Abstract

The purpose of this study was to investigate the error which numerical solution of a differential equation inevitably produced in the simulation of a human movement with a mathematical model. The experimental simulation was to flex the forearm with maximum effort and the rotate a heavy fly-wheel with a thin wire running from the hand to around the wheel. The mathematical model used in this study was composed of an idealized bone-muscle system of the upper extremity and the fly-wheel. In this model the biceps brachii muscle was considered to be the equivalent elbow flexor (2). The muscle model for this equivalent flexor was Hills two component model. The non-linear differential equations used in the mathematical model were solved numerically at short, regular intervals as in the ordinary numerical solution. These differential equations were solved by repeatedly employed different time intervals. Time courses for the force of the elbow flexor and the angular velocity of the forearm throughout the flexion against 5, 50 and 250 kg of equivalent mass of the wheel were determined by solving the equation at different time intervals which were 20, 10, 5, 2.5 and 1.25 ms. The E (i) j was determined from these time courses. Verf.-Referat