Expert-driven genetic algorithms for simulating evaluation functions

In this paper we demonstrate how genetic algorithms can be used to reverse engineer an evaluation function’s parameters for computer chess. Our results show that using an appropriate expert (or mentor), we can evolve a program that is on par with top tournament-playing chess programs, outperforming a two-time World Computer Chess Champion. This performance gain is achieved by evolving a program that mimics the behavior of a superior expert. The resulting evaluation function of the evolved program consists of a much smaller number of parameters than the expert’s. The extended experimental results provided in this paper include a report on our successful participation in the 2008 World Computer Chess Championship. In principle, our expert-driven approach could be used in a wide range of problems for which appropriate experts are available.     

Performance evaluation of multilayer perceptron,radial basis function,fuzzy inference system,and an adaptively tuned fuzzy wavelet neural network in parameter prediction of multiphase flow measurement instrumentation

Application of predictive models in industrial multiphase flow metering has attracted an increasing attention recently. Void fraction (VF), water–liquid ratio (WLR), and flow regime are key parameters, measured by oil/water/gas multiphase flow metres (MPFM) in petroleum industry. Artificial neural networks and fuzzy inference systems (FIS) are reliable and efficient computational models, which can be simply implemented on microprocessors of MPFMs, having the advantages of trainability, adaptability, and capability to model non‐linear functions. In this paper, a wavelet‐based adaptive neuro‐FIS (WANFIS) is introduced and validated by the prediction of multiphase flow measurement critical parameters including flow regime, VF, and WLR. The performance of the proposed WANFIS model is then compared with multilayer perceptron (MLP), radial basis function (RBF) network, and an FIS trained by fuzzy c‐means and a subtractive clustering method in the prediction of flow parameters in a customized designed structure of oil/water/gas MPFM. Structural parameters of all predictive models are first optimized to yield the most efficient structure for the available dataset. Comparison is then made between the optimized predictive models, in terms of mean squared error of parameter prediction, computation time, and repeatability of the prediction process. According to the obtained results, MLP model using Levenberg–Marquardt training algorithm and WANFIS model using gradient‐based back propagation dynamical iterative learning algorithm are the most efficient models, which give the best performance compared with other used models. All predictive models can predict the flow regime with 100% accuracy, whereas the highest inaccuracy is related to the prediction of WLR. The results of this study can be used to select and develop the most appropriate predictive model applicable in predicting and identifying flow measurement parameters in industrial MPFMs.     

Implicit branching and parameterized partial cover problems

Covering problems are fundamental classical problems in optimization, computer science and complexity theory. Typically an input to these problems is a family of sets over a finite universe and the goal is to cover the elements of the universe with as few sets of the family as possible. The variations of covering problems include well-known problems like Set Cover, Vertex Cover, Dominating Set and Facility Location to name a few. Recently there has been a lot of study on partial covering problems, a natural generalization of covering problems. Here, the goal is not to cover all the elements but to cover the specified number of elements with the minimum number of sets. In this paper we study partial covering problems in graphs in the realm of parameterized complexity. Classical (non-partial) version of all these problems has been intensively studied in planar graphs and in graphs excluding a fixed graph H as a minor. However, the techniques developed for parameterized version of non-partial covering problems cannot be applied directly to their partial counterparts. The approach we use, to show that various partial covering problems are fixed parameter tractable on planar graphs, graphs of bounded local treewidth and graph excluding some graph as a minor, is quite different from previously known techniques. The main idea behind our approach is the concept of implicit branching. We find implicit branching technique to be interesting on its own and believe that it can be used for some other problems.     

