Analysis and Implementation of a Novel Optimized Space Vector Modulation Strategy for Multilevel Inverter

• Main Authors: Kuo-Hsien Liu, 劉國賢
• Other Authors: Yie-Tone Chen
• Format: Others
• Language: zh-TW
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/49819057712954708632

碩士 === 國立雲林科技大學 === 電機工程系碩士班 === 95 === Multilevel inverters have been widely used in the high power applications and various multilevel modulation strategies are developing gradually. Among them, space vector pulse width modulation (SVPWM) has better performance, high degree of design freedom and complicated modulation process. Implementation of the SVPWM for multilevel inverter is complicated, tedious and calculation numerously. This is due to the difficulty in determining the location of reference voltage , the calculation for each vector’s dwelling time numerously, and as the number of levels increases, redundant switching states and complicated choice in determining the switching sequence increase dramatically and brings difficulty in selecting switching state and determining switching sequence. Due to the complexity of realization, this thesis presents a novel space vector modulation strategy for the multilevel inverter. n level inverter is equivalent to n-1 two-level inverters by the concept of the equivalent inverter. Since the redistribution of the space vector plane and the establishment of vector decomposition rules, space vector plane and reference voltage will be simplified to n-1 two-level planes and n-1 two-level reference voltages for resynthesize the original reference voltage and controlling n-1 equivalent two-level inverters. Therefore, the multilevel SVPWM with high complexity can be finished briefly by the two-level SVPWM method. Comparing with the conventional nearest three vector approach, this strategy is easy to realize without complex calculation. Besides, this modulation strategy certainly achieve the optimized switching sequence with the only selection way of determine switching sequence. Simulations and experiments will verify its validity and the realization of all functions uses TI Company’s TMS320LF2407A digit signal processor to implement the digital control of whole system.