Instant flow distribution network optimization in liquid composite molding using deep reinforcement learning

CFRP (Carbon fiber reinforced plastics) exhibit superior mechanical properties, and are recently applied to large structural materials using the VaRTM (Vacuum assisted Resin Transfer Molding) method. In the vicinity of the resin inlet of this method, and at the joint of CFRP structures, it is pointed out that the disorder in fiber orientation often occurs. In this study, therefore, the effect of local off-axis fibers array on the tensile properties of unidirectional CFRP was basically investigated. An experimental device for making a local off-axis fibers array on a CFRP lamina was newly developed, and various unidirectional CFRP laminae were prepared by changing the fiber orientation angle and the size. The results show that the tensile strength of CFRP laminae with off-axis fibers array decreases with increasing the fiber orientation angle and the length, and depends on the degree of fiber bundle waving. On the other hand, Young's moduli of the CFRP lamnae show an almost constant value, which is relatively close to that of CFRP lamina without off-axis fibers array. Finally, Young's moduli and tensile strength of the CFRP laminae were analyzed from a proposed composite model with an off-axis fibers array, and its validity was discussed.

Bending strength of CFRP laminated adhesive joints fabricated by vacuum-assisted resin transfer molding

Carbon fiber reinforced plastics (CFRP) having both high strength and lightness has been widely used in various fields, however, the delamination failure is the major problem. Since the 3D textile CFRP has a unique mechanical properties such as good peeling resistance and high toughness, it has aroused a number of advantages. For the easy weaving and cost performance 2.5D textile will be also worth in applications. Meanwhile, the feature of the carbon fiber with excellent electrical conductivity and electromagnetic shielding performance was considered to be used as an electromagnetic shield material. Thus, the electromagnetic shielding performance of 3D fabric structure CFRP is investigated for improving electromagnetic shielding effect. The carbon fiber woven fabric was designed for this new multi-layer structure in an efficient vacuum assisted resin transfer molding (VARTM) process to produce a carbon fiber woven composites. The mechanical properties and electromagnetic shielding performance were discussed.

Mode I fracture toughness of asymmetric metal-composite adhesive joints

Vacuum-assisted resin transfer molding (VARTM) has several inherent shortcomings such as long mold filling times, low fiber volume fraction, and high void content in fabricated laminates. These problems in VARTM mainly arise from the limited compaction of the laminate and low resin pressure. Pressurized infusion (PI) molding introduced in this paper overcomes these disadvantages by (i) applying high compaction pressure on the laminate by an external pressure chamber placed on the mold and (ii) increasing the resin pressure by pressurizing the inlet resin reservoir. The effectiveness of PI molding was verified by fabricating composite laminates at various levels of chamber and inlet pressures and investigating the effect of these parameters on the fill time, fiber volume fraction, and void content. Furthermore, spatial distribution of voids was characterized by employing a unique method, which uses a flatbed scanner to capture the high-resolution planar scan of the fabricated laminates. The results revealed that PI molding reduced fill time by 45%, increased fiber volume fraction by 16%, reduced void content by 98%, improved short beam shear (SBS) strength by 14%, and yielded uniform spatial distribution of voids compared to those obtained by conventional VARTM.

Learning ability is the basic characteristic of human intelligence. The July 1, 2005 issue of Science published a list of 125 important questions in science. Among them, the question 94 “What are the limits of learning by machines?”. The annotation “Computers can already beat the world’s best chess players, and they have a wealth of information on the Web to draw on. But abstract reasoning is still beyond any machine”. In recent artificial intelligence has made great progresses. In 1997, the rise of the man-machine war, IBM Supercomputer Deep Blue defeated the chess master Garry Kasparov. On February 14, 2011, IBM’s Watson supercomputer won a practice round against Jeopardy champions Ken Jennings and Brad Rutter. In March 2016, Google DeepMind’s AlphaGo sealed a 4-1 victory over a South Korean Go grandmaster Lee Se-dol. This paper focuses on the machine learning methods of AlphaGo, including reinforcement learning, deep learning, deep reinforcement learning, analysis of the existing problems and the latest research progress. Deep reinforcement learning is the combination of deep learning and reinforcement learning, which can realize the learning algorithm from the perception to action. Simply said, this is the same as human behavior, input sensing information such as vision, and then, direct output action through the deep neural network. Deep reinforcement learning has the potential to learn a variety of skills for the robot to achieve full autonomy. Even though reinforcement learning is practiced successfully, but feature states need to manually set, for complex scene is a difficult thing, especially easy to cause the dimension disaster, and expression is not good. In 2010, Sascha Lange and Martin Riedmiller proposed deep auto-encoder neural networks in reinforcement learning to extract feature, which is used to control the visual correlation. In 2013, DeepMind proposed deep Q-network (DQN) in NIPS 2013, using convolution neural network to extract features, and then applied in reinforcement learning. They continue to improve and published an improved version of DQN on Nature in 2015, which has aroused widespread concern. In order to break through the limits of learning by machines, cognitive machine learning is proposed, which is the combination of machine learning and brain cognition, so that the machine intelligence is constantly evolving, and gradually reaches the human level of artificial intelligence. A cognitive model entitled Consciousness And Memory (CAM) is proposed by author, which consists of memory, consciousness, high-level cognitive functions, perception and motor. High-level cognitive functions of the brain include learning, language, thinking, decision making, emotion, and so on. Learning is a course to accept the stimulus through the nervous system and obtain new behavior, habits and accumulation experience. According to the current research progress of brain science and cognitive science, cognitive machine learning may be interested in learning emergence, procedural memory knowledge learning, learning evolution and so on. For intelligence, so-called evolution is refers to the learning of learning and the structure also follows the change. It is important to record the learning result by structure changing and improve the learning method.

