Gel-casted porous Al<Subscript>2</Subscript>O<Subscript>3</Subscript> ceramics by use of natural fibres as pore developers

An approach using ceramic shaping by gel-casting and addition of natural tropical fibres as burnable pore developers has been explored for development of porous alumina materials with predetermined pore structures. The fibres used have been characterized in terms of size, shape, water uptake and surface charge. The degree of fibres loading in slurry has been varied. It has been shown that the amount of added fibres correlates well with the measured porosity, the shape of the developed pores however deviating slightly with that of the introduced fibres. The ceramic bodies have been characterised in terms of porosity, shrinkage and flexural strength. It could be assumed that at fibres amount in slurry above 30% V, an open porosity appears which is accompanied by a sharp drop in strength. In order to explore the benefits of the gel-casting method for complex bodies development, two key issues need to be addressed: air evacuation prior to casting and lowering slurry viscosity at fibre loadings higher than the tested ones. For reaching the latter objective, surface modification of fibres has been undertaken.     

砖土拱城门结构的安全性分析与加固

结合南京中山门城门结构 ,对多跨连续砖拱与芯墙土共同组成的复合结构作了有限元模拟及安全性分析。研究表明 ,基础不均匀沉降和芯土进水恶化对此类结构具有明显的不利影响。同时 ,提出内衬混凝土空间结构的加固方法 ,通过置换芯土 ,达到减轻砖拱及侧墙的负载、阻止裂缝扩展 ,提高结构整体性和稳定性的目的     

New Approach for Estimating the Deflection of Beams Reinforced with FRP Reinforcement

Due to concerns with corrosion, the use of fiber-reinforced polymer (FRP) as a replacement to conventional steel reinforcement has greatly increased over the last decade. Researchers have identified the distinctive mechanical and bond properties of FRP reinforcement that prevent the use of existing relationships to establish serviceability of concrete structures reinforced with such products. Although studies have modified these empirical relationships to describe the behavior of structures reinforced with FRP reinforcement, this paper will provide a new approach to estimate deflection of concrete beams by considering material properties of the reinforcement and incorporating the effects of tension stiffening. Accuracy and precision of the approach was established by performing a statistical analysis on a database containing 171 FRP-reinforced concrete beams. Results were compared to those from existing proposed relationships and indicate the potential of the method to estimate deflection at various service conditions.     

Advanced Boundary Element Analysis of Three-Dimensional Elastic Solids with Fiber Reinforcements

An advanced boundary element methodology for the analysis of three-dimensional fiber-reinforced elastic solids using the concept of fiber elements has been presented in this paper as an extension of the earlier work of Banerjee and co-workers. The previous simplified formulation was based on the assumption that the Poisson ratios of the matrix and the fiber are equal. However, this may not be a valid assumption for all values of elastic stiffness ratios and fiber to matrix volume ratios. Moreover, such restrictions do not allow for any future extensions to nonlinear analysis. Also, because of limiting computing power available at that time, their implementation was restricted to only a small number of fiber elements in a given analysis. The new algorithm proposed in this work does not make any such assumptions. In this algorithm, the fibers are semianalytically integrated and the resulting equations are back-substituted into the equations for the composite matrix without any further approximation. The present work has been implemented in a general purpose multiregion boundary element computer program and is capable of handling very large numbers of fiber elements in a given analysis. Several numerical examples are presented to validate the proposed method of fiber composite analysis and its applicability is demonstrated via practical engineering problems.     

Approximate dynamic programming-based approaches for input-output data-driven control of nonlinear processes

We propose two approximate dynamic programming (ADP)-based strategies for control of nonlinear processes using input-output data. In the first strategy, which we term ‘J-learning,’ one builds an empirical nonlinear model using closed-loop test data and performs dynamic programming with it to derive an improved control policy. In the second strategy, called ‘Q-learning,’ one tries to learn an improved control policy in a model-less manner. Compared to the conventional model predictive control approach, the new approach offers some practical advantages in using nonlinear empirical models for process control. Besides the potential reduction in the on-line computational burden, it offers a convenient way to control the degree of model extrapolation in the calculation of optimal control moves. One major difficulty associated with using an empirical model within the multi-step predictive control setting is that the model can be excessively extrapolated into regions of the state space where identification data were scarce or nonexistent, leading to performances far worse than predicted by the model. Within the proposed ADP-based strategies, this problem is handled by imposing a penalty term designed on the basis of local data distribution. A CSTR example is provided to illustrate the proposed approaches.     

AlphaQO: Robust Learned Query Optimizer

Recently, learned query optimizers typically driven by deep learning models have attracted wide attention as they can offer similar or even better performance than state-of-the-art commercial optimizers. A successful learning optimizer often relies on enough high-quality load queries as training data, and poor-quality training will lead to the query failure of learned query optimizers. In this paper, we propose a novel training framework AlphaQO for robust learned query optimizers based on Reinforcement Learning (RL), and the robustness of the optimizers can be improved by finding the bad queries in advance. AlphaQO is a loop system consisting of two main components, namely the query generator and the learned optimizer. A query generator aims at generating ``difficult'' queries (i.e., queries that the learned optimizer provides poor estimates). The learned optimizer will be trained using these generated queries, as well as providing feedback (in terms of numerical rewards) to the query generator for updates. If the generated queries are good, the query generator will get a high reward; otherwise, the query generator will get a low reward. The above process is performed iteratively, with the main goal that within a small budget, the learned optimizer can be trained and generalized well to a wide range of unseen queries. Extensive experiments show that AlphaQO can generate a relatively small number of queries and train a learned optimizer to outperform commercial optimizers. Moreover, learned optimizers require much fewer queries from AlphaQO than randomly generated queries for the quality training of the learned optimizer.     

