Search results for: supplementary reinforcement

Authors: Peng Liu

Abstract:

Deep Q networks (DQNs) represent a significant advancement in reinforcement learning, utilizing neural networks to approximate the optimal Q-value for guiding sequential decision processes. This paper presents a comprehensive introduction to reinforcement learning principles, delves into the mechanics of DQNs, and explores its application in portfolio optimization. By evaluating the performance of DQNs against traditional benchmark portfolios, we demonstrate its potential to enhance investment strategies. Our results underscore the advantages of DQNs in dynamically adjusting asset allocations, offering a robust portfolio management framework.

Keywords: deep reinforcement learning, deep Q networks, portfolio optimization, multi-period optimization

Procedia PDF Downloads 35

Authors: Mohammed Albahttiti, Eliana Aguilar

Abstract:

A push for sustainability in the concrete industry is increasing. Cow manure itself is becoming a problem and having the potential solution to use it in concrete as a cementitious replacement would be an ideal solution. For cow manure ash to become a well-rounded substitute, it would have to meet the right criteria to progress in becoming a more popular idea in the concrete industry. This investigation primarily focuses on how the replacement of cow manure ash affects the air content and air void distribution in concrete. In order to assess these parameters, the Super Air Meter (SAM) was used to test concrete in this research. In addition, multiple additional tests were performed, which included the slump test, temperature, and compression test. The strength results of the manure ash in concrete were promising. The manure showed compression strength results that are similar to that of the other supplementary cementitious materials tested. On the other hand, concrete samples made with cow manure ash showed 2% air content loss and an increasing SAM number proportional to cow manure content starting at 0.38 and increasing to 0.8. In conclusion, while the use of cow manure results in loss of air content, it results in compressive strengths similar to other supplementary cementitious materials.

Keywords: air content, biomass ash, cow manure ash, super air meter, supplementary cementitious materials

Procedia PDF Downloads 149

Authors: Zeineb Ben Salem, Wissem Frikha, Mounir Bouassida

Abstract:

The aim of this paper is to assess the effectiveness of stone columns as a liquefaction countermeasure focusing on shear reinforcementbenefit. In fact, stone columns which have high shear modulus relative to the surrounding soils potentially can carry higher shear stress levels. Thus, stone columns provide shear reinforcement and decrease the Cyclic Shear Stress Ratio CSR to which the treated soils would be subjected during an earthquake. In order to quantify the level of shear stress reduction in reinforced soil, several approaches have been developed. Nevertheless, the available approaches do not take into account the improvement of the soil parameters, mainly the shear modulusdue to stone columns installation. Indeed, in situ control tests carried out before and after the installation of stone columns based upon the results of collected data derived from 24 case histories have given evidence of the improvement of the existing soil properties.In this paper, the assessment of shear reinforcement of stone columns that accounts such improvement of the soil parameters due to stone column installation is investigated. Comparative results indicate that considering the improvement effects considerably affect the assessment of shear reinforcement for liquefaction analysis of reinforced soil by stone columns.

Keywords: stone column, liquefaction, shear reinforcement, CSR, soil improvement

Procedia PDF Downloads 153

Authors: Dhyaalddin Bahaalddin Noori Zangana

Abstract:

This study involves the behavior of reinforced sand under a square footing. A series of bearing capacity tests were performed on a small-scale laboratory model, which filled with a poorly-graded homogenous bed of sand, which was placed in a medium dense state using sand raining technique. The sand was reinforced with 40 mm wide household aluminum foil strips. The main studied parameters was to consider the effect of reinforcing strip length, with various linear density of reinforcement, number of reinforcement layers and depth of top layer of reinforcement below the footing, on load-settlement behavior, bearing capacity ratio and settlement reduction factor. The relation of load-settlement generally showed similar trend in all the tests. Failure was defined as settlement equal to 10% of the footing width. The recommended optimum reinforcing strip length, linear density of reinforcement, number of reinforcement layers and depth of top layer of reinforcing strips that give the maximum bearing capacity improvement and minimum settlement reduction factor were presented and discussed. Different bearing capacity ration versus length of the reinforcing strips and settlement reduction factor versus length of the reinforcing strips relations at failure were showed improvement of bearing capacity ratio by a factor of 3.82 and reduction of settlement reduction factor by a factor of 0.813. The optimum length of reinforcement was found to be 7.5 times the footing width.

