Research on Task Scheduling Algorithms in Control System Oriented Parallel Testing Techniques

Task scheduling algorithms have consistently been the foundation of parallel testing technology, particularly in the domain of control systems. The complexity of testing tasks in these systems presents a significant challenge to the task scheduling of parallel testing technology. Existing algorithms often fail to fully consider the back-and-forth dependencies between tasks and resource preemption when dealing with parallel testing tasks in control systems. Consequently, the generated scheduling schemes lack efficiency and flexibility. To address this issue, a set of task planning rules has been proposed, based on an analysis of test task information and common conflict patterns. These rules aim to enhance the efficacy of parallel test scheduling in control systems. A multiple swarm genetic algorithm, which incorporates the ant colony mechanism, is thus devised. This algorithm combines the pheromone guidance mechanism of the ant colony algorithm with the crossover and mutation operations of the genetic algorithm. Furthermore, the multiple swarms parallel search strategy is employed in order to achieve the dual-objective optimization of the total test time and the balance of the test resources. The simulation results demonstrate that the designed algorithm is capable of achieving stable convergence while simultaneously enhancing the dual-objective optimization rate and optimization stability.