Combining genetic and deterministic algorithms for locating actuators on space structures

Genetic algorithms are a powerful tool for the solution of combinatorial problems such as the actuator placement problem. However, these algorithms require a large number of analyses with the attendant high computational costs. Therefore, it is useful to tune the operators and parameters of the algorithm on problems with inexpensive analyses that are similar to computationally more expensive problems. An easy-to-calculate measure of actuator effectiveness is employed to evaluate several genetic algorithms for a problem of placing actuators at 8 of 1507 possible locations. Even with the best of the algorithms and with optimum mutation rates, tens of thousands of analyses are required for obtaining near-optimum locations. A hybrid procedure is proposed that first estimates near-optimum locations with a deterministic algorithm and then seeds these locations in the initial population of a genetic algorithm. A simulated annealing technique is also applied as a mutation operator for the genetic algorithm. The hybrid procedure reduces the cost of the genetic optimization by an order of magnitude.