A Self-Adaptive Cuckoo Search Algorithm for Energy Consumption Minimization Problem with Deadline Constraint

This work presents a self-adaptive cuckoo search algorithm with a new encoding mechanism to minimize the energy consumption in a heterogeneous distributed embedded system that runs tasks with arbitrary precedence constraints. We use the heterogeneous earliest-finish-time rule to construct a relatively high-quality initial solution. For the first time, a parameter feedback control scheme based on Monte-Carlo policy evaluation is used to balance the global and local search, in which way its search ability is greatly enhanced. In the end, the proposed self-adaptive cuckoo search approach is validated with two benchmarks and extensively randomly generated cases, and the experimental results demonstrate that our proposed approach have better performance than its counterparts.