Bond behaviour of corroded reinforcing steel bars in concrete

The effect of steel corrosion on bond between steel bars and the surrounding concrete was investigated for different corrosion levels. Both pullout tests and finite element analysis were used and the results from the two were compared. An electrolyte corrosion technique was used to accelerate steel corrosion. For confined deformed bars, a medium level (around 4%) of corrosion had no substantial influence on the bond strength, but substantial reduction in bond took place when corrosion increased thereafter to a higher level of around 6%. It is demonstrated that the confinement supplied an effective way to counteract bond loss for corroded steel bars of a medium (around 4% to 6%) corrosion level. The results of finite element analyses, where it was assumed that rust behaved like a granular material, showed a reasonably good agreement with the experiments regarding bond strength and bond stiffness.     

On integral generalized policy iteration for continuous-time linear quadratic regulations

This paper mathematically analyzes the integral generalized policy iteration (I-GPI) algorithms applied to a class of continuous-time linear quadratic regulation (LQR) problems with the unknown system matrix A$A$. GPI is the general idea of interacting policy evaluation and policy improvement steps of policy iteration (PI), for computing the optimal policy. We first introduce the update horizon ?$?$, and then show that (i) all of the I-GPI methods with the same ?$?$ can be considered equivalent and that (ii) the value function approximated in the policy evaluation step monotonically converges to the exact one as ?→∞$?\to \infty$. This reveals the relation between the computational complexity and the update (or time) horizon of I-GPI as well as between I-PI and I-GPI in the limit ?→∞$?\to \infty$. We also provide and discuss two modes of convergence of I-GPI; I-GPI behaves like PI in one mode, and in the other mode, it performs like value iteration for discrete-time LQR and infinitesimal GPI (?→0$?\to 0$). From these results, a new classification of the integral reinforcement learning is formed with respect to ?$?$. Two matrix inequality conditions for stability, the region of local monotone convergence, and data-driven (adaptive) implementation methods are also provided with detailed discussion. Numerical simulations are carried out for verification and further investigations.     

Data-based approximate policy iteration for affine nonlinear continuous-time optimal control design

This paper addresses the model-free nonlinear optimal control problem based on data by introducing the reinforcement learning (RL) technique. It is known that the nonlinear optimal control problem relies on the solution of the Hamilton–Jacobi–Bellman (HJB) equation, which is a nonlinear partial differential equation that is generally impossible to be solved analytically. Even worse, most practical systems are too complicated to establish an accurate mathematical model. To overcome these difficulties, we propose a data-based approximate policy iteration (API) method by using real system data rather than a system model. Firstly, a model-free policy iteration algorithm is derived and its convergence is proved. The implementation of the algorithm is based on the actor–critic structure, where actor and critic neural networks (NNs) are employed to approximate the control policy and cost function, respectively. To update the weights of actor and critic NNs, a least-square approach is developed based on the method of weighted residuals. The data-based API is an off-policy RL method, where the “exploration” is improved by arbitrarily sampling data on the state and input domain. Finally, we test the data-based API control design method on a simple nonlinear system, and further apply it to a rotational/translational actuator system. The simulation results demonstrate the effectiveness of the proposed method.     

Dyeing of natural and synthetic textiles in supercritical carbon dioxide with disperse reactive dyes

Polyester, nylon, silk and wool were dyed with disperse reactive dyes in supercritical carbon dioxide (scCO₂). The dyes were substituted with either vinylsulphone or dichlorotriazine reactive groups. Since earlier research showed that water, distributed over the scCO₂ and the textile, increased the colouration, experiments were done with the vinylsulphone dye with varying amounts of water in the dyeing vessel, to investigate if there is an optimum water concentration. The amounts were such, that no liquid water was present. The maximum colouration was obtained when both the scCO₂ and the textiles were saturated with water. At the saturation point, deep colours were obtained with the vinylsulphone dye for polyester, nylon, silk and wool, with fixation percentages between 70 and 92% when the dyeing time was 2 h. The positive effect of water was due to its ability to swell fibres or due to an effect of water on the reactivity of the dye–fibre system. Also the dichlorotriazine dye showed more colouration when the scCO₂ was moist. With this dye, experiments were conducted in water-saturated scCO₂, varying the pressure from 225 to 278 bar and the temperature from 100 to 116 °C. The colouration of polyester increased with pressure, the results for silk and wool were not sensitive to pressure. Increasing the temperature had no influence on the dyeing of polyester, silk and wool. The fixations on polyester, silk and wool, being between 71 and 97%, were also independent of pressure and temperature.