Hits:

Affiliation of Author(s):自动化学院

Title of Paper:Condition Monitoring and Prognosis of Power Converters based on CSA-LSSVM

Journal:2017 INTERNATIONAL CONFERENCE ON SENSING, DIAGNOSTICS, PROGNOSTICS, AND CONTROL (SDPC)

Key Words:Boost converters Condition monitoring Parameter estimation Crow search algorithm(CSA) Prognosis

Abstract:Condition monitoring and prognosis are effective methodologies to reduce the downtime and maintenance cost and improve the reliability and lifespan for power electronic converter systems. This paper presents a model-based method to implement the condition estimation and a data-driven method to conduct the prognosis for boost converter. Firstly, the equivalent circuit for non-ideal boost converter should be simplified and the state space equation should be obtained. Then, constructing the objective function for crow search algorithm (CSA) and several parameters like inductance, on-resistance of Metal-Oxide Semiconductor Field-Effect Transistor (MOSFET), capacitance and equivalent series resistance of capacitor are estimated based on CSA. Considering components degradation and variable operating conditions, several simulation experiments have been conducted to validate the presented approach. Finally, the prognosis for capacitor of the boost converter has been conducted based on the least square support vector machine (LSSVM) algorithm. The results show that this technique characterizes high computation efficiency and good estimation accuracy. Also, it will be a basis of further study for the circuit-level remaining useful life prediction.

Translation or Not:no

Date of Publication:2017-01-01

Co-author:Sun, Quan,Jiang, Yuanyuan,Shao, Liwei

Correspondence Author:wang you ren