Browsing by Author "Saanane, Bonaventure Bernard"

Now showing 1 - 1 of 1
  • No Thumbnail Available
    Item
    Multi-objective optimisation of core losses in AC rotating electrical machines by use of fuzzy logic
    (University of Dar es Salaam, 2007) Saanane, Bonaventure Bernard
    Electric motors use more than half of all electricity, worldwide International research bodies calculate that over 50% of drivepower electricity could be saved if motors and drive systems become more efficient. Electrical machines designs present a mathematically indeterminate problem. There is no single optimum design, but there exist different ways to solve a design problem in electrical machines. As a substantial part of the losses in electric machine is the loss in the iron core, it is therefore important to consider them for optimization in order to still improve the motor efficiency. In this thesis, an Iron loss prediction and optimization model was developed which ensures minimized from loss for any standard motor frame. Through the generated optimal points of stator bore (internal stator core diameter), optD, and airgrap magnetic flux density, optB, a new motor geometry is reconfigured which has a guaranteed lowered iron loss and also total motor loss. Through a statistical tool and a curve fitting method novel empirical formulate have been formulated for each motor frame instigated. Experimental results of original motor frames were compared with the new result of the design. This showed an improvement of motor efficiency. Also empirical formulate were developed which can greatly assist motor designers in a much more easier way to search for an optimal motor design, which no necessity every time to develop initially a computer model for every change of a design.