Keywords: square footing, relative density, linear density of reinforcement, bearing capacity ratio, load-settlement behaviour

Procedia PDF Downloads 98

Authors: Adam Beaunoyer, Cory Beaunoyer, Mohammed Elmorsy, Hanan Saleh

Abstract:

This research aims to create a reinforcement learning agent capable of racing in challenging simulated environments with a low collision count. We present a reinforcement learning agent that can navigate challenging tracks using both a Deep Q-Network (DQN) and a Soft Actor-Critic (SAC) method. A challenging track includes curves, jumps, and varying road widths throughout. Using open-source code on Github, the environment used in this research is based on the 1995 racing game WipeOut. The proposed reinforcement learning agent can navigate challenging tracks rapidly while maintaining low racing completion time and collision count. The results show that the SAC model outperforms the DQN model by a large margin. We also propose an alternative multiple-car model that can navigate the track without colliding with other vehicles on the track. The SAC model is the basis for the multiple-car model, where it can complete the laps quicker than the single-car model but has a higher collision rate with the track wall.

Keywords: reinforcement learning, soft actor-critic, deep q-network, self-driving cars, artificial intelligence, gaming

Procedia PDF Downloads 49

Authors: Hatem Ghith

Abstract:

This research paper experimentally investigates the effect of burning by fire flame from one face on the behavior and load carrying capacity for reinforced columns. Residual ultimate load carrying capacity, axial deformation, crack pattern and maximum crack width for column specimens with and without burning were recorded and discussed. Tested six reinforced concrete columns were divided into control specimen and two groups. The first group was exposed to a fire with a different temperature (300, 500, 700 °C) for an hour with reinforcement ratio 0.89% and the second group was exposed to a fire with a temperature 500 °C for an hour with different reinforcement ratio (0.89%, 2.18%, and 3.57%), then all columns were tested under short-term axial loading. From the obtained results, it could be concluded that the fire parameters significantly influence the fire resistance of R.C columns. The fire parameters cause axial deformation and moment on the column due to the eccentricity that generated from the difference in temperature and consequently the compressive stresses of both faces of the columns but the increased reinforcement ratio enhanced the resistance of columns for axial deformation and moment on the column due to the eccentricity.

Keywords: columns, reinforcement ratio, strength, time exposure

Procedia PDF Downloads 246

Authors: Ravi Prasad Darapureddi

Abstract:

The use of industrial waste in making concrete reduce the consumption of natural resources and pollution of the environment. These materials possess problems of disposal and health hazards. An attempt has been made to use paper and thermal industrial wastes such as lime sludge and flyash. Present investigation is aimed at the utilization of Lime Sludge and Flyash as Supplementary Cementitious Materials (SCM) and influence of these materials on strength properties of concrete. Thermal industry waste fly ash is mixed with lime sludge and used as a replacement to cement at different proportions to obtain the strength properties and compared with ordinary concrete prepared without any additives. Grade of concrete prepared was M₂₅ designed according to Indian standard method. Cement has been replaced by paper industry waste and fly ash in different proportions such as 0% (normal concrete), 10%, 20%, and 30% by weight. Mechanical properties such as compressive strength, splitting tensile strength and flexural strength were assessed. Test results indicated that the use of lime sludge and Fly ash in concrete had improved the properties of concrete. Better results were observed at 20% replacement of cement with these additives.

