Precision modeling of form errors for cylindricity evaluation using genetic algorithms

Numerical optimization of heat transfer enhancement in a wavy channel using nanofluids

A heuristic approach is presented in this paper for efficient evaluation of exact straightness and flatness values using a hybrid approach combining genetic algorithms (GAs) with the geometric characterization method. The procedure of the hybrid approach begins with the evolution of GAs, and then the geometric characterization method, based on minimum zone criteria, is inserted into the process to improve the speed of convergence of the GAs. Numerical results indicate that the hybrid approach provides better accuracy and efficiency.

The main problem with all fractal compression implementation is execution time. Algorithms can spend hours to compress a single image. Most of the major variants of the standard algorithm for speeding up computation time have led to a bad-quality or a lower compression ratio. For example, the Fisher's [7] proposed classification pattern greatly accelerated the algorithm, but image quality was poor due to the search-space reduction imposed by the classification, which eleminates a lot of good solutions. By using genetic algorithms to address the problem, we optimize the domain blocks search. We explore all domain blocks present in the image but not in exhaustive way (like a standard algorithm) and without omitting any possible block (solution) as a classification pattern does. A genetic algorithm is the unique method for satisfying these constraints. And it is a way to do be a random search because the genetic one is directed by fitness selection, which produces optimal solutions. Our goal in this work is to use a genetic algorithm to solve the IFS inverse problem and to build a fractal compression algorithm based on the genetic optimization of a domain blocks search. we have also implemented standard Barnsley algorithm, the Y. Fisher based on classification, and the genetic compression algorithm with quadtree partitioning. A population of transformations was evolved for each range block, and the result is compared with the standard Barnsely algorithm and the Fisher algorithm = based classification. We deduced an optimal set of values for the best parameters combination, and we can also specify the best combination for each desired criteria: best compression ratio, best image quality, or quick compression process. By running many test images, we experimentally found the following set of optimal values of all the algorithm parameters that ensure compromise between execution time and solutions optimality: Population size = 100, Maximum generations = 20, Crossover rate = 0.7, Mutation rate = 0.1, RMS limit = 5, Decomposition error limit = 10, Flips and isometrics count = 8. In our proposed algorithm, results were much better than those obtained both vences and Rudomin [5] and Lankhorst [4] approaches.

To explore the essential relation between insulator leakage current and influence factors, the genetic algorithm (GA) combined with the Least square method (LSM) is used to obtain the relationship of the leakage current <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\boldsymbol{I_{\mathrm{h}}}$</tex> , the relative humidity density RH, the equivalent salt deposit density <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\boldsymbol{\rho_{\text{EDSS}}}$</tex> and the applied voltage <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\boldsymbol{U}$</tex> . Firstly, artificial pollution tests were carried out on single suspension insulator, waveforms of leakage current were recorded by a leakage current monitoring system under different operating conditions. Then the function relationship of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\boldsymbol{RH,\rho_{\text{EDSS}},U}$</tex> , and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\boldsymbol{I_{\mathrm{h}}}$</tex> was obtained by the GA and the LSM respectively. The result shows that when the leakage current is greater than 10 mA, the leakage current can be accurately estimated by GA, while when the leakage current is less than 10 mA, the LSM can estimate the leakage current more quickly. Finally, a new method for estimating leakage current is obtained by combining the GA with the LSM. The research results provide a reference for the evaluation of leakage current pollution degree in engineering.

Chemical heuristics have been fundamental to the advancement of chemistry and materials science. These heuristics are typically established by scientists using knowledge and creativity to extract patterns from limited datasets. Machine learning offers opportunities to perfect this approach using computers and larger datasets. Here, we discuss the relationships between traditional heuristics and machine learning approaches. We show how traditional rules can be challenged by large-scale statistical assessment and how traditional concepts commonly used as features are feeding the machine learning techniques. We stress the waste involved in relearning chemical rules and the challenges in terms of data size requirements for purely data-driven approaches. Our view is that heuristic and machine learning approaches are at their best when they work together.