Deep reinforcement learning (DRL) is one of the emerging areas of machine learning which focuses on maximized rewards. DRL is a type of machine learning that combines reinforcement learning and deep learning. It uses a series of algorithms to enable an agent to learn how to make decisions in a complex environment. DRL is a subset of artificial intelligence that focuses on making decisions based on the environment and the rewards associated with each action.The goal of DRL is to maximize the long-term reward of an agent. In order to do this, the agent must use a combination of deep learning, reinforcement learning and other AI techniques to learn which actions will lead to the highest reward. DRL is used to solve a variety of problems, from playing video games to controlling robots. It is also used in autonomous driving and robotics, as well as for financial trading. DRL is a powerful tool for solving complex problems and has been used in a variety of research projects. DRL has the potential to revolutionize the way we interact with machines and the environment.

Deep Reinforcement Learning (DRL) has emerged as a transformative paradigm in the field of artificial intelligence (AI), offering unprecedented capabilities in decision-making across diverse domains. This article explores the profound impact of DRL on enhancing the decision-making capabilities of AI systems, elucidating its underlying principles, applications, and implications.DRL represents a fusion of deep learning and reinforcement learning, enabling machines to learn complex behaviors and make decisions by interacting with their environment. The utilization of neural networks allows DRL algorithms to handle high-dimensional input spaces, making it well-suited for tasks that involve intricate decision-making processes.One of the key strengths of DRL lies in its ability to address problems with sparse and delayed rewards, common challenges in traditional reinforcement learning. Through a process of trial and error, DRL algorithms can learn optimal decision strategies by navigating through a vast decision space, adapting to dynamic environments, and maximizing cumulative rewards over time.The applications of DRL span various domains, including robotics, finance, healthcare, gaming, and autonomous systems. In robotics, DRL facilitates the development of intelligent agents capable of autonomously navigating complex environments, performing intricate tasks, and adapting to unforeseen circumstances. In finance, DRL is leveraged for portfolio optimization, algorithmic trading, and risk management, demonstrating its potential to revolutionize traditional financial strategies.

In recent years, deep reinforcement learning (DRL) has made impressive achievements in many fields. However, existing DRL algorithms usually require a large amount of exploration to obtain a good action policy. In addition, in many complex situations, the reward function cannot be well designed to meet task requirements. These two problems will make it difficult for DRL to learn a good action policy within a relatively short period. The use of expert data can provide effective guidance and avoid unnecessary exploration. This study proposes a deep imitation reinforcement learning (DIRL) algorithm that uses a certain amount of expert demonstration data to speed up the training of DRL. In the proposed method, the learning agent imitates the expert's action policy by learning from demonstration data. After imitation learning, DRL is used to optimise the action policy in a self‐learning way. By experimental comparison on a video game called the Mario racing game, it is shown that the proposed DIRL algorithm with expert demonstration data can obtain much better performance than previous DRL algorithms without expert guidance.

: Vacuum assisted resin transfer molding (VARTM) and resin transfer molding (RTM) processes have been successfully developed to enable infusion of thermoplastic pre-polymers into glass and carbon fiber performs at temperatures up to 210 deg C. The pre-polymers used in this study were CBT (Cyclic Butylene Teraphthalate oligomer) leading to a PBT composite, APLC-12 (Anionically-Polymerised Lactam -12) leading to a PA-12 composite, and APLC-6 (Anionically-Polymerised Lactam -6) leading to a PA-6 composite. Mixtures of PA-12 and PA-6 were also examined. Traditionally it has not been easy to mold thermoplastic-reinforced composites due to their high viscosities. High temperatures and pressures were always required to fully impregnate fiber beds. Liquid composite molding processes, such as those demonstrated in this report, were virtually impossible. The employment of an activated pre-polymer melts eliminates the problems associated with high viscosity. All of the materials processed here were using vacuum pressure only.

