Biomechanical Modeling of a Seated Human Body Exposed to Vertical and Horizontal Vibrations Using Genetic Algorithms

Abstract

Investigation of vibrations transmitted from automotive to the human body, for evaluating comfort parameters, always has been a concern for car manufacturers. Hence, the study of vertical vibration is more important in comparison with other directions. To fined and study vertical vibrations in a sitting position in two states, some error criteria and the backrest support effect on the apparent mass responses, as well as biomechanical responses consisting of apparent mass, driving point mechanical impedance and seat-to-head transmissibility. In this paper, two new biomechanical models are presented in order to study vertical vibrations in the seated posture in both with and without backrest support states. The first model is an optimized five-degrees-of-freedom (5-DoF) lumped-parameter model with a unique structure to display vertical vibrations in one direction. The other one is a new type of model called the matrix model, on which the stiffness and damping matrixes are employed instead of the spring and damper scalar parameters to present vertical vibrations in two directions: vertical and horizontal. The optimal parameters for 5-DoF and matrix models have been determined using genetic algorithms. Some criteria such as sum of square error, root mean square error, mean absolute percentage error and coefficient of determination (R2) were used to evaluate models’ accuracy. The backrest support showed different effects on the module apparent mass responses, and the module apparent mass responses values without backrest support were higher than those of values obtained in the presence of backrest support. As the extracted scientific results from this research we could said that matrix model is a new method which has many advantages over previous methods, such as simplicity, fewer degrees of freedom and high accuracy.