This paper presents an approach to optimal power harness design based on a modified ant colony optimisation algorithm. The optimisation of the harness routing topology is formulated as a constrained multi-objective optimisation problem in which the main objectives are to minimise the length (and therefore the mass) of the harness. The modified ant colony optimisation algorithm automatically routes different types of wiring, creating the optimal harness layout. During the optimisation the length, mass and bundleness of the cables are computed and used as cost functions. The optimisation algorithm works incrementally on a finite set of waypoints, forming a tree, by adding and evaluating one branch at a time, utilising a set of heuristics using the cable length and cable bundling as criteria to select the optimal path. Constraints are introduced as forbidden waypoints through which digital agents (hereafter called ants) cannot travel. The new algorithm developed will be applied to the design of the harness of a small satellite, with results highlighting the capabilities and potentialities of the code.
展开▼
