Power system analysis and optimization for BioApps RoMicP® foot prosthesis / Mohammad Bilal Goolfee

Mohammad Bilal, Goolfee (2021) Power system analysis and optimization for BioApps RoMicP® foot prosthesis / Mohammad Bilal Goolfee. Masters thesis, Universiti Malaya.

	PDF (The Candidate’s Agreement) Restricted to Repository staff only Download (750Kb)
Preview	PDF (Thesis M.A) Download (1426Kb) | Preview

Abstract

A critical component of a motor-powered prosthesis is the power system that generates the required torque at the joint by converting chemical energy stored in batteries into electrical energy and finally mechanical torque. Prior studies focus mainly on the control system of prostheses and barely cover the power component. In order to achieve high power efficiency in the RoMicP® prosthesis, this study investigates the required battery specifications through calculations, as well as the power consumption of the LMG5200 inverter system and the feasibility of a multilevel inverter configuration using the simulation model and software (available online) from the manufacturer, Texas Instrument. The results showed that to fulfil the design requirement of weighing below 2.5kg, the battery size required is 10Ah at 24V and can power the prosthesis for 5k steps per day. Moreover, using the LMG5200, an efficiency of 99.72% was achieved through PWM switching with harmonic distortion of 2.15% for a three-phase output. Furthermore, the proposed multi-level inverter design achieved an efficiency of 59.46% with harmonic distortion of 0.82% and could therefore not be recommended for use in this case, although it demonstrated that a multi-level system could generate cleaner output. In summary, this project successfully investigated the battery requirements, power consumption and efficiency of the simulated system, as well as the feasibility of a multi-level inverter topology. Future studies should focus on investigating other multi-level topologies, and increasing the number of levels, as well as different advanced switching techniques such as Space Vector Modulation.

Actions (For repository staff only : Login required)