Taguchi统计学分析方法优化Al 7068-TiC纳米复合材料的凝固行为

采用机械合金化和热压工艺制备高强Al 7068?5%TiC(质量分数)纳米复合材料。基于致密化的重要性和晶粒生长的影响,以获得较高抗压强度为目标,采用Taguchi统计法对制备Al 7068?5%TiC块体纳米复合材料的热压条件进行优化。结果表明：在500 MPa和385°C下热压30 min能获得抗压强度为938 MPa、硬度为HV 265的Al 7068?TiC纳米复合材料。此外,方差分析结果表明,外加压力是影响纳米复合材料热压过程最关键的因素。各因素对纳米复合材料热压过程影响贡献率为外加压力(61.3%)、热压温度(29.53%)和热压时间(4.49%)。     

Automated inspection in robotic additive manufacturing using deep learning for layer deformation detection

Journal of Intelligent Manufacturing - In this paper, an automated layer defect detection system for construction 3D printing is proposed. Initially, a step-by-step procedure is implemented to...     

Effect of Shot Peening Operation on the Microstructure and Wear Behavior of AZ31 Magnesium Alloy

Protection of Metals and Physical Chemistry of Surfaces - Shot peening is a treatment used to increase surface hardness and wear resistance. In this study, the effect of shot peening on the...     

A neural approach for estimation of per capita electricity consumption due to age and income

Electricity consumption is influenced by number of adults and children and their relationship at household level. Household income also plays a critical role on expenditure on electricity. Accordingly, this article presents a joint probability model of electricity demand based on occupants’ age grades and household income levels. A bottom-up strategy is developed using a micro level database of 70 Australian households. A neural regression-generalization technique is devised to estimate electricity demand using back-propagation and cognitive mapping. The aggregated result is then validated against 2012 Australian national census. Accordingly, the model is improved based on a top-down review. The results show per capita electricity demand by adult and child at 0.408 kW (69 kWh/week) and 0.226 kW (38 kWh/week), respectively. The equivalent dollar values are $13.6/week and $7.6/week in 2012. At macro level, the model reveals per capita demand by all individuals at 0.324 kW (54.35 kWh/week) equivalent to dollar value of $10.87/week, across Australia. The results also show higher percentage of per capita demand for adults in high and medium income classes, and the otherwise for low income class. Ratio of child’s demand over adult’s demand is highest among the low income households, and lowset among the middle income households, while best balance between adult and child per capita demand belongs to the high income.

     

Modeling and Optimization of SE and SI of Copper Flotation via Hybrid GA–ANN

In the present paper, a new attitude has been proposed for optimization of the separation efficiency (SE) and the Gaudin’s selectivity index (SI) in a flotation process by Hybrid artificial neural network (ANN) and genetic algorithm (GA). The chemical reagent’s dosage (collector, frother and fuel oil), pH, solid percentage, feed rate, Cu, Mo, and Fe grades in the flotation feed were selected as input variables and the SE-Cu and SI-Mo and SI-Fe were selected as output ones. Multilayer NN with back propagation (BP) algorithm was trained by the standard Bayesian regulation algorithm in which the validation data set did not required to be apart from its training. This algorithm with four-layer was used to relate output and input variables. Employment of Hybrid GA–ANN method resulted in significant improvement on GA fitnesses, as SE-Cu = 88, SI-Mo = 4.47 and SI of Fe = 12.85 were achieved. The input parameters corresponding to the fitnesses were as follows: pH = 12.25; the grade of Cu = 0.55%, Mo = 0.04% and Fe = 5.53%; the collector, frother and fuel–oil concentrations being 16.55, 15.54 and 2.71 (g/ton), respectively; the solid percentage was 25.84% and feed rate was 38,380 ton/day. The best fitness of GA was obtained after 10 generations by MSE value of 2.23.     

Document Image Coding for Processing and Retrieval

Document images belong to a unique class of images where the information is embedded in the language represented by a series of symbols on the page rather than in the visual objects themselves. Since these symbols tend to appear repeatedly, a domain-specific image coding strategy can be designed to facilitate enhanced compression and retrieval. In this paper we describe a coding methodology that not only exploits component-level redundancy to reduce code length but also supports efficient data access. The approach identifies and organizes symbol patterns which appear repeatedly. Similar components are represented by a single prototype stored in a library and the location of each component instance is coded along with the residual between it and its prototype. A representation is built which provides a natural information index allowing access to individual components. Compression results are competitive and compressed-domain access is superior to competing methods. Applications to network-related problems have been considered, and show promising results.