Optimization study on the siting of maintenance stations for distributed radars based on an improved K-means algorithm

In this study, the siting of radar maintenance stations was optimized concerning the distributed networking deployment patter n of high-precision radars. A genetic algorithm was introduced into the K-means algorithm, forming a GA-K-means algorithm. Besides, a K-means algorithm with constraints was designed to satisfy the actual task demands. With the global search ability of the genetic algorithm, the GA-K-means algorithm was provided with ideal initial centers of mass so that this algorithm became insensitive to the initial values and obtained a good clustering effect. By simulation experiments and changing the radar positions constantly, it was verified that the proposed GA-K-means algorithm was robust and generalized. The experimental results demonstrate that the maintenance station siting scheme based on the GA-K-means algorithm shortened the distance from a maintenance station to the corresponding radar sites and thus reduced the maintenance cost while satisfying the task constraint type.