As an indispensable constituent of the premises of highly precious control of vertical takeoff and landing (VTOL) aircrafts, parameter identification has received an increasingly considerable attention from academic community and practitioners. In an effort to tackle the matter better, we herewith put forward a PID controlling particle swarm optimizer (PSO) which we call the proportional integral derivative (PID) controller inspired particle swarm optimizer (P idSO). It uses a novel evolutionary strategy whereby a specified PID controller is used to improve particles' local and global best positions information. Empirical experiments were conducted on both analytically unimodal and multimodal test functions. The experimental results demonstrate that PidSO features better search effectiveness and efficiency in solving most of the multimodal optimization problems when compared with other recent variants of PSOs, and its performance can be upgraded by adopting proper control law based controllers. Moreover, PidSO, together with least squares (LS) method and genetic algorithm (GA), is applied to the parameter estimation of the VTOL aircraft. In comparison with LS method and GA, PidSO is a more effective tool in estimating the parameters of the VTOL aircraft.

Differential Evolution (DE) algorithm is a new heuristic approach which has been proposed particulary for numeric optimization problems. It is a population based algorithm like genetic algorithms using the similar operators; crossover, mutation and selection. In this work, DE algorithm has been applied to the design of fixed point digital Finite Impuls Response (FIR) filters and its performance has been compared to that of Genetic Algorithm (GA) and Least Squares Algorithm (LSQ).

A study on flowshop scheduling problem combining Taguchi experimental design and genetic algorithm

In this study, an artificial intelligence technique is proposed to be implemented in the parameter tuning of a PVD process. Due to its previous adaptation in similar optimization problems, genetic algorithm (GA) is selected to optimize the parameter tuning of the RF magnetron sputtering process. The most optimized parameter combination obtained from GA’s optimization result is expected to produce the desirable zinc oxide (ZnO) thin film from the sputtering process. The parameters involved in this study were RF power, deposition time and substrate temperature. The algorithm was tested to optimize the 25 datasets of parameter combinations. The results from the computational experiment were then compared with the actual result from the laboratory experiment. Based on the comparison, GA had shown that the algorithm was reliable to optimize the parameter combination before the parameter tuning could be done to the RF magnetron sputtering machine. In order to verify the result of GA, the algorithm was also been compared to other well known optimization algorithms, which were, particle swarm optimization (PSO) and gravitational search algorithm (GSA). The results had shown that GA was reliable in solving this RF magnetron sputtering process parameter tuning problem. GA had shown better accuracy in the optimization based on the fitness evaluation.

The profile accuracy of arc grinding wheels directly affects the precision of complex optics during grinding. However, existing studies have yet to systematically elucidate the distinct effects of wheel profile errors across different frequency bands on workpiece form errors. In this study, the generation mechanism of workpiece form error during the arc-envelope grinding process is revealed, considering the influence of wheel profile error across different frequency bands through kinematic simulation. The mapping relationship between profile errors and form errors is further analyzed, showing that the mapping coefficient is governed by both amplitude and frequency. The effects of neglecting wheel profile errors on the compensation effectiveness are analyzed for the conventional form error reverse compensation methods, revealing that conventional approaches cannot achieve complete convergence of error profiles and magnitudes. To address this issue, what we believe to be a novel compensation method considering profile errors of arc grinding wheels is proposed. Experimental results demonstrate that the cylindrical workpiece form error was significantly reduced from 8.8 µm PV to 2.1 µm PV after only a single compensation iteration, achieving a 38.2% improvement in compensation accuracy compared to the conventional shape-correction compensation method after two iterations. This method effectively enhances the grinding precision of complex optics while reducing iteration cycles of grinding-measurement-compensation processes.

New approach for evaluating form errors of engineering surfaces

Instance selection plays an important role in improving scalability of data mining algorithms, but it can also be used to improve the quality of the data mining results. In this dissertation we present a new optimization-based approach for instance selection that uses a genetic algorithm (GA) to select a subset of instances to produce a simpler decision tree with acceptable accuracy. The resultant trees are likely to be easier to comprehend and interpret by the decision maker and hence more useful in practice. We present numerical results for several difficult test datasets that indicate that GA-based instance selection can often reduce the size of the decision tree by an order of magnitude while still maintaining good prediction accuracy. The results suggest that GA-based instance selection works best for low entropy datasets. With higher entropy, there will be less benefit from instance selection. A comparison between GA and other heuristic approaches such as Rmhc (Random Mutation Hill Climbing) and simple construction heuristic, indicates that GA is able to obtain a good solution with low computation cost even for some large datasets. One advantage of instance selection is that it is able to increase the average instances associated with the leaves of the decision trees to avoid overfitting, thus instance selection can be used as an effective alternative to prune decision trees. Finally, the analysis on the selected instances reveals that instance selection helps to reduce outliers, reduce missing values, and select the most useful instances for separating classes.

