Induction motor variable speed drive performance and impact on energy savings

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

ESKOM, together with Municipalities distribute power at a line-to-line voltage of 400+/-10%. Hence in theory, the user ends up with a voltage of 340 to 440. In urban areas 2% voltage unbalance is considered acceptable, while 3% voltage unbalance is considered acceptable for rural areas. The combined effects of under or overvoltages coupled with "acceptable maximum" voltage unbalance is not understood or documented. This project is a comprehensive study of the effects of voltage unbalance in combination with under or overvoltages on induction machines (1M) variable speed drives. In addition the performance with PWM inverters connected to 1M via long cables is also studied. The project starts by an evaluation of previous work on the subject and related topics. Relevant conclusions are made based on the literature review in Chapter One. The second step of the project was to develop two test beds in the UCT Machine's Laboratory. The test beds are for performance testing of low (up to 15kW) and medium (up to 75kW) power 1M and their VSDs. The test bed losses are determined at 4 different operating points and documented. Several tests such as the blocked rotor and no load tests were undertaken to determine the machine parameters. A PSpice simulation model was developed and tested for the low voltage test bed. The test results compared well to simulation predictions. Furthermore, tests under unbalanced and overvoltage conditions were performed and the results verified with simulations. This proves that the developed PSpice model can reliably simulate the actual test bed for both balanced and unbalanced conditions. Moreover, it can be easily applied to model different size machines and drives. The information required includes: the motor parameters, drive dc-link parameters, modulation ratio (ma) and the switching frequency. The impact of voltage unbalance in drives with long cables between the inverter and motor is examined. A 36m long, 4-core, PVC insulated cable is used on the low voltage test bed for a case study. The model used for the cable is a lumped parameter model. The 1M model is also taken from the literature to include a high frequency branch, to model the reflected voltage wave phenomenon. A basic analysis of energy savings in pumps and fans load associated with the application of variable speed drives is also done. Boiler feed-pumps from Tutuka power station are used as a case study that may be replicated for other applications in utilities or industry. The developed Excel Spreadsheet program to project the value of savings and determine the payback period is compared to an ABB program. The two results agree within an acceptable range. Therefore, the knowledge of the pumping system specifications and the cost of energy in kWh with the Energy Savings Program would enable one to project energy savings.

Bibliography: p. 132-134.