Task Offloading Scheduling with Time Constraint for Optimizing Energy Consumption in Edge Cloud Computing

In this paper, an improved genetic algorithm with delay constraint was designed. When initializing the population, a greedy strategy was adopted to ensure that there were enough excellent genes in the initial population, a normal distribution ordering selection strategy was adopted when selecting the next generation, so that high-quality chromosomes have a greater probability of being selected, and an adaptive cross-mutation strategy was proposed to achieve dynamic probability when cross-mutation was carried out to avoid the problem that the algorithm was easy to fall into the local optimal solution in the later stage, and at the same time, a maximum subsegment crossover strategy and a discardable mutation strategy were proposed to solve the problem of individual solution deterioration after crossover.