Keywords: supplementary cementitious materials, lime sludge, fly ash, strength properties

Procedia PDF Downloads 198

Authors: Waqed H. Almohammed, Mohammed Y. Fattah, Sajjad E. Rasheed

Abstract:

Dynamic traffic loads cause deformation of underground pipes, resulting in vehicle discomfort. This makes it necessary to reinforce the layers of soil above underground pipes. In this study, the subbase layer was reinforced. Finite element software (PLAXIS 3D) was used to in the simulation, which includes geocell reinforcement, vehicle loading, soil layers and Glass Fiber Reinforced Plastic (GRP) pipe. Geocell reinforcement was modeled using a geogrid element, which was defined as a slender structure element that has the ability to withstand axial stresses but not to resist bending. Geogrids cannot withstand compression but they can withstand tensile forces. Comparisons have been made between the numerical models and experimental works, and a good agreement was obtained. Using the mathematical model, the performance of three different pipes of diameter 600 mm, 800 mm, and 1000 mm, and three different vehicular speeds of 20 km/h, 40 km/h, and 60 km/h, was examined to determine their impact on surface settlement and vertical pressure at the pipe crown for two cases: with and without geocell reinforcement. The results showed that, for a pipe diameter of 600 mm under geocell reinforcement, surface settlement decreases by 94 % when the speed of the vehicle is 20 km/h and by 98% when the speed of the vehicle is 60 km/h. Vertical pressure decreases by 81 % when the diameter of the pipe is 600 mm, while the value decreases to 58 % for a pipe with diameter 1000 mm. The results show that geocell reinforcement causes a significant and positive reduction in surface settlement and vertical stress above the pipe crown, leading to an increase in pipe safety.

Keywords: dynamic loading, finite element, geocell-reinforcement, GRP pipe, PLAXIS 3D, surface settlement

Procedia PDF Downloads 248

Authors: Qinggui Wu, Xinming Cao, Guyue Guo, Jiajun Ding

Abstract:

In efforts to study the shear ductility of regional confined concrete beam, 5 reinforced concrete beams were tested to examine its shear performance. These beams has the same shear span ratio, concrete strength, different ratios of tension reinforcement and shapes of stirrup. The purpose of the test is studying the effects of stirrup shape and tension reinforcement ratio on failure mode and shear ductility. The test shows that the regional confined part can be used as an independent part and the rest of the beam is good to work together so that the ductility of the beam is more one time higher than that of the normal confined concrete beam. The related laws of the effect of tension reinforcement ratio and stirrup shapes on beam’s shear ductility are founded.

Keywords: ratio of tension reinforcement, stirrup shapes, shear ductility, failure mode

Procedia PDF Downloads 335

Authors: Khalil Bachiri, Ali Yahyaouy, Nicoleta Rogovschi

Abstract:

Improved user experience is a goal of contemporary recommender systems. Recommender systems are starting to incorporate reinforcement learning since it easily satisfies this goal of increasing a user’s reward every session. In this paper, we examine the most effective Reinforcement Learning agent tactics on the Movielens (1M) dataset, balancing precision and a variety of recommendations. The absence of variability in final predictions makes simplistic techniques, although able to optimize ranking quality criteria, worthless for consumers of the recommendation system. Utilizing the stochasticity of Leonard-Ornstein processes, our suggested strategy encourages the agent to investigate its surroundings. Research demonstrates that raising the NDCG (Discounted Cumulative Gain) and HR (HitRate) criterion without lowering the Ornstein-Uhlenbeck process drift coefficient enhances the diversity of suggestions.

Keywords: recommender systems, reinforcement learning, deep learning, DDPG, Leonard-Ornstein process

Procedia PDF Downloads 145

Authors: Jee-Sang Kim, Jong Ho Park

Abstract:

Geopolymer concretes are a new class of construction materials that have emerged as an alternative to Ordinary Portland cement concrete. Considerable researches have been carried out on material development of geopolymer concrete, however, a few studies have been reported on the structural use of them. This paper presents the bond behaviors of reinforcement embedded in fly ash based geopolymer concrete. The development lengths of reinforcement for various compressive strengths of concrete, 20, 30 and 40 MPa, and reinforcement diameters, 10, 16, and 25 mm are investigated. Total 27 specimens were manufactured and pull-out test according to EN 10080 was applied to measure bond strength and slips between concrete and reinforcements. The average bond strengths decreased from 23.06MPa to 17.26 MPa, as the diameters of reinforcements increased from 10mm to 25mm. The compressive strength levels of geopolymer concrete showed no significant influence on bond strengths in this study. Also, the bond-slip relations between geopolymer concrete and reinforcement are derived using non-linear regression analysis for various experimental conditions.

