Evolving flexible beta basis function neural tree using extended genetic programming & Hybrid Artificial Bee Colony

In this paper, a new hybrid learning algorithm is introduced to evolve the flexible beta basis function neural tree (FBBFNT). The structure is developed using the Extended Genetic Programming (EGP) and the Beta parameters and connected weights are optimized by the Hybrid Artificial Bee Colony algorithm. This hybridization is essentially based on replacing the random Artificial Bee Colony (ABC) position with the guided Opposite-based Particle Swarm Optimization (OPSO) position. Such modification can minimize the delay which might be lead by the random position, in reaching the global solution. The performance of the proposed model is evaluated for benchmark problems drawn from time series prediction area and is compared with those of related methods.     

一种基于敏感分量融合的液压泵退化特征提取方法

液压泵退化特征提取是实现故障预测的关键环节。在液压泵性能退化过程中，其振动信号复杂度高、非线性强，难以有效地提取退化特征，为此，本文提出一种基于敏感分量融合的退化特征提取方法。采用改进局部特征尺度分解（ILCD）方法对振动信号进行分解，并利用贝叶斯信息准则与所构建的敏感因子，对内禀尺度分量进行筛选，以减少干扰分量的影响，得到敏感分量，有效抓取特征信息；在此基础上，引入离散余弦变换代替传统复合谱分析中的傅里叶变换，提出离散余弦变换-复合谱（DCS）算法，以解决信息遗漏问题，并利用DCS对敏感分量进行融合，提取复合谱熵作为退化特征，以提高对退化过程的表征能力；最后，通过对液压泵性能退化试验实测振动信号的应用分析，验证了该方法的有效性。     

Prediction of heavy oil viscosity using a radial basis function neural network

Heavy oil and extra heavy oil resources comprise about 75% of petroleum resources. The most important characteristic of heavy oils is their viscosity. Consequently, to extract and prepare these kinds of crude oil for use, great emphasis should be put on viscosity. The present study highlights the application of intelligent model named radial basis function (RBF) network optimized by genetic algorithm for estimation of diluted heavy oil viscosity in presence on kerosene. The input parameters of model were temperature and mass fraction of kerosene. The output of model was viscosity of heavy oil. Genetic algorithm was utilized to optimize the tuning parameters of RBF model. The outcomes of this study showed that the proposed model is accurate in estimation of target data.     

Effect of in-situ processing parameters on microstructure and mechanical properties of TiC particulate reinforced Al–4.5Cu alloy MMC fabricated by stir-casting technique – Optimization using grey based differential evolution algorithm

Fabrication of metal matrix composite materials by using the traditional casting process is a very promising task of developing near net shape products at a very low cost. Among these casting processes, the stir-casting technique has fine microstructures because of fast cooling, less porosity and good quality bonding among particle and the base matrix alloy. The major aim of this present paper is to fabricate titanium carbide reinforced metal matrix composite (MMC) by using micron sized activated charcoal powder, to get react with titanium material for producing TiC intermetallic particles in the Al–4.5%Cu matrix by in-situ technique so as to produce Al–4.5%Cu–5%TiC MMC, with the intention of finding the optimum processing parameter i.e. pouring temperature, stirring speed in liquid melt-stirring process, reaction time with the aim of achieving better microstructure and mechanical properties, i.e. micro-hardness, toughness and ultimate tensile strength by using an efficient grey based differential evolution algorithm. Mechanical properties of optimum processing condition are verified by doing confirmation tests. ANOVA analysis is performed to analyze the role of each factor on multiple performance characteristics and it was found that stirrer rotational speed and reaction time is playing a key role in affecting the mechanical property. Additionally, to predict the response parameters regression equations were developed and the validity of regression model is tested by plotting scatter diagram. Finally microstructure study has been performed for the optimal sample by using optical microscope.     

A global shooting algorithm for the facility location and capacity acquisition problem on a line with dense demand

This paper describes the development of an exact allocation-based solution algorithm for the facility location and capacity acquisition problem (LCAP) on a line with dense demand data. Initially, the n-facility problem on a line is studied and formulated as a dynamic programming model in the allocation decision space. Next, we cast this dynamic programming formulation as a two-point boundary value problem and provide conditions for the existence and uniqueness of solutions. We derive sufficient conditions for non-empty service regions and necessary conditions for interior facility locations. We develop an efficient exact shooting algorithm to solve the problem as an initial value problem and illustrate on an example. A computational study is conducted to study the effect of demand density and other problem parameters on the solutions.     