Reinforcement learning is a mathematical framework for agents to interact intelligently with their environment. Unlike supervised learning, where a system learns with the help of labeled data, reinforcement learning agents learn how to act by trial and error only receiving a reward signal from their environments. A field where reinforcement learning has been prominently successful is robotics [3]. However, real-world control problems are also particularly challenging because of the noise and high- dimensionality of input data (e.g., visual input). In recent years, in the field of supervised learning, deep neural networks have been successfully used to extract meaning from this kind of data. Building on these advances, deep reinforcement learning was used to solve complex problems like Atari games and Go. Mnih et al. [1] built a system with fixed hyper parameters able to learn to play 49 different Atari games only from raw pixel inputs. However, in order to apply the same methods to real-world control problems, deep reinforcement learning has to be able to deal with continuous action spaces. Discretizing continuous action spaces would scale poorly, since the number of discrete actions grows exponentially with the dimensionality of the action. Furthermore, having a parametrized policy can be advantageous because it can generalize in the action space. Therefore with this thesis we study state-of-the-art deep reinforcement learning algorithm, Deep Deterministic Policy Gradients. We provide a theoretical comparison to other popular methods, an evaluation of its performance, identify its limitations and investigate future directions of research. The remainder of the thesis is organized as follows. We start by introducing the field of interest, machine learning, focusing our attention of deep learning and reinforcement learning. We continue by describing in details the two main algorithms, core of this study, namely Deep Q-Network (DQN) and Deep Deterministic Policy Gradients (DDPG). We then provide implementatory details of DDPG and our test environment, followed by a description of benchmark test cases. Finally, we discuss the results of our evaluation, identifying limitations of the current approach and proposing future avenues of research.

A new ensemble deep graph reinforcement learning network for spatio-temporal traffic volume forecasting in a freeway network

This paper introduces Porous Mold Process (PMP) as a reliable low-cost manufacturing process of fiber-reinforced composites. PMP using porous aluminum is applied to the vacuum-assisted resin transfer molding (VaRTM) process of carbon fiber reinforced plastics (CFRPs). Experimental evaluation on resin infusion behavior and quality and mechanical properties of the cured plates is performed with varying the dimensions of the plates. The results show that quality and mechanical properties of the cured plates using PMP are satisfactory, stable, and almost independent of plate dimensions. It is concluded that PMP provides a reliable and knowhow-less resin infusion process of composite materials.

Vacuum-assisted resin transfer molding (VARTM) process is one of the liquid composite molding (LCM) processes aimed at producing high-quality composite parts. The void content and fiber volume fraction of a VARTM part can be affected by many parameters and is critical to the mechanical properties and the quality of the part. In this paper, a series of experiments were conducted with a heated dual pressure control VARTM setup for investigating the effects of vacuum pressure, inlet pressure, and mold temperature on the void content and fiber volume fraction of polyester/E-glass fiber composite. It was found that stronger vacuum and higher mold temperature can better control and increase the fiber volume fraction; however, such a combination of strong vacuum and high mold temperature may also require a reduced inlet pressure for minimizing the void content. The need of pressure reduction can be explained with the compatibility between Darcy's flow and capillary flow in the fiber preform and can be calculated based on the room temperature VARTM results. The experimental results suggest that high mold temperature, high vacuum, and appropriately reduced inlet pressure can produce a VARTM part with high fiber volume fraction and low void content.

The objective of this study is to enhance the efficiency of urban air mobility (UAM) air transportation by determining the optimal design, materials, and processes of seven different manufactured fiber-reinforced plastics (FRP), and to confirm their applicability for next-generation UAM aircraft seats. Structural designs for the aircraft seat models were carried out using carbon fiber-reinforced plastic (CFRP), glass fiber-reinforced plastic (GFRP), and GFRP chop materials, employing Autoclave, Hot-press, and vacuum-assisted resin transfer molding (VaRTM) processes, resulting in a total of seven FRP configurations. It was confirmed that CFRP seats were 50% lower weight than aluminum models, while GFRP seats showed a 30% reduction. Although Autoclave processing resulted in the highest tensile strength at 985[Formula: see text]MPa, VaRTM processing also produced strength levels comparable to Autoclave processing. Structural integrity assessments of the seat models, utilizing the Korean aviation standards (KAS), confirmed that the designed seat models exhibited no failure or deformation under the conditions required by the technical standards. This study provides insights into the potential application of the seven types of FRP materials in the design of aircraft seats, offering weight reduction benefits and meeting the structural integrity requirements outlined by KAS.