Search results for: reduce the quantity of reinforcement

Authors: Sana Rezig, Yosr Ben Mlik, Mounir Jaouadi, Foued Khoffi, Slah Msahli, Bernard Durand

Abstract:

Increasing interest in environmental concerns, natural fibers are once again being considered as reinforcements for polymer composites. The main objective of this study is to explore another natural resource, Typha fiber; which is renewable without production cost and available abundantly in nature. The aim of this study was to study the flexural properties of composite resin with and without reinforcing Typha leaf and stem fibers. The specimens were made by the hand-lay-up process using polyester matrix. In our work, we focused on the effect of various treatment conditions (sea water, alkali treatment and a combination of the two treatments), as a surface modifier, on the flexural properties of the Typha fibers reinforced polyester composites. Moreover, weight ratio of Typha leaf or stem fibers was investigated. Besides, both fibers from leaf and stem of Typha plant were used to evaluate the reinforcing effect. Another parameter, which is reinforcement structure, was investigated. In fact, a first composite was made with air-laid nonwoven structure of fibers. A second composite was with a mixture of fibers and resin for each kind of treatment. Results show that alkali treatment and combined process provided better mechanical properties of composites in comparison with fiber treated by sea water. The fiber weight ratio influenced the flexural properties of composites. Indeed, a maximum value of flexural strength of 69.8 and 62,32 MPa with flexural modulus of 6.16 and 6.34 GPawas observed respectively for composite reinforced with leaf and stem fibers for 12.6 % fiber weight ratio. For the different treatments carried out, the treatment using caustic soda, whether alone or after retting seawater, show the best results because it improves adhesion between the polyester matrix and the fibers of reinforcement. SEM photographs were made to ascertain the effects of the surface treatment of the fibers. By varying the structure of the fibers of Typha, the reinforcement used in bulk shows more effective results as that used in the non-woven structure. In addition, flexural strength rises with about (65.32 %) in the case of composite reinforced with a mixture of 12.6% leaf fibers and (27.45 %) in the case of a composite reinforced with a nonwoven structure of 12.6 % of leaf fibers. Thus, to better evaluate the effect of the fiber origin, the reinforcing structure, the processing performed and the reinforcement factor on the performance of composite materials, a statistical study was performed using Minitab. Thus, ANOVA was used, and the patterns of the main effects of these parameters and interaction between them were established. Statistical analysis, the fiber treatment and reinforcement structure seem to be the most significant parameters.

Keywords: flexural properties, fiber treatment, structure and weight ratio, SEM photographs, Typha leaf and stem fibers

Procedia PDF Downloads 391

Authors: M. Amrita, R. R. Srikant, A. V. Sita Rama Raju

Abstract:

Water miscible cutting fluids are conventionally used to lubricate and cool the machining zone. But issues related to health hazards, maintenance and disposal costs have limited their usage, leading to application of Minimum Quantity Lubrication (MQL). To increase the effectiveness of MQL, nanocutting fluids are proposed. In the present work, water miscible nanographite cutting fluids of varying concentration are applied at cutting zone by two systems A and B. System A utilizes high pressure air and supplies cutting fluid at a flow rate of 1ml/min. System B uses low pressure air and supplies cutting fluid at a flow rate of 5ml/min. Their performance in machining is evaluated by measuring cutting temperatures, tool wear, cutting forces and surface roughness and compared with dry machining and flood machining. Application of nano cutting fluid using both systems showed better performance than dry machining. Cutting temperatures and cutting forces obtained by both techniques are more than flood machining. But tool wear and surface roughness showed improvement compared to flood machining. Economic analysis has been carried out in all the cases to decide the applicability of the techniques.

Keywords: economic analysis, machining, minimum quantity lubrication, nanofluid

Procedia PDF Downloads 359

Authors: B. Z. Manuel, H. D. Esmeralda, H. S. Felipe, D. R. Héctor, D. de la Torre Sebastián, R. L. Diego

Abstract:

Aluminum Matrix Composites reinforced with nanocrystalline Ni3Al carbon-coated intermetallic particles, were synthesized by powder metallurgy. Powder mixture of aluminum with 0.5-volume fraction of reinforcement particles was compacted by spark plasma sintering (SPS) technique and the compared with conventional sintering process. The better results for SPS technique were obtained in 520ºC-5kN-3min.The hardness (70.5±8 HV) and the elastic modulus (95 GPa) were evaluated in function of sintering conditions for SPS technique; it was found that the incorporation of these kind of reinforcement particles in aluminum matrix improve its mechanical properties. The densities were about 94% and 97% of the theoretical density. The carbon coating avoided the interfacial reaction between matrix-particle at high temperature (520°C) without show composition change either intermetallic dissolution.

