Supervisor: Dr. Marcus Mueller, University of Edinburgh
PhD Student: Eddie Yew Chuan Chong, University of Edinburgh
Status: Closed
The main objective of this project is to develop a generic design tool to model the air-flow within an electrical generator. Current flowing in an electrical generator produces heat, which must be removed in order to ensure reliable operation. The amount of heat that can be removed is one of the limiting factors in the output of a generator. An understanding of the airflow within the machine allows the designer to produce a more optimised generator design.
The electrical generator used in a wave and tidal current device tends to be an off-the-shelf induction or synchronous generator, which has not necessarily been designed for the specific application. An off-the-shelf machine tends to be designed for an ambient of 20ºC and a temperature rise of 100ºC above ambient. In a wave or tidal current device it is very likely that the ambient will be less. In some devices a cooling fluid will be available to assist with the generator cooling so that the temperature rise will be less than 100ºC. In current PhD research, Hodgins showed that the generator in an oscillating water column (OWC) device on Islay experiences a maximum ambient of 14ºC and the airflow over the generator from the OWC provides an effective cooling medium so that the temperature rise was less than 60ºC. The generator used was not designed taking this information into account. As a result the user could have operated the machine beyond its rated value without exceeding the maximum temperature rise and hence could potentially generate more revenue.
Air-flow in electrical machines is complex, and could be modelled using CFD, but current CFD tools do not lend themselves to the design office environment or to the development of optimisation design tools. In this proposal the PhD student will develop a design tool in which the airflow is modelled by pipes, with different shaped pipes being used to model the different air path geometries within the machine. CFD will be used as a means of verifying and refining the air-flow pipe models. Experimental results will also be used to provide further validation.
The proposed project will build upon work on two previous PhD projects (Hodgins and Mejuto), both of which have been successfully submitted. In these projects thermal models have been developed for the induction machine and the synchronous machine. These models are based upon lumped parameter models, and were validated using experimental results using machines instrumented with thermocouples. Hodgins developed a thermal model and investigated the impact of the external air-flow within the OWC environment on the heat transfer of the machine. Much of this work was done experimentally on the Islay OWC device in collaboration with Wavegen. He did not include any internal air-flow modeling. Mejuto developed a thermal model of the synchronous generator without taking into account any external airflows. Even though a basic internal airflow model was used, good temperature results were obtained. However, the conclusion from both theses was that an improved internal and external air-flow model would improve the combined electromagnetic-thermal optimization.
The air-flow design tool developed in this project will be integrated into the induction and synchronous machine thermal models developed in the previous projects. For both of theses projects an induction machine and synchronous machine were fully instrumented with thermocouples to allow experimental validation. These machines will be used in this project to allow validation of the air-flow models, which will require flow sensors to be integrated into the various paths within the machine. In addition to these conventional generators, a novel low speed permanent magnet generator developed for direct drive wind applications will also be used to demonstrate the application of the air-flow design tool.