复杂条件下矿体开采方法及采场顶板维护研究

在复杂矿体的回采中,经常会遇到顶板冒落问题,作为复杂矿体的反倾节理矿体,其顶板管理更需加强,为了处理反倾节理矿体开采中遇到的各种问题和情况,有针对性地提出了一些建议和措施。     

Meso-scale numerical investigation on cracking of cover concrete induced by corrosion of reinforcing steel

Concrete cover cracking induced by corrosion of steel reinforcement is a major influencing factor for durability and serviceability of reinforced concrete structures. Here in this study, the influence of concrete meso-structure on the failure pattern of concrete cover is accounted for. The concrete is assumed to be a three-phase composite composed of aggregate, mortar matrix and the interfacial transition zone (ITZ). And a concrete random aggregate structure is established for the study on the mechanical behavior of radial corrosion expansion. In the present simulations, the plasticity damaged model is used to describe the mechanical behavior of the mortar matrix and the ITZ, and it is assumed that the corrosion of steel reinforcement is uniform. The cracking of concrete cover due to steel reinforcement corrosion is numerically simulated. The simulation results have a good agreement with the available test data and they are between the two analytical results. The failure patterns obtained from the macro-scale homogeneous model and the meso-scale heterogeneous model are compared. Furthermore, the influences of ratio of cover thickness and reinforcement diameter (i.e. c/d), the location of the steel reinforcement (i.e., placed at the middle and corner zones) and the concrete tensile strength on the steel corrosion rate when the concrete cover cracks are investigated. Finally, some useful conclusions are drawn.     

Tuning fuzzy PD and PI controllers using reinforcement learning

In this paper, we propose a new auto-tuning fuzzy PD and PI controllers using reinforcement Q-learning (QL) algorithm for SISO (single-input single-output) and TITO (two-input two-output) systems. We first, investigate the design parameters and settings of a typical class of Fuzzy PD (FPD) and Fuzzy PI (FPI) controllers: zero-order Takagi–Sugeno controllers with equidistant triangular membership functions for inputs, equidistant singleton membership functions for output, Larsen’s implication method, and average sum defuzzification method. Secondly, the analytical structures of these typical fuzzy PD and PI controllers are compared to their classical counterpart PD and PI controllers. Finally, the effectiveness of the proposed method is proven through simulation examples.     

Predicting investment behavior: An augmented reinforcement learning model

The goal of this paper is to augment the ordinal temporal-difference type (TD-type) reinforcement learning model in order to detect the most suitable learning model of the human decision-making process in financial investment tasks. The simplicity and robustness of the TD-type learning model is fascinating. However, the available evidence and our observation suggest the necessity of introducing the nonlinear effect in learning and the possibility that additional factors might play important roles in the investment decision-making process. To extend the ordinal TD-type learning model, we adopt a three-layered perceptron as the basis function and the hierarchical Bayesian method to calibrate the parameter values. The result of the predictive test suggests that the augmented TD-type learning model constructed in this paper can evade the overfitting and can predict people's investment behavior well as compared to other familiar learning models.     

Optimality and convergence of adaptive optimal control by reinforcement synthesis

Adaptive Optimal Control (AOC) by reinforcement synthesis is proposed to facilitate the application of optimal control theory in feedback controls. Reinforcement synthesis uses the critic–actor architecture of reinforcement learning to carry out sequential optimization. Optimality conditions for AOC are formulated using the discrete minimum principle. A proof of the convergence conditions for the reinforcement synthesis algorithm is presented. As the final time extends to infinity, the reinforcement synthesis algorithm is equivalent to the Dual Heuristic dynamic Programming (DHP) algorithm, a version of approximate dynamic programming. Thus, formulating DHP with the AOC approach has rigorous proofs of optimality and convergence. The efficacy of AOC by reinforcement synthesis is demonstrated by solving a linear quadratic regulator problem.     

