Research on low power scheduling in open CNC system using hybrid genetic algorithm

Aiming at the NP-hard characteristics of low-energy scheduling problems on multi-core platforms, a hybrid genetic algorithm with variable neighborhood search (HGA-VNS) for periodic dependent tasks in open CNC systems is proposed. The mathematical model of the optimization problem was formulated, and the flowchart of the algorithm was designed. Considering the periodic dependence of the tasks, the chromosome is generated by the operation-based coding method, and the segment boundary is added to the chromosome. To preserve the topology of offspring, specialized initial population generation algorithm as well as crossover operation was presented. Furthermore, a variable neighborhood search was carried out on current optimal salutation to enhance the local optimization ability of algorithm. The experimental results show that the proposed algorithm can effectively reduce the system energy consumption under the premise of guaranteeing the schedulability of system, which proves that HGA-VNS can effectively solve the low-energy real-time scheduling problem on multi-processors.