A parallel computing framework for common services of power grid computational deduction

With the construction of new power system, the original hierarchical structure of the power grid has been broken. The operation mode of power grid is becoming increasingly complex. The traditional power grid computational deduction methods are no longer suitable for the actual business requirement of the power grid. Therefore, advanced technology is needed to meet the real-time requirement of power grid computational deduction. Parallel computing is an efficient computing mode. This technology utilizes multiple types of computing resources to collaboratively execute complex computing tasks, which can achieve efficient utilization of computing resources and effectively improve solution speed. The computational deduction of power grid involves multiple types of practical business scenarios. When power grid is in abnormal situation, it can have an impact on related business scenarios and lead to problems such as local or even large-scale power outages. In response to the real-time requirements of computational deduction, this article proposes a parallel computing framework for the common service of power grid computational deduction. This method includes three parts: behavior prediction of computational deduction services, adaptive strategy generation of parallel computing, and strategy execution of adaptive optimization, which is to achieve precise and efficient dynamic strategy adaptation.