Keywords: aluminum matrix composites, intermetallics, spark plasma sintering, nanocrystalline

Procedia PDF Downloads 420

Authors: M. Linek, P. Nita

Abstract:

In the previous airfield construction industry, pavements made of reinforced concrete have been used very rarely; however, the necessity to use this type of pavements in an emergency situations justifies the need reference to this issue. The paper concerns the problem of airfield pavement dimensioning made of reinforced concrete and the evaluation of selected dimensioning methods of reinforced concrete slabs intended for airfield pavements. Analysis of slabs dimensioning, according to classical method of limit states has been performed and it has been compared to results obtained in case of methods complying with Eurocode 2 guidelines. Basis of an analysis was a concrete slab of class C35/45 with reinforcement, located in tension zone. Steel bars of 16.0 mm have been used as slab reinforcement. According to comparative analysis of obtained results, conclusions were reached regarding application legitimacy of the discussed methods and their design advantages.

Keywords: rainforced concrete, cement concrete, airport pavements, dimensioning

Procedia PDF Downloads 232

Authors: Jain Jyoti, Jain Shorab, Sinha Shishir

Abstract:

In the current scenario, development of new biodegradable composites with the reinforcement of some plant derived natural fibers are in major research concern. Abundant quantity of these natural plant derived fibers including sisal, ramp, jute, wheat straw, pine, pineapple, bagasse, etc. can be used exclusively or in combination with other natural or synthetic fibers to augment their specific properties like chemical, mechanical or thermal properties. Among all natural fibers, wheat straw, bagasse, kenaf, pineapple leaf, banana, coir, ramie, flax, etc. pineapple leaf fibers have very good mechanical properties. Being hydrophilic in nature, pineapple leaf fibers have very less affinity towards all types of polymer matrixes. Not much work has been carried out in this area. Surface treatments like alkaline treatment in different concentrations were conducted to improve its compatibility towards hydrophobic polymer matrix. Pineapple leaf fiber epoxy composites have been prepared using hand layup method. Effect of variation in fiber loading up to 20% in epoxy composites has been studied for mechanical properties like tensile strength and flexural strength. Analysis of fiber morphology has also been studied using FTIR, XRD. SEM micrographs have also been studied for fracture surface.

Keywords: composite, mechanical, natural fiber, pineapple leaf fiber

Procedia PDF Downloads 215

Authors: Migena Mana, Ahmed Khalid Syed, Abdul Malik, Nikhil Cherian

Abstract:

In recent years, advancements in deep learning enabled researchers to tackle the problem of self-driving cars. Car companies use huge datasets to train their deep learning models to make autonomous cars a reality. However, this approach has certain drawbacks in that the state space of possible actions for a car is so huge that there cannot be a dataset for every possible road scenario. To overcome this problem, the concept of reinforcement learning (RL) is being investigated in this research. Since the problem of autonomous driving can be modeled in a simulation, it lends itself naturally to the domain of reinforcement learning. The advantage of this approach is that we can model different and complex road scenarios in a simulation without having to deploy in the real world. The autonomous agent can learn to drive by finding the optimal policy. This learned model can then be easily deployed in a real-world setting. In this project, we focus on three RL algorithms: Q-learning, Deep Deterministic Policy Gradient (DDPG), and Proximal Policy Optimization (PPO). To model the environment, we have used TORCS (The Open Racing Car Simulator), which provides us with a strong foundation to test our model. The inputs to the algorithms are the sensor data provided by the simulator such as velocity, distance from side pavement, etc. The outcome of this research project is a comparative analysis of these algorithms. Based on the comparison, the PPO algorithm gives the best results. When using PPO algorithm, the reward is greater, and the acceleration, steering angle and braking are more stable compared to the other algorithms, which means that the agent learns to drive in a better and more efficient way in this case. Additionally, we have come up with a dataset taken from the training of the agent with DDPG and PPO algorithms. It contains all the steps of the agent during one full training in the form: (all input values, acceleration, steering angle, break, loss, reward). This study can serve as a base for further complex road scenarios. Furthermore, it can be enlarged in the field of computer vision, using the images to find the best policy.

