Myung-Won Suh

Pareto‐based continuous evolutionary algorithms for multiobjective optimization

In an attempt to solve multiobjective optimization problems, many traditional methods scalarize an objective vector into a single objective by a weight vector. In these cases, the obtained solution is highly sensitive to the weight vector used in the scalarization process and demands a user to have knowledge about the underlying problem. Moreover, in solving multiobjective problems, designers may be interested in a set of Pareto‐optimal points, instead of a single point. In this paper, Pareto‐based Continuous Evolutionary Algorithms for Multiobjective Optimization problems having continuous search space are introduced. These algorithms are based on Continuous Evolutionary Algorithms, which were developed by the authors to solve single‐objective optimization problems with a continuous function and continuous search space efficiently. For multiobjective optimization, a progressive reproduction operator and a niche‐formation method for fitness sharing and a storing process for elitism are implemented in the algorithm. The operator and the niche formulation allow the solution set to be distributed widely over the Pareto‐optimal tradeoff surface. Finally, the validity of this method has been demonstrated through some numerical examples.

Hardware-in-the-loop simulation for ABS based on PC

The prevalence of microprocessor-based controllers in automotive systems has greatly increased the need for tools which can be used to validate and tune the control systems over their full range of operation. The objective of this paper is to develop a real time simulator for an anti-lock brake system (ABS) based on the methodology of hardware-in-the-loop simulation using a personal computer. By use of this simulator, the analysis of a commercial electronic control unit as well as the validation of the developed control logic for ABS was performed successfully. The idea of the simulator can be applied to the development of more advanced control systems, such as traction control system, vehicle dynamic control system and so forth.

Crack identification using evolutionary algorithms in parallel computing environment

Abstract It is well known that a crack has an important effect on the dynamic behavior of a structure. This effect depends mainly on the location and depth of the crack. To identify the location and depth of a crack in a structure, classical optimization technique was adopted by previous researchers. That technique overcame the difficulty of finding the intersection point of the superposed contours that correspond to the eigenfrequency caused by the crack presence. However, it is hard to select the trial solution initially for optimization because the defined objective function has heavily local minima. A method is presented in this paper which uses a continuous evolutionary algorithm (CEA), which is suitable for solving inverse problems and implemented on PC clusters to shorten calculation time. With finite element model of the structure to calculate eigenfrequencies, it is possible to formulate the inverse problem in optimization format. CEAs are used to identify the crack location and depth minimizing the difference from the measured frequencies. We have tried this new idea on beam structures and the results are promising with high parallel efficiency over about 91%.

A Study on Multiobjective Optimization Technique for Inverse and Crack Identification Problems

It is well known that a crack has an important effect on the dynamic behavior of a structure. This effect depends mainly on the location and depth of the crack. To identify the location and depth of a crack in a structure, a variety of techniques were developed. However, these approaches are not enough to identify the crack profiles because the crack identification problem is heavily ill-posed (the existence, uniqueness and stability of solutions cannot be assured). A method is presented in this article which uses Pareto-based Continuous Evolutionary Algorithms for Multiobjective Optimization (MOPCEAs), which are deriving solutions without weighting factors such as a regularization parameter and can work effectively for problems of concern. With finite element model of the structure to calculate eigenfrequencies, it is possible to formulate the inverse problem in multiobjective optimization format. MOPCEAs are used to identify the crack location and depth by minimizing three objective functions, which are defined by the difference of the calculated eigenfrequency and the measured/reference one, respectively. We have tried this new idea on beam structures and the results are promising.

A study on the factors that improve the velocity perception of a virtual reality‐based vehicle simulator

The Virtual Reality (VR) system of a real‐time VR‐linked vehicle simulator that was used in this study provides visual information and sound effects to participants. The VR system of a VR‐linked vehicle simulator should provide a perceived velocity similar with the perceived velocity in actual driving. To achieve these goals, modeling and rendering methods that offer an improved performance for complex VR applications, such as the 3D road model, were implemented and evaluated. We also evaluated the influences of graphic and engine sound effects on the driver and analyzed each result according to a drivers viewpoint, the dot densities of road texture provided, the lateral distance between a virtual driver and environmental objects, and the engine sound. Each factor was individually analyzed through an experiment that evaluated the influence of visual images or sound effects in the vehicle simulator. Through the experimental evaluation, the research results could be used for improving the effectiveness of VR‐based vehicle simulators.

