Presentation of a virtual power plant environment and its application with combined first-principle and data-driven models intended for the diagnostics of a power plant - Part 2

摘要

This work aims to present and discuss the architecture and application aspects of the Virtual Power Plant (VPP) environment, a tool created for the purpose of the modeling, simulation, data management, and visualization of power plant systems. The VPP environment is based on a general-purpose simulation package that provides a framework for incorporating a broad variety of models, ranging from real-time models to detailed models that require off-line execution. The VPP, as a tool, opens up the possibility to develop, test, and validate the feasibility of new software and hardware components within simulated operational scenarios. An original approach to tuning a first-principle model using the example of a feedwater heater by means of a parameter estimation technique is presented, along with a case study of modeling a 225 MW coal-fired power unit. The objectives are as follows: (i) develop a tuning method for this model; (ii) propose key indicators of feedwater heater performance using a model-based approach; and finally (iii) automate the calculation process of the indicators.