Keywords: autonomous driving, DDPG (deep deterministic policy gradient), PPO (proximal policy optimization), reinforcement learning

Procedia PDF Downloads 120

Authors: Manal Bouasria, Mohammed-Hichem Benzaama, Valérie Pralong, Yassine El Mendili

Abstract:

Reducing the quantity of cement in cementitious composites can help to reduce the environmental effect of construction materials. By-products such as ferronickel slags (FNS), fly ash (FA), and Crepidula fornicata (CR) are promising options for cement replacement. In this work, we investigated the relevance of substituting cement with FNS-CR and FA-CR on the mechanical properties of mortar and on the thermal properties of concrete. Foraging intervals ranging from 2 to 28 days, the mechanical properties are obtained by 3-point bending and compression tests. The chosen mix is used to construct a prototype in order to study the material’s hygrothermal performance. The data collected by the sensors placed on the prototype was utilized to build an artificial neural network.

Keywords: artificial neural network, cement, circular economy, concrete, by products

Procedia PDF Downloads 93

Authors: Shunichi Ohmori, Sirawadee Arunyanart, Kazuho Yoshimoto

Abstract:

We consider a periodic review inventory control problem of minimizing production cost, inventory cost, and lost-sales under demand uncertainty, in which product demands are not specified exactly and it is only known to belong to a given uncertainty set, yet the constraints must hold for possible values of the data from the uncertainty set. We propose a robust optimization formulation for obtaining lowest cost possible and guaranteeing the feasibility with respect to range of order quantity and inventory level under demand uncertainty. Our formulation is based on the adaptive robust counterpart, which suppose order quantity is affine function of past demands. We derive certainty equivalent problem via second-order cone programming, which gives 'not too pessimistic' worst-case.

Keywords: robust optimization, inventory control, supply chain managment, second-order programming

Procedia PDF Downloads 381

Authors: Ali Reza Ghanbarnezhad Ghazvini, Seyyed Hamid Reza Mosayyebi

Abstract:

In recent years, an earthquake has occurred approximately every five years in certain regions of Iran. To mitigate the impact of these seismic events, it is crucial to identify and thoroughly assess the vulnerability of buildings and infrastructure, ensuring their safety through principled reinforcement. By adopting new methods of risk assessment, we can effectively reduce the potential risks associated with future earthquakes. In our research, we have observed that the coefficient of behavior in the fourth chapter is 1.65 for the initial structure and 1.72 for the Superframe structure. This indicates that the Superframe structure can enhance the strength of the main structural members by approximately 10% through the utilization of super beams. Furthermore, based on the comparative analysis between the two structures conducted in this study, we have successfully designed a stronger structure with minimal changes in the coefficient of behavior. Additionally, this design has allowed for greater energy dissipation during seismic events, further enhancing the structure's resilience to earthquakes. By comprehensively examining and reinforcing the vulnerability of buildings and infrastructure, along with implementing advanced risk assessment techniques, we can significantly reduce casualties and damages caused by earthquakes in Iran. The findings of this study offer valuable insights for civil engineering professionals in the field of structural engineering, aiding them in designing safer and more resilient structures.

Keywords: modal pushover analysis, response modification factor, high-strength concrete, concrete shear walls, high-rise building

Procedia PDF Downloads 114

Authors: Burkhard Hehenkamp, Tahmina Faizi

Abstract:

In this paper, we analyze the formation of cartels along with the stability of the cartel for the case of three firms that produce differentiated products and differ in their cost of production. Both cost and demand are linear, and firms compete in quantities once a cartel has been formed (or not). It turns out that the degree of product differentiation has a direct effect on the incentive to form a cartel. Firstly, when goods are complements or close substitutes, firms form a grand coalition. Secondly, for weak and medium substitutes, the firm with the lowest cost prefers to remain independent, while both other firms form a coalition. We also find that the producer profit of the stable coalition structure is nonmonotonic in the degree of product differentiation.

Keywords: collusion, cartel formation, cartel stability, differentiated market, quantity competition, oligopolies

Procedia PDF Downloads 80

Authors: Marwa Hussein Mohamed, Mohamed Helmy Khafagy, Samah Ahmed Senbel

Abstract:

Map Reduce is a programming model used to handle and support massive data sets. Rapidly increasing in data size and big data are the most important issue today to make an analysis of this data. map reduce is used to analyze data and get more helpful information by using two simple functions map and reduce it's only written by the programmer, and it includes load balancing , fault tolerance and high scalability. The most important operation in data analysis are join, but map reduce is not directly support join. This paper explains two-way map-reduce join algorithm, semi-join and per split semi-join, and proposes new algorithm hash semi-join that used hash table to increase performance by eliminating unused records as early as possible and apply join using hash table rather than using map function to match join key with other data table in the second phase but using hash tables isn't affecting on memory size because we only save matched records from the second table only. Our experimental result shows that using a hash table with hash semi-join algorithm has higher performance than two other algorithms while increasing the data size from 10 million records to 500 million and running time are increased according to the size of joined records between two tables.

