Abstract In the intelligent monitoring of operations in outdoor power scenarios, cloud services play a crucial role as essential support technology for task processing and resource allocation. This paper takes quality of service (QoS) and energy consumption as the key indicators to measure the cloud service composition, constructs a bi-objective optimization function including QoS utility value and energy consumption, and then solves the model through an improved non-dominated sorting genetic algorithm to obtain the best cloud service composition. The algorithm improvement includes the introduction of chaotic mapping sequences to generate the initial population, eliminating uncertainties associated with random population generation, and proposing a guided crossover method to steer individuals toward more favorable directions. Exemplary results demonstrate the effectiveness of the improved algorithm through comparative analysis.
Wu et al. (Fri,) studied this question.