Thermodynamic analysis and optimization of a geothermal Kalina cycle system using Artificial Bee Colony algorithm

In this paper, thermodynamic analysis is carried out for a geothermal Kalina cycle employed in Husavic power plant. Afterwards, the optimum operating conditions in which the cycle is at its best performance are calculated. In order to reach the optimum thermal and exergy efficiencies of the cycle, Artificial Bee Colony (ABC) algorithm, a new powerful multi-objective and multi-modal optimization algorithm, is conducted. Regarding the mechanism of ABC algorithm, convergence speed and precision of solutions have been remarkably improved when compared to those of GA, PSO and DE algorithms. Such a relative improvement is indicated by a limit parameter and declining probability of premature convergence. In this research, exergy efficiency including chemical and physical exergies and thermal efficiency are chosen as the objective functions of ABC algorithm where optimum values of the efficiencies for the Kalina cycle are found to be 48.18 and 20.36%, respectively, while the empirical thermal efficiency of the cycle is about 14%. At the optimum thermal and exergy efficiencies, total exergy destruction rates are respectively 4.17 and 3.48 MW. Finally, effects of the separator inlet pressure, temperature, basic ammonia mass fraction and mass flow rate on the first and second law efficiencies are investigated.     

基于遗传算法的大厂矿区地震定位研究

基于遗传算法对大厂矿区的地震进行重新定位。为验证该方法在大厂矿区的应用效果,首先对已知的5次人工爆破进行定位,得出平均定位水平误差在146m范围内(矿区台网定位水平误差为500m),进而对大厂矿区的81次ML1.0级以上地震进行重新定位,结果表明:大厂矿震主要呈北西向展布,且主要集中在大厂断裂右侧;较大地震均与大厂断裂呈平行状分布,推断其附近可能有隐伏断层存在;大厂矿区地震主要受北西向断层的影响,受北东向断层的影响不大。定位后震源平均深度为2.8km,最大深度6.5km,最小深度1.5km。     

An integrated supply chain production–distribution planning with stochastic demands

In this paper a three echelon supply chain with multiple distribution centers, production sites and suppliers is modeled. For this supply chain several commodities with defined items is produced. The model is categorized as a capacitated facility location model. The hierarchical approach is used to modeling and based on decision types the model is divided into two levels. The solution approach is based on the Lagrangian Relaxation approach, improved by an efficient heuristic to solve complex sub-problems. Computational results indicated that the proposed method yields high-quality solutions within a reasonable computational time for various real size problems.     

Identification of time‐delay Markov jump autoregressive exogenous systems with expectation‐maximization algorithm

This paper is concerned with the identification problem of the Markov jump autoregressive exogenous system with an unknown time delay. The considered problem is solved using the expectation‐maximization algorithm, which estimates the parameters of local models, Markov transition probabilities, and time delay simultaneously. A numerical example and a simulated continuous fermentation reactor example are given to illustrate the capability of the proposed method. It shows that the influences of time delay during identification can be overcome by the proposed algorithm effectively.     

From evaluation to learning: Some aspects of designing a cyber-university

Research is described on a system for web-assisted education and how it is used to deliver on-line drill questions, automatically suited to individual students. The system can store and display all of the various pieces of information used in a class-room (slides, examples, handouts, drill items) and give individualized drills to participating students. The system is built on the basic theme that it is for learning rather than evaluation.Experimental results shown here imply that both the item database and the item allocation methods are important and examples are given on how these need to be tuned for each course. Different item allocation methods are discussed and a method is proposed for comparing several such schemes. It is shown that students improve their knowledge while using the system. Classical statistical models which do not include learning, but are designed for mere evaluation, are therefore not applicable.A corollary of the openness and emphasis on learning is that the student is permitted to continue requesting drill items until the system reports a grade which is satisfactory to the student. An obvious resulting challenge is how such a grade should be computed so as to reflect actual knowledge at the time of computation, entice the student to continue and simultaneously be a clear indication for the student. To name a few methods, a grade can in principle be computed based on all available answers on a topic, on the last few answers or on answers up to a given number of attempts, but all of these have obvious problems.