Keywords: map reduce, hadoop, semi join, two way join

Procedia PDF Downloads 488

Authors: Leila Jafari, Viliam Makis

Abstract:

In this paper, the joint optimization of the economic manufacturing quantity (EMQ), safety stock level, and condition-based maintenance (CBM) is presented for a partially observable, deteriorating system subject to random failure. The demand is stochastic and it is described by a Poisson process. The stochastic model is developed and the optimization problem is formulated in the semi-Markov decision process framework. A modification of the policy iteration algorithm is developed to find the optimal policy. A numerical example is presented to compare the optimal policy with the policy considering zero safety stock.

Keywords: condition-based maintenance, economic manufacturing quantity, safety stock, stochastic demand

Procedia PDF Downloads 444

Authors: S. Tireche, A. Tairi abdelaziz

Abstract:

Environmental damage has increased steadily in recent decades: Depletion of natural resources, destruction of the ozone layer, greenhouse effect, degradation of the quality of life, land use etc. New terms have emerged as: "Prevention rather than cure" or "polluter pays" falls within the principles of common sense, their practical implementation still remains fragmented. Among the avenues to be explored, one of the most promising is certainly one that focuses on product design. Indeed, where better than during the design phase, can reduce the source of future impacts on the environment? What choices or those of design, they influence more on the environmental characteristics of products? The most currently recognized at the international level is the analysis of the life cycle (LCA) and Life Cycle Assessment, subject to International Standardization (ISO 14040-14043). LCA provides scientific and objective assessment of potential impacts of the product or service, considering its entire life cycle. This approach makes it possible to minimize impacts to the source in pollution prevention. It is widely preferable to curative approach, currently majority in the industrial crops, led mostly by a report of pollution. The "product" is to reduce the environmental impacts of a given product, taking into account all or part of its life cycle. Currently, there are emerging tools, known as eco-design. They are intended to establish an environmental profile of the product to improve its environmental performance. They require a quantity sufficient information on the product for each phase of its life cycle: raw material extraction, manufacturing, distribution, usage, end of life (recycling or incineration or deposit) and all stages of transport. The assessment results indicate the sensitive points of the product studied, points on which the developer must act.

Keywords: eco design, impact, life cycle analysis (LCA), sustainability

Procedia PDF Downloads 394

Authors: Ki-Seok Kwon, Heung-Youl Kim

Abstract:

This research evaluated fire resistances of high-strengthened reinforced concrete (RC) column, spiral wire rope which applied with 60, and 100MPa. The fire resistance test of RC column with loading condition was conducted following the ISO 834 (3 hours). This experiment set mixing of fiber (PP fiber, Steel fiber) and types of horizontal reinforcement as a variable of reinforcement method. The fire resistance test measured the main steel bar’s max and mean temperatures also the shrinkage and shrinking ratio of columns(500 X 500 X 3,000mm) with loadings. As a result, the specimen of 60MPa attained three hours fire resistance with only spiral wire rope. Also, the specimen of 100MPa must be reinforced with fibers and spiral wire rope to attain three hours fire resistance.

Keywords: reinforced concrete column, high strength concrete, wire rope, fire resistance test

Procedia PDF Downloads 298

Authors: Metwally Abdel Aziz Ahmed, Ahmed Shaban Abdel Hay Gabr, Inas Mohamed Saleh

Abstract:

The main objective of structural design is to ensure safety and functional performance requirements of a structural system for its target reliability levels. In this study, the reliability index for the reinforcement concrete slender columns with rectangular cross section is studied. The variable parameters studied include the loads, the concrete compressive strength, the steel yield strength, the dimensions of concrete cross-section, the reinforcement ratio, and the location of steel placement. Risk analysis program was used to perform the analytical study. The effect of load eccentricity on the reliability index of reinforced concrete slender column was studied and presented. The results of this study indicate that the good quality control improve the performance of slender reinforced columns through increasing the reliability index β.

Keywords: reliability, reinforced concrete, safety, slender column

Procedia PDF Downloads 423

Authors: Chinnawat Satsananan

Abstract:

This research aimed to find amount of heavy metal in herb used in Dusit community and compare of heavy metal in each part by quantity in herb and standard determination in Thai herb books to develop a sustainable quality of life, the result of study in 14 herbs do not find sample of heavy metal., by quantity of heavy contamination of 4 kinds: Cd, Co, Fe and Pb have lower than standard of 2 organizations: Thai herb standard, and World Health Organization, from the test 14 herbs have Fe in every part of herbs and all 14 kinds has Fe that is necessary for our health.

