracticing. The ultimate model can be expressed with an explicit formulation and used to predict tunnelinginduced ground response in real time, facilitating its application in engineering practice.
Keywords: Tunnel | Ground response | Reinforcement learning | Extreme learning machine | Optimization
Dynamic luminance tuning method for tunnel lighting based on data mining of real-time traffic flow
روش تنظیم پویا درخشندگی برای روشنایی تونل بر اساس داده کاوی جریان ترافیک در زمان واقعی-2020
Tunnel lighting constitutes one of the major expenses incurred in transportation lighting, and hence substantial research has been conducted to improve the efficiency of lighting and thus to minimize operating costs. This paper investigates an intelligent method for adjusting tunnel lighting with dynamic control based on data mining of traffic flow distribution, traffic composition, and vehicle speed distribution. Field monitoring data of traffic flow in five real expressway tunnels, which are in HeDa expressway, Jilin Province, China, was used in the analysis. The K-MEANS clustering algorithm was used to group (or cluster) the distribution of daily traffic volume into six-time periods, in which the traffic volume includes two peak periods (8:01–11:23 and 14:31–19:01). A dynamic luminance regulation method is proposed that distinguishes operational strategies under different time periods. Furthermore, the impact of tunnel length and traffic flow on the effect of energysaving and system sustainability of the proposed method was assessed. The results show that when using the proposed method, the energy-savings in tunnel lighting could be between about 50% and 60% for a daily traffic volume between 750 and 2500 vehicles. The results also show that the switching frequency of the lighting system is significantly reduced, which would significantly enhance the sustainability of the lighting system.
Keywords: Data mining | Energy management | Intelligent control | Tunnel lighting