Stator fault detection and parameter estimation in induction machines

Sammanfattning: In this licentiate thesis the fascinating area of induction machines is considered. Three different problems are treated, each one of them being more or less unsolved although many researchers around the world have addressed them extensively. The first problem is the mathematical modeling of induction machines with faults in one or more of the stator phases. This author believes that good models are absolutely necessary for successful fault detection. The second problem is fault detection of induction machines. Solving this problem in a satisfactory manner would significantly reduce the costs of many industries, saving lots of money for the world economy. The third problem is parameter estimation in induction machines. Knowing the actual parameters of an induction machine is an absolute necessity for tuning model-based fault detection algorithms. For solving the second problem, which is the ultimate goal of this study, the other two problems have to be addressed and solved beforehand. In this thesis an attempt to bring the matter of fault detection in induction machines closer to its solution is made.