Keywords: herbs plants, heavy metal, Dusit district, sustainable quality of life

Procedia PDF Downloads 351

Authors: A. Benyahia, M. Zergoug, M. Amir, M. Fodil

Abstract:

The main objective of this work is to enhance the Pulsed Eddy Current (PEC) response from the aluminum structure using signal processing. Cracks and metal loss in different structures cause changes in PEC response measurements. In this paper, time-frequency analysis is used to represent PEC response, which generates a large quantity of data and reduce the noise due to measurement. Power Spectral Density (PSD) after Wavelet Decomposition (PSD-WD) is proposed for defect detection. The experimental results demonstrate that the cracks in the surface can be extracted satisfactorily by the proposed methods. The validity of the proposed method is discussed.

Keywords: DT, pulsed eddy current, continuous wavelet transform, Mexican hat wavelet mother, defect detection, power spectral density.

Procedia PDF Downloads 207

Authors: Vivian A. Galindo, Maria C. Galvis, Jaime R. Obando, Alvaro Guarin

Abstract:

In Colombia, most of the roads of the national tertiary road network are unpaved roads with granular rolling surface. These are very important ways of guaranteeing the mobility of people, products, and inputs from the agricultural sector from the most remote areas to urban centers; however, it has not paid much attention to the search for alternatives to avoid the occurrence of deteriorations that occur shortly after its commissioning. In recent years, geosynthetics have been used satisfactorily to reinforce unpaved roads on soft soils, with geotextiles and geogrids being the most widely used. The interaction of the geogrid and the aggregate minimizes the lateral movement of the aggregate particles and increases the load capacity of the material, which leads to a better distribution of the vertical stresses, consequently reducing the vertical deformations in the subgrade. Taking into account the above, the research aimed at the mechanical behavior of the granular material, used in unpaved roads with and without the presence of geogrids, from the development of laboratory tests through the loaded wheel tester (LWT). For comparison purposes, the reinforced conditions and traffic conditions to which this type of material can be accessed in practice were simulated. In total four types of geogrids, were tested with granular material; this means that five test sets, the reinforced material and the non-reinforced control sample were evaluated. The results of the numbers of load cycles and depth rutting supported by each test body showed the influence of the properties of the reinforcement on the mechanical behavior of the assembly and the significant increases in the number of load cycles of the reinforced specimens in relation to those without reinforcement.

Keywords: geosynthetics, load wheel tester LWT, tertiary roads, unpaved road, vertical deformation

Procedia PDF Downloads 219

Authors: Masato Nishi, Tetsushi Kaburagi, Masashi Kurose, Tei Hirashima, Tetsusei Kurasiki

Abstract:

The process of thermoforming a carbon fiber reinforced thermoplastic (CFRTP) has increased its presence in the automotive industry for its wide applicability to the mass production car. A non-isothermal forming for CFRTP can shorten its cycle time to less than 1 minute. In this paper, the textile reinforcement FE model which the authors proposed in a previous work is extended to the CFRTP model for non-isothermal forming simulation. The effect of thermoplastic is given by adding shell elements which consider thermal effect to the textile reinforcement model. By applying Reuss model to the stress calculation of thermoplastic, the proposed model can accurately predict in-plane shear behavior, which is the key deformation mode during forming, in the range of the process temperature. Using the proposed model, thermoforming simulation was conducted and the results are in good agreement with the experimental results.

Keywords: carbon fiber reinforced thermoplastic, finite element analysis, pre-impregnated textile composite, non-isothermal forming

Procedia PDF Downloads 407

Authors: Nihal Murali, Kunal Gupta, Surekha Bhanot

Abstract:

Training of artificial neural networks (ANNs) using reinforcement learning (RL) techniques is being widely discussed in the robot learning literature. The high model complexity of ANNs along with the model-free nature of RL algorithms provides a desirable combination for many robotics applications. There is a huge need for algorithms that generalize using raw sensory inputs, such as vision, without any hand-engineered features or domain heuristics. In this paper, the standard control problem of line following robot was used as a test-bed, and an ANN controller for the robot was trained on images from a live video feed using Q-learning. A virtual agent was first trained in simulation environment and then deployed onto a robot’s hardware. The robot successfully learns to traverse a wide range of curves and displays excellent generalization ability. Qualitative analysis of the evolution of policies, performance and weights of the network provide insights into the nature and convergence of the learning algorithm.