The article deals with the definition and estimation of the parameters of a nonlinear regression model using a genetic algorithm. Parameter estimation is a type of optimization problem that can be solved using stochastic algorithms, which presents the possibility of using such algorithms to fit nonlinear regression models. The trees data set, which displays the non-linear relationship between traits, was investigated. The nonlinear method of least squares (NLS) was used to find the coefficients of the regression equation using a genetic algorithm on the trees data set belonging to the spuRs library. The set contains 1200 observations and three variables: tree ID; age of the tree; volume of wood. Visual analysis revealed the presence of additional signs that affect the volume of wood. Only data for trees with a certain location are selected for further work. As a result, the sample has 12 observations and two variables. The dependence of the volume of wood (Vol) on the age of the tree (Age) was analyzed. Attention to tree locations made it possible to choose the Richards logistic function as a functional dependence, nonlinear both in terms of parameters and variables. A genetic algorithm was used to estimate the parameters of the Richards function. The result of the work of the genetic algorithm depends on how its parameters are configured. When implementing the genetic algorithm, the following parameters were set: population size; the maximum number of iterations after which the work of the genetic algorithm stops; the number of consecutive generations without any improvement in the value of the fitness function before stopping the algorithm. A random scheme was used for the selection operation. A mutation operation involves changing a randomly selected bit. The type of crossover operation is one-point. Coefficients of nonlinear models were calculated for randomly selected tree locations. The coefficient of determination R2 was used to assess the quality of the models.

With the development of measurement technology, the Flexible Measuring Arm (FMA) is widely used in quality test of automobile processing and industrial production. FMA is a kind of nonlinear system with many parameters. Low cost and efficient calibration method have become the focuses of attention. This article presents a fast calibration method for FMA based on an adaptive Genetic Algorithm (GA) just with several standard balls and a ball plate. It can greatly reduce the calibration cost than common external calibration method which needs high precision instruments and sensors. Firstly, the kinematic model of FMA is established by RPY theory. Secondly, the common GA is optimized and improved, and an adaptive mechanism is added to the algorithms which can realize the automatic adjustment of crossover and mutation operators. A Normalized Genetic Algorithm (NGA) with adaptive mechanism is proposed to complete the optimization calculation. It can improve the numbers of optimal individuals and the convergence speed. So, the search efficiency will be enhanced greatly. Finally, the Least square method (LSM), the General Genetic Algorithm (GGA), and the proposed NGA are respectively used to finish the calibration work. The compensation accuracy and the search efficiency with the above three different algorithms have been systematically analyzed. Experiment indicates that the performance of NGA is much better than LSM and GGA. The data also has proved that the LSM is suitable to complete optimization calculation for linear system. Its convergence stability is much poorer than NGA and GGA because of the ill-condition Jacobin matrix. GGA is easy to fall into local optimization because of the fixed operators. The proposed NGA obviously owns fast convergence speed, high accuracy and better stability than GGA. The position error is reduced from 3.17 to 0.5 mm after compensation with the proposed NGA. Its convergence rate is almost two time of GGA which applies constant genetic factors. The effectiveness and feasibility of proposed method are verified by experiment.

Rapid advances in the world of high-speed optical network and high-speed electronic routing has led to consideration of advanced models such as genetic algorithms (GAs). GAs have proved to be a practical and robust optimization and search tool for network design. However, a GA is not generally employed in traffic grooming over mesh networks. In this paper, we propose a new GA model to handle the traffic grooming problem in an all-optical WDM mesh network. The new model extends classical GAs with heuristic approach to support network cost optimization for combining multiple traffic streams into a single lightpath. In addition, to support our new GA model, we have developed a set of new investigation including position-based bit matrix encoding, genetic heuristic operators, and fitness evaluation function using the clustering method. The experimental results show that our GA model is superior to traditional heuristic approaches for the 5, 6 and 14-node benchmark networks.