Keywords: bond-slip relation, bond strength, geopolymer concrete, pull-out test

Procedia PDF Downloads 350

Authors: Suwaree Yordchim

Abstract:

The objective is to study the satisfaction on English with an online learning. Online learning system mainly consists of English lessons, exercises, tests, web boards, and supplementary lessons for language practice. The sample groups are 80 Thai students studying English for Business Communication, majoring in Hotel and Lodging Management. The data are analyzed by mean, standard deviation (S.D.) value from the questionnaires. The results were found that the most average of satisfaction on academic aspects are technological searching tool through E-learning system that support the students’ learning (4.51), knowledge evaluation on prepost learning and teaching (4.45), and change for project selections according to their interest, subject contents including practice in the real situations (4.45), respectively.

Keywords: English language learning, online system, online learning, supplementary lessons

Procedia PDF Downloads 466

Authors: C. Shim, C. Koem, S. Park, S. Lee

Abstract:

This paper deals with experimental studies on pre stressed precast concrete columns with continuous reinforcing bars and pre stressing tendons. Design requirements on minimum transverse reinforcement ratio are not included in current design codes. Pre stressing introduces additional compression to the column. Precast columns with different transverse reinforcement ratios were tested to derive adequate design requirement. Displacement ductility of the pre stressed precast columns was evaluated and compared with previous studies. Design of axial steels including reinforcing bars and pre stressing tendons influenced on the seismic performance. Without significant increase of transverse reinforcement ratio, the specimens showed required displacement ductility without reduction of their flexural strength. Design recommendations for precast bridge piers were derived.

Keywords: displacement ductility, flexural strength, prestressed precast column, transverse reinforcement

Procedia PDF Downloads 279

Authors: Jawairia Rasheed

Abstract:

Modern robots are often used for search and rescue purpose. One of the key areas of interest in such cases is learning complex environments. One of the key methodologies for robots in such cases is reinforcement learning. In reinforcement learning robots learn to move the path to reach the goal while avoiding obstacles. Q-learning, one of the most advancement of reinforcement learning is used for making the robots to learn the path. Robots learn by interacting with the environment to reach the goal. In this paper simulation model of bee-like robots is implemented in NETLOGO. In the start the learning rate was less and it increased with the passage of time. The bees successfully learned to reach the goal while avoiding obstacles through Q-learning technique.

Keywords: reinforlearning of bee like robots for reaching the goalcement learning for randomly placed obstacles, obstacle avoidance through q-learning, q-learning for obstacle avoidance,

Procedia PDF Downloads 106

Authors: Edward W. Staley, Corban G. Rivera, Ashley J. Llorens

Abstract:

Advances in reinforcement learning (RL) have resulted in recent breakthroughs in the application of artificial intelligence (AI) across many different domains. An emerging landscape of development environments is making powerful RL techniques more accessible for a growing community of researchers. However, most existing frameworks do not directly address the problem of learning in complex operating environments, such as dense urban settings or defense-related scenarios, that incorporate distributed, heterogeneous teams of agents. To help enable AI research for this important class of applications, we introduce the AI Arena: a scalable framework with flexible abstractions for distributed multi-agent reinforcement learning. The AI Arena extends the OpenAI Gym interface to allow greater flexibility in learning control policies across multiple agents with heterogeneous learning strategies and localized views of the environment. To illustrate the utility of our framework, we present experimental results that demonstrate performance gains due to a distributed multi-agent learning approach over commonly-used RL techniques in several different learning environments.

Keywords: reinforcement learning, multi-agent, deep learning, artificial intelligence

Procedia PDF Downloads 160

Authors: Delaram Ghodsi, Nasrin Omidvar, Hassan Eini-Zinab, Arash Rashidian, Hossein Raghfar

Abstract:

Food supplementary programs are policy approaches that aim to reduce financial barriers to healthy diets and tackle food insecurity. This study aimed to evaluate the effect of the supportive section of Multidisciplinary Supplementary Program for Improvement of Nutritional Status of Children (MuPINSC) on households’ food security status and nutritional status of mothers. MuPINSC is a national integrative program in Iran that distributes supplementary food basket to malnourished or growth retarded children living in low-income families in addition to providing health services, including sanitation, growth monitoring, and empowerment of families. This longitudinal study is part of a comprehensive evaluation of the program. The study participants included 359 mothers of children aged 6 to 72 month under coverage of the supportive section of the program in two provinces of Iran (Semnan and Qazvin). Demographic and economic characteristics of families were assessed by a questionnaire. Data on food security of family was collected by locally adapted Household Food Insecurity Access Scale (HFIAS) at the baseline of the study and six month thereafter. Weight and height of mothers were measured at the baseline and end of the study and mother’s BMI was calculated. Data were analysed, using paired t-test, GEE (Generalized Estimating Equation), and Chi-square tests. Based on the findings, at the baseline, only 4.7% of families were food-secure, while 13.1%, 38.7% and, 43.5% were categorized as mild, moderate and severe food insecure. After six months follow up, the distribution of different levels of food security changed significantly (P<0.001) to 7.9%, 11.6%, 42.6%, and 38%, respectively. At the end of the study, the chance of food insecurity was significantly 20% lower than the beginning (OR=0.796; 0.653-0.971). No significant difference was observed in maternal BMI based on food security (P>0.05). The findings show that the food supplementary program for children improved household food security status in the studied households. Further research is needed to assess other factors that affect the effectiveness of this large scale program on nutritional status and household’s food security.

Keywords: food security, food supplementary program, household, malnourished children

Procedia PDF Downloads 402

Authors: Yitao Lei, Xingxiang Zhai, Burra Venkata Durga Kumar

Abstract:

As the quantity and complexity of computing in large-scale software systems increase, distributed system computing becomes increasingly important. The distributed system realizes high-performance computing by collaboration between different computing resources. If there are no efficient resource scheduling resources, the abuse of distributed computing may cause resource waste and high costs. However, resource scheduling is usually an NP-hard problem, so we cannot find a general solution. However, some optimization algorithms exist like genetic algorithm, ant colony optimization, etc. The large scale of distributed systems makes this traditional optimization algorithm challenging to work with. Heuristic and machine learning algorithms are usually applied in this situation to ease the computing load. As a result, we do a review of traditional resource scheduling optimization algorithms and try to introduce a deep reinforcement learning method that utilizes the perceptual ability of neural networks and the decision-making ability of reinforcement learning. Using the machine learning method, we try to find important factors that influence the performance of distributed system computing and help the distributed system do an efficient computing resource scheduling. This paper surveys the application of deep reinforcement learning on distributed system computing resource scheduling proposes a deep reinforcement learning method that uses a recurrent neural network to optimize the resource scheduling, and proposes the challenges and improvement directions for DRL-based resource scheduling algorithms.

Keywords: resource scheduling, deep reinforcement learning, distributed system, artificial intelligence

Procedia PDF Downloads 113

Authors: Hyeongbok Kim, Lingling Zhao, Xiaohong Su

Abstract:

Deep reinforcement learning has been applied to address various problems in robotics, such as autonomous driving and unmanned aerial vehicle. However, because of the sparse reward penalty for a collision with obstacles during the navigation mission, the agent fails to learn the optimal policy or requires a long time for convergence. Therefore, using obstacles and enemy agents, in this paper, we present a curriculum-based boost learning method to effectively train compound skills during multi-agent reinforcement learning. First, to enable the agents to solve challenging tasks, we gradually increased learning difficulties by adjusting reward shaping instead of constructing different learning environments. Then, in a benchmark environment with static obstacles and moving enemy agents, the experimental results showed that the proposed curriculum learning strategy enhanced cooperative navigation and compound collision avoidance skills in uncertain environments while improving learning efficiency.

Keywords: curriculum learning, hard exploration, multi-agent reinforcement learning, robotic navigation, sparse reward

Procedia PDF Downloads 93

Authors: Haodong Huang, Zida Zhao, Shilong Sun, Chiyao Li, Wenfu Xu

Abstract:

The control methods for legged robots based on deep reinforcement learning have seen widespread application; however, the inherent black-box nature of neural networks presents challenges in understanding the decision-making motives of the robots. To address this issue, we propose a fully interpretable deep reinforcement learning training method to elucidate the underlying principles of legged robot motion. We incorporate the dynamics of legged robots into the policy, where observations serve as inputs and actions as outputs of the dynamics model. By embedding the dynamics equations within the multi-layer perceptron (MLP) computation process and making the parameters trainable, we enhance interpretability. Additionally, Bayesian optimization is introduced to train these parameters. We validate the proposed fully interpretable motion control algorithm on a legged robot, opening new research avenues for motion control and learning algorithms for legged robots within the deep learning framework.