Keywords: artificial neural networks, q-learning, reinforcement learning, robot learning

Procedia PDF Downloads 351

Authors: Valentin Père, Mathieu Milhé, Fabien Baillon, Jean-Louis Dirion

Abstract:

Microgrids offer potential solutions to meet the need for local grid stability and increase isolated networks autonomy with the integration of intermittent renewable energy production and storage facilities. In such a context, sizing production and storage for a given network is a complex task, highly depending on input data such as power load profile and renewable resource availability. This work aims at developing an operating cost computation methodology for different microgrid designs based on the use of deep reinforcement learning (RL) algorithms to tackle the optimal operation problem in stochastic environments. RL is a data-based sequential decision control method based on Markov decision processes that enable the consideration of random variables for control at a chosen time scale. Agents trained via RL constitute a promising class of Energy Management Systems (EMS) for the operation of microgrids with energy storage. Microgrid sizing (or design) is generally performed by minimizing investment costs and operational costs arising from the EMS behavior. The latter might include economic aspects (power purchase, facilities aging), social aspects (load curtailment), and ecological aspects (carbon emissions). Sizing variables are related to major constraints on the optimal operation of the network by the EMS. In this work, an islanded mode microgrid is considered. Renewable generation is done with photovoltaic panels; an electrochemical battery ensures short-term electricity storage. The controllable unit is a hydrogen tank that is used as a long-term storage unit. The proposed approach focus on the transfer of agent learning for the near-optimal operating cost approximation with deep RL for each microgrid size. Like most data-based algorithms, the training step in RL leads to important computer time. The objective of this work is thus to study the potential of Batch-Constrained Q-learning (BCQ) for the optimal sizing of microgrids and especially to reduce the computation time of operating cost estimation in several microgrid configurations. BCQ is an off-line RL algorithm that is known to be data efficient and can learn better policies than on-line RL algorithms on the same buffer. The general idea is to use the learned policy of agents trained in similar environments to constitute a buffer. The latter is used to train BCQ, and thus the agent learning can be performed without update during interaction sampling. A comparison between online RL and the presented method is performed based on the score by environment and on the computation time.

Keywords: batch-constrained reinforcement learning, control, design, optimal

Procedia PDF Downloads 94

Authors: Szymon Rutkowski, Tomasz Korbak

Abstract:

We present a reinforcement-learning-based method of training neural machine translation models without parallel corpora. The standard encoder-decoder approach to machine translation suffers from two problems we aim to address. First, it needs parallel corpora, which are scarce, especially for low-resource languages. Second, it lacks psychological plausibility of learning procedure: learning a foreign language is about learning to communicate useful information, not merely learning to transduce from one language’s 'encoding' to another. We instead pose the problem of learning to translate as learning a policy in a communication game between two agents: the translator and the classifier. The classifier is trained beforehand on a natural language inference task (determining the entailment relation between a premise and a hypothesis) in the target language. The translator produces a sequence of actions that correspond to generating translations of both the hypothesis and premise, which are then passed to the classifier. The translator is rewarded for classifier’s performance on determining entailment between sentences translated by the translator to disciple’s native language. Translator’s performance thus reflects its ability to communicate useful information to the classifier. In effect, we train a machine translation model without the need for parallel corpora altogether. While similar reinforcement learning formulations for zero-shot translation were proposed before, there is a number of improvements we introduce. While prior research aimed at grounding the translation task in the physical world by evaluating agents on an image captioning task, we found that using a linguistic task is more sample-efficient. Natural language inference (also known as recognizing textual entailment) captures semantic properties of sentence pairs that are poorly correlated with semantic similarity, thus enforcing basic understanding of the role played by compositionality. It has been shown that models trained recognizing textual entailment produce high-quality general-purpose sentence embeddings transferrable to other tasks. We use stanford natural language inference (SNLI) dataset as well as its analogous datasets for French (XNLI) and Polish (CDSCorpus). Textual entailment corpora can be obtained relatively easily for any language, which makes our approach more extensible to low-resource languages than traditional approaches based on parallel corpora. We evaluated a number of reinforcement learning algorithms (including policy gradients and actor-critic) to solve the problem of translator’s policy optimization and found that our attempts yield some promising improvements over previous approaches to reinforcement-learning based zero-shot machine translation.

Keywords: agent-based language learning, low-resource translation, natural language inference, neural machine translation, reinforcement learning

Procedia PDF Downloads 103

Authors: Mohammad Reza Koushki Ardestani, Saeed Daneshmand, Mohammad Heydari Vini

Abstract:

In the present study, AA7075/SiO2 composites have been fabricated via liquid metallurgy process. Using the degassing process, the wet ability of the molten aluminum alloys increased which improved the bonding between aluminum matrix and reinforcement (SiO2) particles. AA7075 alloy and SiO2 particles were taken as the base matrix and reinforcements, respectively. Then, contents of 2.5 and 5 wt. % of SiO2 subdivisions were added into the AA7075 matrix. To improve wettability and distribution, reinforcement particles were pre-heated to a temperature of 550°C for each composite sample. A uniform distribution of SiO2 particles was observed through the matrix alloy in the microstructural study. A hardened EN32 steel disc as the counter face was used to evaluate the wear rate pin-on-disc, a wear testing machine containing. The results showed that the wear rate of the AA/SiO2 composites was lesser than that of the monolithic AA7075 samples. Finally, The SEM worn surfaces of samples were investigated.

