OPTIMAL DESIGN OF ELECTROMAGNETIC SUSPENSION OF THE MAGLEV TRAIN USING A GENETIC ALGORITHM
DOI:
https://doi.org/10.53297/18293336-2022.2-23Keywords:
MAGLEV train, electromagnet suspension, electromagnet with straight armature, optimal design, minimal cost of active materials, minimal consumed electric power, genetic algorithm, optimization, multiobjective optimizationAbstract
MAGLEV high-speed trains (trains on a magnetic cushion) float over the road surface while moving, not touching the surface of the rails. The necessary air gap between the track and the train is provided by an electromagnetic suspension.
The process of designing an electromagnetic suspension (an electromagnet with a straight-line armature) is considered at given values of the weight of the train car, the number of suspensions per car, the height of the train above the road surface, and the ambient temperature necessary for safe and reliable driving. As a result, the values of the structural dimensions and parameters of the suspension are determined, providing the value of the electromagnetic force acting on the anchor of the suspension, which is necessary for lifting the wagon from the canvas and holding it at the required height.
In order to develop a methodology for the optimal design of an electromagnetic suspension, two one-objective optimization problems of an electromagnet and corresponding target functions are formulated - the minimum cost V of active materials (steel magnetic wire and winding wire of the control coil), related to production costs, and the minimum electric power consumed by the suspension Q, related to operational costs A system of inequalities representing the limits of values of construction coefficients and parameters of the device is made up. A genetic algorithm is selected as the optimization method, as a result of which the numerical values of the design coefficients and parameters of the projected electromagnet, including the required minimum values of the target functions, are obtained for two formulated single-criterion problems.
A multiobjective optimization problem with objective functions - minimal cost V and minimal consumed power P and with the obtained method of the ideal point of the generalized objective function σ - is also formulated and solved. This multijbjective problem is also solved by genetic algorithm.
