Balancing exploration and exploitation in robust multiobjective electromagnetic design optimisation

Design optimisation of electromagnetic and electromechanical devices is usually aided by numerical simulations, such as the finite element method, which often carry high computational costs, especially if three-dimensional transient modelling is required. Thus in addition to the task of finding the global optimum, while avoiding local minima traps, there is the additional requirement of achieving the final solution efficiently with as few objective function evaluations as possible. With this in mind several surrogate modelling techniques have been developed to replace, under controlled environment, the computationally expensive accurate field modelling by fast approximate substitutes. This thesis looks at a particular technique known as kriging, which in other applications has been demonstrated to provide accurate representations, even of complicated responses, based on a limited set of observations whilst providing an error estimate of the predictions and hence increasing the confidence in the answer. In the iterative optimisation process the critical issue is where to position the next point for evaluation to find a sensible compromise between conflicting goals to explore the search space thoroughly but at the same time exploit information already available. This thesis proposes several novel algorithms based on reinforcement learning theory using the concept of rewards for balancing exploration and exploitation automatically and adaptively. The performance of these algorithms has been assessed comprehensively using carefully selected test functions and real engineering problems (taken from TEAM workshops) and compared with the results published in literature. The kriging approach has generally been found to outperform significantly other available methods. One of the practical limitations, however, was found to be large-scale multi-dimensional or multi-objective tasks because of the need to create special correlation matrices for the kriging predictions to work. Several techniques have been developed and implemented to alleviate such problems and control the memory space occupied by such matrices. Finally, in practical design problems, the issue of robustness of the design has to be considered – related to manufacturing tolerances, material variability, etc – which requires the designer not only to find the theoretical optimum but also assess its quality (sensitivity) within specified uncertainties of variables. Several strategies for evaluation of design robustness assisted by kriging modelling have been developed and implemented in combination with commercial electromagnetic design software.

[1]  Shiyou Yang,et al.  An improved Tabu search for the global optimizations of electromagnetic devices , 2001 .

[2]  Byung Chai Lee,et al.  Development of a simple and efficient method for robust optimization , 2002 .

[3]  Jan K. Sykulski,et al.  Strategies for balancing exploration and exploitation in electromagnetic optimisation , 2013 .

[4]  S. Brisset,et al.  A new tabu search method for optimization with continuous parameters , 2004, IEEE Transactions on Magnetics.

[5]  Wolfgang Ponweiser,et al.  Clustered multiple generalized expected improvement: A novel infill sampling criterion for surrogate models , 2008, 2008 IEEE Congress on Evolutionary Computation (IEEE World Congress on Computational Intelligence).

[6]  Minh-Trien Pham,et al.  Robust Global Optimization of Electromagnetic Devices With Uncertain Design Parameters: Comparison of the Worst Case Optimization Methods and Multiobjective Optimization Approach Using Gradient Index , 2013, IEEE Transactions on Magnetics.

[7]  Dimitris K. Tasoulis,et al.  Dynamic Multi-Armed Bandit with Covariates , 2008, European Conference on Artificial Intelligence.

[8]  N. Hu Tabu search method with random moves for globally optimal design , 1992 .

[9]  Shiyou Yang,et al.  An improved population-based incremental learning method for inverse problems , 2008, 2008 World Automation Congress.

[10]  Harold J. Kushner,et al.  A New Method of Locating the Maximum Point of an Arbitrary Multipeak Curve in the Presence of Noise , 1964 .

[11]  J. K. Sykulski,et al.  Considerations of accuracy and uncertainty with kriging surrogate models in single-objective electromagnetic design optimisation , 2007 .

[12]  Stephen Chen,et al.  Locust Swarms - A new multi-optima search technique , 2009, 2009 IEEE Congress on Evolutionary Computation.

[13]  Jan K. Sykulski,et al.  Adaptive Weighted Expected Improvement With Rewards Approach in Kriging Assisted Electromagnetic Design , 2013, IEEE Transactions on Magnetics.