Model-matching control applied to longitudinal and lateral automated driving

Although there has been substantial research on longitudinal and lateral controllers for an automated driving system, stability issues with respect to the effect of uncertainties due to parameter variations (e.g. in the vehicle mass and the cornering stiffness) and disturbances or perturbations to the vehicle system (e.g. in the road gradient and the wind) still need to be addressed. Thus, an automated driving system needs to be made robust to those influences. For this purpose, the model-matching control applied to longitudinal and lateral automated driving is investigated by vehicle dynamics simulation. The design of the model-matching controller is obtained by using the characteristics of a two-degree-of-freedom controller. It can make various characteristics of automated driving vehicles equivalent to a specific transfer function, which is suggested as the reference model. The vehicle dynamics models including the model-matching controller are constructed for computer simulation. Then, simple examples of open-loop simulation and closed-loop simulation are solved to check the robustness of the model-matching controller. As a practical example, an automated driving system is adopted. It is proved that the model-matching control is effective and adequate for uncertainties due to parameter variation and disturbances or perturbations to the vehicle system, which are shown in the responses of the changes in the vehicle mass, the road gradient and the cornering stiffness.

Development of Optimization Algorithm for Unconstrained Problems Using the Sequential Design of Experiments and Artificial Neural Network

The conventional approximate optimization method, which uses the statistical design of experiments(DOE) and response surface method(RSM), can derive an approximated optimum results through the iterative process by a trial and error. The quality of results depends seriously on the factors and levels assigned by a designer. The purpose of this study is to propose a new technique, which is called a sequential design of experiments(SDOE), to reduce a trial and error procedure and to find an appropriate condition for using artificial neural network(ANN) systematically. An appropriate condition is determined from the iterative process based on the analysis of means. With this new technique and ANN, it is possible to find an optimum design accurately and efficiently. The suggested algorithm has been applied to various mathematical examples and a structural problem.

A Study on Maintenance Reliability Allocation of Urban Transit Brake System Using Hybrid Neuro-Genetic Technique

For reasonable establishing of maintenance strategies, safety security and cost limitation must be considered at the same time. In this paper, the concept of system reliability introduces and optimizes as the key of reasonable maintenance strategies. This study aims at optimizing component’s reliability that satisfies the target reliability of brake system in the urban transit. First of all, constructed reliability evaluation system is used to predict and analyze reliability. This data is used for the optimization. To identify component reliability in a system, a method is presented in this paper which uses hybrid neuro-genetic technique. Feed-forward multi-layer neural networks trained by back propagation are used to find out the relationship between component reliability (input) and system reliability (output) of a structural system. The inverse problem can be formulated by using neural network. Genetic algorithm is used to find the minimum square error. Finally, this paper presents reasonable maintenance cycle of urban transit brake system by using optimal system reliability.

Development of the GPS Simulator for Driving Simulators

Driving simulators are useful tools not only to test the components of future cars but also to evaluate the telematics service and HMI (human-machine interface). However driving simulators cannot be implemented to test and evaluate the telematics service system because the GPS (Global Positioning System) which contains basic functional support for the telematics module dose not work in the VR (virtual reality) environment. This paper presents a method to implement telematics service to a driving simulator by developing the GPS simulator which is able to emulate GPS satellite signals consist of NMEA-0183 protocol and RS232C communication standards. It is expected that the driving simulator with the GPS simulator can be used to study HMI and human-factor evaluations of the commercial telematics system to realize the HiLES (human-in-the-loop evaluation system)

Development of Urban Driving Cycle for Performance Evaluation of Electric Vehicles Part II : Verification of Driving Cycle

Abstract : Recently, due to various environmental problems such as global warming, increases of international oil prices, exhaustion of resource, a paradigm of world automobile market is rapidly changing from conventional vehicles using internal combustion engine to eco-friendly vehicles using electric power such as EV, HEV, PHEV and FCEV. Generally, in order to measure fuel consumption and pollutant emissions of cars, chassis dynamometer tests are performed on various driving cycles before actual driving test. There are many driving cycles for performance evaluation of conventional vehicles. However, there is a lack of researches on driving cycle for EV. In this study, the urban driving cycle for performance evaluation of electric vehicles was developed. This study is composed of two parts. In the part 1, the urban driving cycle ’GUDC-EV(Gwacheon-city Urban Driving Cycle for Electric Vehicles)’ was developed by using driving data, which were obtained through actual driving experiment, and statistic analysis with chronological table. In this paper part 2, in order to verify the developed driving cycle GUDC-EV, virtual EV platforms were configured and simulations were performed with actual driving data using In addition, simulation results were compared with existing driving cycles such as FTP-72, NEDC and Japan 10-15.