Research On Visual Analysis Method Of Control Optimization Of Complex Parameters In Thermal Power Plants

Nowadays,thermal power generation still plays an important role in China's power production.As a large energy consumer,energy conservation and emission reduction are particularly important.Therefore,how to optimize the power unit running and improve the production efficiency is a continuous concern of the electric power enterprises.With the help of automatic distributed control system(DCS)equipped in the power plants,a large amount of monitoring data has been accumulated,which is valuable for operating control optimization.However,it has not been effectively used in the actual production optimization.The monitoring data contains both complex mechanism background and stochastic characteristics,and also involves dense sampling and numerous number of parameters,which makes traditional analytical methods hard to investigate valuable information.In this thesis,an interactive visual analysis tool(iDCS)is designed to facilitate the analyzing process of parameter optimization in the thermal power units.iDCS can support the exploration in large-scale multivariate time-series monitoring data to discover hidden patterns and to develop the optimizing strategy.Firstly,through analyzing the needs of operating optimization of the thermal power units,a visual analysis data flow model for complex parameter optimization is proposed.Then,a series of coordinated views is designed to support observation,exploration and evaluation.Through these views and the interaction techniques between each other,the complex correlation and changing rules among parameters are revealed and multi-sides evaluation are supported.Finally,to test the practicality and effectiveness of iDCS,a real thermal power plant dataset is adapted in the case study,and domain experts are also invited during the test and evaluation.This thesis makes a meaningful attempt in the application of visual analysis in industrial big data,which provides a valuable reference for related research of visual analysis on similar industrial monitoring data.