Research on Asset Management of Power Grid Engineering Based on Asset Group

Abstract

In recent years, with the slowdown in power generation, the reform of transmission and distribution prices has continued to deepen, and objectively requires more precise management of projects invested by power grid companies. This paper takes the asset group as the evaluation unit of grid engineering asset management, starts with the characteristics of the asset group, expounds the scope and division criteria of the asset group, and determines the asset group classification. The cost and benefit sharing method of the asset group is proposed. The actual cost method is used to collect the components of each asset. The electricity and cash flow based methods are used to separate the asset groups of the transmission and distribution networks. The proceeds are apportioned. The asset group input-output evaluation system was constructed. Through the multi-dimensional selection of indicators reflecting the input and output of the asset group, a comprehensive evaluation of the operation of the asset group was made. The cost-benefit sharing method and input-output evaluation system proposed in this paper can effectively guide the asset management of power grid projects and assist the power grid companies in making decisions.

A Review of Asset Management using Artificial Intelligence based Machine Learning models with applications for the Electric Power and Energy System

“…Future studies might use a hierarchical control framework based on reinforcement learning for complex systems. Modern power systems [18,19] are getting larger, more sophisticated, and have more operating conditions & controlling options. The single-agent reinforcement learning algorithms use centralized frameworks that primarily rely on uninterrupted transmission lines, making them unable to scale up to huge systems or handle communication delays.…”

A review of asset management using artificial intelligence‐based machine learning models: Applications for the electric power and energy system

“…Future studies might use a hierarchical control framework based on reinforcement learning for complex systems. Modern power systems [19,20] are getting larger, more sophisticated, and have more operating conditions and controlling options. Single-agent reinforcement learning algorithms use centralized frameworks that primarily rely on uninterrupted transmission lines, making them unable to scale up to huge systems or handle communication delays.…”