Keywords: Al7075, SiO₂, wear, composites, stir casting

Procedia PDF Downloads 54

Authors: Parisa Ahmadi Oliaei, Seyed Abolhassan Naeini

Abstract:

The knowledge of soil–reinforcement interaction parameters is particularly important in the design of reinforced soil structures. The pull-out test is one of the most widely used tests in this regard. The results of tensile tests may be very sensitive to boundary conditions, and more research is needed for a better understanding of the Pull-out response of reinforcement, so numerical analysis using the finite element method can be a useful tool for the understanding of the Pull-out response of soil-geogrid interaction. The main objective of the present study is to compare the numerical and experimental results of Pull- out a test on geogrid-reinforced sandy soils interactions. Plaxis 2D finite element software is used for simulation. In the present study, the pull-out test modeling has been done on sandy soil. The effect of geogrid hardness was also investigated by considering two different types of geogrids. The numerical results curve had a good agreement with the pull-out laboratory results.

Keywords: plaxis, pull-out test, sand, soil- geogrid interaction

Procedia PDF Downloads 142

Authors: Sidra Saleemi, Arsalan Khan, Urooj Baig, Tahir Jamil

Abstract:

In this research, it was examined that some residual amount of bleaching chemicals left in the liquor, this amount is more in Batch-wise process as compared to continuous process. These chemicals can be recovered and reused for bleaching by adding more quantity of fresh bleaching chemicals and water, this quantity will be required to balance the recipe for fabric. This liquor is recovered and samples were bleached with different modified recipe of liquor for both processes i.e. Batch-wise and continuous process. Every time good results were achieved with negligible variation in the quality parameter between the fabric bleached with fresh liquor and the fabric bleached with Recovered Liquor. Additionally, samples were dyed, and found that dyeing can be done easily on samples bleached with recover liquor.

Keywords: bleaching process, hydrogen peroxide, sodium hydroxide, liquor recovery

Procedia PDF Downloads 326

Authors: Alica Höpken, Hergen Pargmann

Abstract:

The resource-efficient planning of the complex production planning processes in the meat industry and the reduction of food waste is a permanent challenge. The complexity of the production planning process occurs in every part of the supply chain, from agriculture to the end consumer. It arises from long and uncertain planning phases. Uncertainties such as stochastic yields, fluctuations in demand, and resource variability are part of this process. In the meat industry, waste mainly relates to incorrect storage, technical causes in production, or overproduction. The high amount of food waste along the complex supply chain in the meat industry could not be reduced by simple solutions until now. Therefore, resource-efficient production planning by conventional methods is currently only partially feasible. The realization of intelligent, automated production planning is basically possible through the application of machine learning algorithms, such as those of reinforcement learning. By applying the adapted design thinking method, machine learning methods (especially reinforcement learning algorithms) are used for the complex production planning process in the meat industry. This method represents a concretization to the application area. A resource-efficient production planning process is made available by adapting the design thinking method. In addition, the complex processes can be planned efficiently by using this method, since this standardized approach offers new possibilities in order to challenge the complexity and the high time consumption. It represents a tool to support the efficient production planning in the meat industry. This paper shows an elegant adaption of the design thinking method to apply the reinforcement learning method for a resource-efficient production planning process in the meat industry. Following, the steps that are necessary to introduce machine learning algorithms into the production planning of the food industry are determined. This is achieved based on a case study which is part of the research project ”REIF - Resource Efficient, Economic and Intelligent Food Chain” supported by the German Federal Ministry for Economic Affairs and Climate Action of Germany and the German Aerospace Center. Through this structured approach, significantly better planning results are achieved, which would be too complex or very time consuming using conventional methods.