Keywords: deep reinforcement learning, interpretation, motion control, legged robots

Procedia PDF Downloads 23

Authors: M. Khlifa, A. Youssef, A. F. Zaed, A. Kraft, V. Arrighi

Abstract:

A series of PS-nanoparticles were prepared by grafting PS from both aggregated silica and colloidally silica using atom-transfer radical polymerisation (ATRP). The mechanical behaviour of the nanocomposites have been examined by differential scanning calorimetry (DSC)and dynamic mechanical thermal analysis (DMTA).

Keywords: ATRP, nanocomposites, polystyrene, reinforcement

Procedia PDF Downloads 626

Authors: Luís Pina

Abstract:

The conservation of marine biodiversity keeps ecosystems in balance and ensures the sustainable use of resources. In this context, technological resources have been used for monitoring marine species to allow biologists to obtain data in real-time. There are different mobile applications developed for data collection for monitoring purposes, but these systems are designed to be utilized only on third-generation (3G) phones or smartphones with Internet access and in rural parts of the developing countries, Internet services and smartphones are scarce. Thus, the objective of this work is to develop a system to monitor marine turtles using Unstructured Supplementary Service Data (USSD), which users can access through basic mobile phones. The system aims to improve the data collection mechanism and enhance the effectiveness of current systems in monitoring sea turtles using any type of mobile device without Internet access. The system will be able to report information related to the biological activities of marine turtles. Also, it will be used as a platform to assist marine conservation entities to receive reports of illegal sales of sea turtles. The system can also be utilized as an educational tool for communities, providing knowledge and allowing the inclusion of communities in the process of monitoring marine turtles. Therefore, this work may contribute with information to decision-making and implementation of contingency plans for marine conservation programs.

Keywords: GSM, marine biology, marine turtles, unstructured supplementary service data (USSD)

Procedia PDF Downloads 206

Authors: Hadi Rouhi Belvirdi

Abstract:

Reinforced concrete structures, in addition to bearing service loads and seismic effects, may also be subjected to impact loads resulting from unforeseen incidents. Understanding the behavior of these structures is crucial, as they serve to protect against such sudden loads and can significantly reduce damage and destruction. In examining the behavior of structures under such loading conditions, a total of eight specimens of single-layer reinforced concrete slabs were subjected to impact loading through the free fall of weights from specified heights. The weights and dimensions of the specimens were uniform, and the amount of reinforcement was consistent. By altering the slabs' overall shape and the reinforcement details, efforts were made to optimize the behavior of the slabs against impact loads. The results indicated that utilizing ductile features in the slabs increased their resistance to impact loading. However, the compressive strength of the reinforcement did not significantly enhance the flexural resistance. Assuming a constant amount of longitudinal steel, changes in the placement of tensile reinforcement led to a decrease in resistance. With a fixed amount of transverse steel, merely adjusting the angle of the transverse reinforcement could help control cracking and mitigate premature failures. An increase in compressive resistance beyond a certain limit resulted in local buckling of the compressive zone, subsequently decreasing the impact resistance.