[14]  M. Sasena,et al.  Exploration of Metamodeling Sampling Criteria for Constrained Global Optimization , 2002 .

[15]  C. Magele,et al.  Managing uncertainties in electromagnetic design problems with robust optimization , 2004, IEEE Transactions on Magnetics.

[16]  J. Sykulski,et al.  A Scalarizing One-Stage Algorithm for Efficient Multi-Objective Optimization , 2008, IEEE Transactions on Magnetics.

[17]  Charles E. Taylor Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence. Complex Adaptive Systems.John H. Holland , 1994 .

[18]  Donald R. Jones,et al.  Efficient Global Optimization of Expensive Black-Box Functions , 1998, J. Glob. Optim..

[19]  Alessandro Formisano,et al.  Tolerance Analysis of NMR Magnets , 2010, IEEE Transactions on Magnetics.

[20]  J. A. Ramírez,et al.  Multiobjective approaches for robust electromagnetic design , 2006, IEEE Transactions on Magnetics.

[21]  G. Spagnuolo,et al.  Robust Design of Electromagnetic Systems Based on Interval Taylor Extension Applied to a Multiquadric Performance Function , 2008, IEEE Transactions on Magnetics.

[22]  Thomas J. Santner,et al.  The Design and Analysis of Computer Experiments , 2003, Springer Series in Statistics.

[23]  Jan K. Sykulski,et al.  Exploration versus exploitation using kriging surrogate modelling in electromagnetic design , 2012 .

[24]  Andy J. Keane,et al.  On the Design of Optimization Strategies Based on Global Response Surface Approximation Models , 2005, J. Glob. Optim..

[25]  Richard S. Sutton,et al.  Introduction to Reinforcement Learning , 1998 .

[26]  C. D. Gelatt,et al.  Optimization by Simulated Annealing , 1983, Science.

[27]  Chang Seop Koh,et al.  Magnetic pole shape optimization of permanent magnet motor for reduction of cogging torque , 1997 .

[28]  G.I. Hawe,et al.  Balancing Exploration Exploitation using Kriging Surrogate Models in Electromagnetic Design Optimization , 2006, 2006 12th Biennial IEEE Conference on Electromagnetic Field Computation.

[29]  Mina Salim,et al.  A modified fast marriage in honey bee optimization algorithm , 2010, 2010 5th International Symposium on Telecommunications.

[30]  Margaret J. Robertson,et al.  Design and Analysis of Experiments , 2006, Handbook of statistics.

[31]  Hans-Paul Schwefel,et al.  Numerical Optimization of Computer Models , 1982 .

[32]  Piergiorgio Alotto,et al.  Optimisation of electromagnetic devices with uncertain parameters and tolerances in the design variables , 2001 .

[33]  G.L. Soares,et al.  Robust Multi-Objective TEAM 22 Problem: A Case Study of Uncertainties in Design Optimization , 2009, IEEE Transactions on Magnetics.

[34]  L. Lebensztajn,et al.  Kriging: a useful tool for electromagnetic device optimization , 2004, IEEE Transactions on Magnetics.

[35]  Archie C. Chapman,et al.  Automated Planning in Repeated Adversarial Games , 2010, UAI.

[36]  Nicholas R. Jennings,et al.  On-Line Adaptation of Exploration in the One-Armed Bandit with Covariates Problem , 2010, 2010 Ninth International Conference on Machine Learning and Applications.

[37]  Shiyou Yang,et al.  A modified tabu search method applied to inverse problems , 2010, Digests of the 2010 14th Biennial IEEE Conference on Electromagnetic Field Computation.

[38]  S.Y. Yang,et al.  An Adaptive Interpolating MLS Based Response Surface Model Applied to Design Optimizations of Electromagnetic Devices , 2007, IEEE Transactions on Magnetics.

[39]  Rong Pan,et al.  Robustness and reliability consideration in product design optimization under uncertainty , 2011, 2011 IEEE International Conference on Industrial Engineering and Engineering Management.