When studying for a doctoral degree (PhD), candidates submit a thesis that provides a critical review of the current state of knowledge of the thesis subject as well as the student’s own contributions to the subject. The distinguishing criterion of doctoral graduate research is a significant and original contribution to knowledge.
Once accepted, the candidate presents the thesis orally. This oral exam is open to the public.
Robust structure, high reliability and low maintenance costs allow induction motors to be widely-used in various industrial applications. In recent decades, due to the increased concerns on global warming, and the effort to enhance the efficiency of tools, equipment, and systems, efficiency of induction machines (IMs) has received a lot of attention. The rated efficiency of an IM can be found on the nameplate. However, it is affected by aging, ambient temperature, load, supplied voltage and other technical reasons. Furthermore, based on NEMA MG 1 standard, the actual efficiency of an IM may vary from the nameplate value. As a result, efficiency estimation of IMs is essential to evaluate the efficiency of the whole system and energy cost.
Applying available international standards to in-situ machines needs load decoupling and in some cases, the no-load/locked-rotor test is required. This is not allowed with in-situ machines. Therefore, having a non-intrusive method which is capable of estimating the efficiency of the machine by using only available data such as the input voltages, currents, active power and nameplate data is necessary.
This thesis investigates in-situ methods to determine the efficiency of IMs and three related subjects are addressed. First, an optimization based algorithm is proposed to determine the efficiency of the IM at different loads. This algorithm is proven to have minimum intrusiveness and only uses the data of one operating point of the machine. Assumptions and techniques to increase the accuracy of the algorithm are addressed. The proposed algorithm is then applied to two conditions. In the first condition, the required input data are recorded when the machine reaches its thermal stability and final temperature rise of the machine is used as an input. In the second condition, the required input data are recorded 30 minutes after start of the machine and then the machine final temperature rise is predicted. Two approaches are proposed to predict final temperature rise and are based on machine insulation class and temperature rise of the machine in the first 30 minutes of operation of the machine after start.
Moreover, a method is proposed to determine the range of IM equivalent circuit parameters and improve the probability of converging to the correct answer. The method is based on the nameplate data of the machine and empirical results provided by Hydro-Québec. The method is also improved by using the operating data of the machine. The proposed range determination is very helpful for in-situ applications where the output power of the machine is not available.
Second, two dimensional finite element analysis (FEA) is used to predict the efficiency of the IMs at different loads. Two methodologies are adopted. In the first methodology, the losses are calculated directly using FEA while in the second one, the equivalent circuit parameters are first estimated using FEA and then the efficiency at different loads are estimated using the equivalent circuit parameters. To improve the results, a simple formula based on the rated power of the machine is proposed to evaluate the friction and windage losses also known as the mechanical loss of the machine. The proposed formula is applicable for 4-pole 60 Hz IMs and achieved after study of more than 100 IMs of this type.
Third, the effects of the adjustable-speed drives on the losses and efficiency of the IMs are addressed. The direct torque control and scalar control schemes implemented by an industrial drive are employed to control two types of IMs. Two IMs designed for direct-fed application and two other IMs designed for PWM applications are studied while method B of IEEE Std-112 is applied to segregate the losses. Variations of different losses at drive-fed and direct-fed conditions are compared and results are discussed.