Thermal Economy Analysis of External Steam Cooler Based on BP Neural Network
Download as PDF
DOI: 10.25236/iwmecs.2020.012
Author(s)
Shidan Chi, Yan Liang, Tao Luan, Xudong Gao, and Yan Gao
Corresponding Author
Shidan Chi
Abstract
Based on 600MWultra supercritical reheat system, the thermal economic calculation model of ESC (external steam cooler) is established. The BP neural networkis introduced to meet the limitations of the traditional algorithm. Nine inputs and tow outputs are determined from unit parameters. More than 1000 sets of related data sets are got from traditional model, which were used as training data to form a general ESC thermal economy model. The results of the study show that the energy saving effect prediction of ESC is better than the outlet feedwater temperature effect. The thermal economy is related to the superheat of the heater which ESC located. The final feedwater temperature is related to the unit efficiency. With the increase of the feed water temperature of the ESC inlet, the efficiency of the unit becomes U-shaped, so the deviation of the prediction result is relatively large.
Keywords
External Steam Cooler, Thermal Economy, Neural Network, Performance Parameter