斜交板桥横桥向位移原因分析及处治措施

针对一座大角度斜交的空心板桥梁出现的横桥向位移等病害，从力学分析的角度分析了该桥产生病害的主要原因。相关调查资料表明，大多数简易支座斜交板桥存在缓慢下滑和平面内转动的现象，其主要原因是横向下滑力和平面内不平衡力矩。据此提出了该桥的加固处理对策，并对今后进行斜交板桥设计和施工提出了建议，以期引起同行对此类型桥梁病害的重视。     

Individual differences in drivers’ cognitive processing of road safety messages

Using Gray and McNaughton's (2000) revised reinforcement sensitivity theory (r-RST), we examined the influence of personality on processing of words presented in gain-framed and loss-framed anti-speeding messages and how the processing biases associated with personality influenced message acceptance. The r-RST predicts that the nervous system regulates personality and that behaviour is dependent upon the activation of the behavioural activation system (BAS), activated by reward cues and the fight-flight-freeze system (FFFS), activated by punishment cues. According to r-RST, individuals differ in the sensitivities of their BAS and FFFS (i.e., weak to strong), which in turn leads to stable patterns of behaviour in the presence of rewards and punishments, respectively. It was hypothesised that individual differences in personality (i.e., strength of the BAS and the FFFS) would influence the degree of both message processing (as measured by reaction time to previously viewed message words) and message acceptance (measured three ways by perceived message effectiveness, behavioural intentions, and attitudes). Specifically, it was anticipated that, individuals with a stronger BAS would process the words presented in the gain-frame messages faster than those with a weaker BAS and individuals with a stronger FFFS would process the words presented in the loss-frame messages faster than those with a weaker FFFS. Further, it was expected that greater processing (faster reaction times) would be associated with greater acceptance for that message. Driver licence holding students (N = 108) were recruited to view one of four anti-speeding messages (i.e., social gain-frame, social loss-frame, physical gain-frame, and physical loss-frame). A computerised lexical decision task assessed participants’ subsequent reaction times to message words, as an indicator of the extent of processing of the previously viewed message. Self-report measures assessed personality and the three message acceptance measures. As predicted, the degree of initial processing of the content of the social gain-framed message mediated the relationship between the reward sensitive trait and message effectiveness. Initial processing of the physical loss-framed message partially mediated the relationship between the punishment sensitive trait and both message effectiveness and behavioural intention ratings. These results show that reward sensitivity and punishment sensitivity traits influence cognitive processing of gain-framed and loss-framed message content, respectively, and subsequently, message effectiveness and behavioural intention ratings. Specifically, a range of road safety messages (i.e., gain-frame and loss-frame messages) could be designed which align with the processing biases associated with personality and which would target those individuals who are sensitive to rewards and those who are sensitive to punishments.     

Development of Mechanical Anchor for CFRP Tendons Using Integrated Sleeve

A durable and very efficient external strengthening system is achieved if steel tendons for posttensioning applications can be replaced with carbon fiber-reinforced polymer (CFRP) tendons, and if reliable anchorage systems are developed. This paper presents a newly developed and simple-to-use, two-piece wedge anchorage for CFRP tendons with an integrated sleeve and a differential angle between barrel and wedge sections. Three longitudinal slits are cut into the one-piece wedge, with one slit open and the other two stopping 1 mm from the inner wedge hole. The integrated sleeve holds the wedge’s sections together during presetting and loading, resulting in a circumferential confined gripping of the CFRP tendon and optimized surface friction area. Therefore, the one-piece wedge differs from conventional wedge systems, where the wedges act separately with adjacent spaces, wedging the separate tendon sleeve in the longitudinal direction. Evaluation of the failure modes during testing was one of the main keys in achieving an increasingly better performance of the anchorage until the final anchorage was developed. The obtained failure modes are therefore described to enlighten the importance of addressing them when testing. The test setup used and measured behavior are described further together with the loading procedure. The anchorage reached the full capacity of the CFRP tendon and was seen to ensure a stable load of fracture.     

预应力锚索对均质岩体的加固效应模拟试验研究

本文采用相似材料模型试验方法研究了单根预应力锚索对均质岩体的加固效应，给出了在三种不同长度的锚索作用下，岩体中受力变形规律，分析了锚索长度、预应力数值等对加固效应的影响，并与有限元计算结果作了对比。     

联系发现：一种新的数据挖掘方法综述

联系发现是数据挖掘中较新的研究领域。联系发现是一种对海量数据进行挖掘，找出其中潜在模式，抽取有用知识并发现隐藏联系的技术。本文首先综述了联系发现的概念、范围、特点和难点等，详细介绍了联系发现的几种主要方法：无监督的联系发现方法（新颖联系发现）、使用归纳逻辑程序技术挖掘关联数据的联系发现方法、多假设反演推理的联系发现方法、基于相关分析的联系发现方法以及KOJAK组队探测器，讨论了联系发现系统性能评估的方法与联系发现的置信区间度量方法，并简要描述了联系发现的一个具体应用的实例一——证据抽取和联系发现研究计划（EELD），最后探讨了目前联系发现研究中出现的问题及未来发展趋势。