Keywords: reinforced concrete slab, high-strength concrete, impact loading, impact resistance

Procedia PDF Downloads 15

Authors: B. Hekner, J. Myalski, P. Wrzesniowski

Abstract:

This study presents aluminum matrix composites reinforced by glassy carbon (GC) and titanium (Ti). In the first step, the heterophase (GC+Ti), spatial form (similar to skeleton) of reinforcement was obtained via own method. The polyurethane foam (with spatial, open-cells structure) covered by suspension of Ti particles in phenolic resin was pyrolyzed. In the second step, the prepared heterogeneous foams were infiltrated by aluminium alloy. The manufactured composites are designated to industrial application, especially as a material used in tribological field. From this point of view, the glassy carbon was applied to stabilise a coefficient of friction on the required value 0.6 and reduce wear. Furthermore, the wear can be limited due to titanium phase application, which reveals high mechanical properties. Moreover, fabrication of thin titanium layer on the carbon skeleton leads to reduce contact between aluminium alloy and carbon and thus aluminium carbide phase creation. However, the main modification involves the manufacturing of reinforcement in the form of 3D, skeleton foam. This kind on reinforcement reveals a few important advantages compared to classical form of reinforcement-particles: possibility to control homogeneity of reinforcement phase in composite material; low-advanced technique of composite manufacturing- infiltration; possibility to application the reinforcement only in required places of material; strict control of phase composition; High quality of bonding between components of material. This research is founded by NCN in the UMO-2016/23/N/ST8/00994.

Keywords: metal matrix composites, MMC, glassy carbon, heterophase composites, tribological application

Procedia PDF Downloads 118

Authors: Mohammad Mehdi Bagheri, Tianying Ma, Meng Gong

Abstract:

Nail laminated timber (NLT) is widely used for constructing timber bridge decks in North America. Butt joints usually exist due to the length limits of lumber, leading to concerns about the decrease of structural performance of NLT. This study aimed at investigating the provisions incorporated in Canadian highway bridge design code on the use of but joints in wooden bridge decks. Three and five layers NLT specimens with various configurations were tested under 3-point bending test. It was found that the standard equation is capable of predicting the bending stiffness reduction due to butt joints and 1-m band limit in which, one but joint in every three adjacent lamination is allowed, sounds reasonable. The strength reduction also followed a pattern similar to stiffness reduction. Also reinforcement of the butt joint through nails and steel side plates was attempted. It was found that nail reinforcement recovers the stiffness slightly. In contrast, reinforcing the butt joint through steel side plate improved the flexural performance significantly when compared to the nail reinforcement.

Keywords: nail laminated timber, butt joint, bending stiffness, reinforcement

Procedia PDF Downloads 186

Authors: Adrian Millea

Abstract:

In hierarchical reinforcement learning, one of the main issues encountered is the discovery of subgoal states or options (which are policies reaching subgoal states) by partitioning the environment in a meaningful way. This partitioning usually requires an expensive global clustering operation or eigendecomposition of the Laplacian of the states graph. We propose a local solution to this issue, much more efficient than algorithms using global information, which successfully discovers subgoal states by computing a simple function, which we call heterogeneity for each state as a function of its neighbors. Moreover, we construct a value function using the difference in heterogeneity from one step to the next, as reward, such that we are able to explore the state space much more efficiently than say epsilon-greedy. The same principle can then be applied to higher level of the hierarchy, where now states are subgoals discovered at the level below.

Keywords: exploration, hierarchical reinforcement learning, locality, options, value functions

Procedia PDF Downloads 171

Authors: Mazin Mohammed S. Sarhan

Abstract:

This study presents the performance of concrete beams reinforced with steel plates as a technique of reinforcement. Three reinforced concrete beams with the dimensions of 200 mm x 300 mm x 4000 mm (width x height x length, respectively) were experimentally investigated under flexural loading. The deformed steel bars were used as the main reinforcement for the first beam. A steel plate placed horizontally was used as the main reinforcement for the second beam. The bond between the steel plate and the surrounding concrete was enhanced by using steel bolts (with a diameter of 20 mm and length of 100 mm) welded to the steel plate at a regular distance of 200 mm. A pair of steel plates placed vertically was used as the main reinforcement for the third beam. The bond between the pair steel plates and the surrounding concrete was enhanced by using 4 equal steel angles (with the dimensions of 75 mm x 75 mm and the thickness of 8 mm) for each vertical steel plate. Two steel angles were welded at each end of the steel plate. The outcomes revealed that the bending stiffness of the beams reinforced with steel plates was higher than that reinforced with deformed steel bars. Also, the flexural ductile behavior of the second beam was much higher than the rest beams.