Keywords: change management, design thinking method, machine learning, meat industry, reinforcement learning, resource-efficient production planning

Procedia PDF Downloads 105

Authors: P. Raajeswari, S. M. Devatha, R. Pragatheeswari

Abstract:

The edible packaging is a new trendsetter in the era of modern packaging. The researchers and food scientist recognise edible packaging as a useful alternative or addition to conventional packaging to reduce waste and to create novel applications for improving product stability. Starch was extracted from different sources that contains abundantly like potato, tapioca, rice, wheat, and corn. The starch based edible films and cutleries are developed as an alternative for conventional packages providing the nutritional benefit when consumed along with the food. The development of starch based edible films by the extraction of starch from various raw ingredients at lab scale level. The films are developed by the employment of plasticiser at different concentrations of 1.5ml and 2ml. The films developed using glycerol as a plasticiser in filmogenic solution to increase the flexibility and plasticity of film. It reduces intra and intermolecular forces in starch, and it increases the mobility of starch based edible films. The films developed are tested for its functional properties such as thickness, tensile strength, elongation at break, moisture permeability, moisture content, and puncture strength. The cutleries like spoons and cups are prepared by making dough and rolling the starch along with water. The overall results showed that starch based edible films absorbed less moisture, and they also contributed to the low moisture permeability with high tensile strength. Food colorants extracted from red onion peel, pumpkin, and red amaranth adds on the nutritive value, colour, and attraction when incorporated in edible cutleries, and it doesn’t influence the functional properties. Addition of a low quantity of glycerol in edible films and colour extraction from onion peel, pumpkin, and red amaranth enhances biodegradability and provides a good quantity of nutrients when consumed. Therefore, due to its multiple advantages, food starch can serve as the best response for eco-friendly industrial products aimed to replace single use plastics at low cost.

Keywords: edible films, edible cutleries, plasticizer, glycerol, starch, functional property

Procedia PDF Downloads 161

Authors: A. Natali, F. V. Lippi, F. Morelli, W. Salvatore, J. H. M. De Paula Filho, P. Pol

Abstract:

Automated Rack Supported Warehouses (ARSWs) are storage buildings whose load-bearing structure is made of the same steel racks where goods are stocked. These racks are made of cold formed elements, and the main supporting structure is repeated several times along the length of the building, resulting in a huge quantity of steel. The possibility of using high strength steel to manufacture the traditional cold-formed profiles used for ARSWs is numerically investigated, with the aim of reducing the necessary steel quantity but guaranteeing optimal structural performance levels.

Keywords: steel racks, automated rack supported warehouse, thin-walled cold-formed elements, high strength steel, structural optimization

Procedia PDF Downloads 126

Authors: Vara Prasad Vemu

Abstract:

The aim of the present work is to investigate the mechanical properties and water absorption capacity of carbon and basalt fibers mixed with matrix epoxy. At present, there is demand for nature friendly products. Basalt reinforced composites developed recently, and these mineral amorphous fibres are a valid alternative to carbon fibres for their lower cost and to glass fibres for their strength. The present paper describes briefly on basalt and carbon fibres (uni-directional) which are used as reinforcement materials for composites. The matrix epoxy (LY 556-HY 951) is taken into account to assess its influence on the evaluated parameters. In order to use reinforced composites for structural applications, it is necessary to perform a mechanical characterization. With this aim experiments like tensile strength, flexural strength, hardness and water absorption are performed. Later the mechanical properties obtained from experiments are compared with ANSYS software results.

Keywords: carbon fibre, basalt fibre, uni-directional, reinforcement, mechanical tests, water absorption test, ANSYS

Procedia PDF Downloads 175

Authors: Rahul Paul, Kedar Nath Das

Abstract:

The paradigm of Reinforcement Learning (RL) has become prominent in training intelligent agents to make decisions in environments that are both dynamic and uncertain. The primary objective of RL is to optimize the policy of an agent in order to maximize the cumulative reward it receives throughout a given period. Nevertheless, the process of optimization presents notable difficulties as a result of the inherent trade-off between exploration and exploitation, the presence of extensive state-action spaces, and the intricate nature of the dynamics involved. Evolutionary Optimization Methods (EOMs) have garnered considerable attention as a supplementary approach to tackle these challenges, providing distinct capabilities for optimizing RL policies and value functions. The ongoing advancement of research in both RL and EOMs presents an opportunity for significant advancements in autonomous decision-making systems. The convergence of these two fields has the potential to have a transformative impact on various domains of artificial intelligence (AI) applications. This article highlights the considerable influence of EOMs in enhancing the capabilities of RL. Taking advantage of evolutionary principles enables RL algorithms to effectively traverse extensive action spaces and discover optimal solutions within intricate environments. Moreover, this paper emphasizes the practical implementations of EOMs in the field of RL, specifically in areas such as robotic control, autonomous systems, inventory problems, and multi-agent scenarios. The article highlights the utilization of EOMs in facilitating RL agents to effectively adapt, evolve, and uncover proficient strategies for complex tasks that may pose challenges for conventional RL approaches.

Keywords: machine learning, reinforcement learning, loss function, optimization techniques, evolutionary optimization methods

Procedia PDF Downloads 54