Optimization and benefit evaluation model of a cloud computing-based platform for power enterprises

To address the challenges associated with the digital transformation of the power industry, this research develops an optimization and benefit evaluation model for cloud computing platforms tailored to power enterprises. It responds to the current lack of systematic optimization mechanisms and evaluation methods in existing cloud computing applications. The proposed model focuses on resource scheduling optimization, task load balancing, and improvements in computational efficiency. A multidimensional optimization framework is constructed, integrating key parameters such as path planning, condition coefficient computation, and the regulation of task and average loads. The model employs an improved lightweight genetic algorithm combined with an elastic resource allocation strategy to dynamically adapt to task changes across various operational scenarios. Experimental results indicate a 46% reduction in failure recovery time, a 78% improvement in high-load throughput capacity, and an average increase of nearly 60% in resource utilization. Compared with traditional on-premise architectures and static scheduling models, the proposed approach offers notable advantages in computational response time and fault tolerance. In addition, through containerized deployment and intelligent orchestration, it achieves a 43% reduction in monthly operating costs. A multi-level benefit evaluation system-spanning power generation, grid operations, and end-user services-is established, integrating historical data, expert weighting, and dynamic optimization algorithms to enable quantitative performance assessment and decision support. In contrast to existing studies that mainly address isolated functional modules such as equipment health monitoring or collaborative design, this research presents a novel paradigm characterized by architectural integration, methodological versatility, and industrial applicability. It thus addresses the empirical gap in multi-objective optimization for industrial-scale power systems. The theoretical contribution of this research lies in the establishment of a highly scalable and integrated framework for optimization and evaluation. Its practical significance is reflected in the notable improvements in operational efficiency and cost control in real-world applications. The proposed model provides a clear trajectory and quantitative foundation for promoting an efficient and intelligent cloud computing ecosystem in the power sector.