Study on Meta-Modeling Method for Performance Analysis of Digital Power Plant

Abstract Digital power plant is the theory and method to improve the operating quality of power plant by quantifying, analyzing, controlling, and deciding the physical and working objects of power plants in the whole life cycle. And the foundation of digital power plant is system modeling and performance analysis. However, there are some problems in the process of modeling establishment and performance analysis. For instance, each component has different dimensions and different types of mathematical description, and the data or information used for modeling are defined differently and belong to different enterprises, who do not want to share their information. Meta-modeling is a potential method to solve these problems. It defines the specification to describe different kinds of elements and the relationship between different elements. In this paper, the collaborative modeling and simulation platform for digital power plant has been established based on the meta-modeling method and the performance of the target power plant has been analyzed from different aspects via field data. The meta-modeling method consists of three parts: syntax definition, model development, and algorithm definition. In the comparative study between the meta-model and the traditional model, maximum average errors of the two methods are 8.72% and 4.74%, which reveals the high accuracy of the meta-modeling-based model. The result shows that the modeling and simulation platform for power plants can be used to reduce costs, decrease equipment failure rate, and improve plant output, so as to guarantee the safety and increase economics.