Keywords: concrete beam, deflection, ductility, plate

Procedia PDF Downloads 160

Authors: Mutadi

Abstract:

Pipes laying on an unstable slope of soft soil are prone to movement. Pipelines that are buried in unstable slope areas will move due to lateral loads from soil movement, which can cause damage to the pipeline. A small-scale laboratory model of the reinforcement system of piles supported by retaining walls was conducted to investigate the effect of lateral load on the reinforcement. In this experiment, the lateral forces of 0.3 kN, 0.35 kN, and 0.4 kN and vertical force of 0.05 kN, 0.1 kN, and 0.15 kN were used. Lateral load from the electric jack is equipped with load cell and vertical load using the cement-steel box. To validate the experimental result, a finite element program named 2-D Plaxis was used. The experimental results showed that with an increase in lateral loading, the displacement of the reinforcement system increased. For a Vertical Load, 0.1 kN and versus a lateral load of 0.3 kN causes a horizontal displacement of 0.35 mm and an increase of 2.94% for loading of 0.35 kN and an increase of 8.82% for loading 0.4 kN. The pattern is the same in the finite element method analysis, where there was a 6.52% increase for 0.35 kN loading and an increase to 23.91 % for 0.4 kN loading. In the same Load, the Reinforcement System is reliable, as shown in Safety Factor on dry conditions were 3.3, 2.824 and 2.474, and on wet conditions were 2.98, 2.522 and 2.235.

Keywords: soft soil, deflection, wall, pipeline

Procedia PDF Downloads 163

Authors: S. Kaymakçı, D. Gündoğdu, H. Özçelik

Abstract:

The truck park of a warehouse for a chain of supermarket was going to be constructed on a poor ground. Rather than using a piled foundation, the client was convinced that a ground improvement using a reinforced foundation raft also known as “basal reinforcement” shall work. The retaining structures supporting the truck park area were designed using a hybrid structure made up of the Terramesh® Wall System and MacGrid™ high strength geogrids. The total wall surface area is nearly 2740 sq.m , reaching a maximum height of 13.00 meters. The area is located in the first degree seismic zone of Turkey and the design seismic acceleration is high. The design of walls has been carried out using pseudo-static method (limit equilibrium) taking into consideration different loading conditions using Eurocode 7. For each standard approach stability analysis in seismic condition were performed. The paper presents the detailed design of the reinforced soil structure, basal reinforcement and the construction methods; advantages of using such system for the project are discussed.

Keywords: basal reinforcement, geogrid, reinforced soil raft, reinforced soil wall, soil reinforcement

Procedia PDF Downloads 304

Authors: Madhu Jain, Rakesh Kumar Meena

Abstract:

This paper investigates a finite M/G/1 fault tolerant multi-component machining system. The system incorporates the features such as standby support, threshold recovery and imperfect coverage make the study closer to real time systems. The performance prediction of M/G/1 fault tolerant system is carried out using recursive approach by treating remaining service time as a supplementary variable. The numerical results are presented to illustrate the computational tractability of analytical results by taking three different service time distributions viz. exponential, 3-stage Erlang and deterministic. Moreover, the cost function is constructed to determine the optimal choice of system descriptors to upgrading the system.

Keywords: fault tolerant, machine repair, threshold recovery policy, imperfect coverage, supplementary variable technique

Procedia PDF Downloads 292

Authors: Egor E. Nuzhin, Nikolay V. Brilliantov

Abstract:

We propose an approach for addressing Reinforcement Learning (RL) problems. It combines the ideas of umbrella sampling, borrowed from Monte Carlo technique of computational physics and chemistry, with optimal control methods, and is realized on the base of neural networks. This results in a powerful algorithm, designed to solve hard RL problems – the problems, with long-time delayed reward, state-traps sticking and a lack of terminal states. It outperforms the prominent algorithms, such as PPO, RND, iLQR and VI, which are among the most efficient for the hard problems. The new algorithm deals with a continuous ensemble of agents and expected return, that includes the ensemble entropy. This results in a quick and efficient search of the optimal policy in terms of ”exploration-exploitation trade-off” in the state-action space.

Keywords: umbrella sampling, reinforcement learning, policy gradient, dynamic programming

Procedia PDF Downloads 25