Operation optimization of cold‐end system of direct air‐cooled units considering the effect of environmental wind

Abstract

AbstractAn air‐cooled island can significantly alter the heat transfer performance of an air‐cooled condenser due to the reflow of hot air caused by environmental wind. This can result in a considerable deviation between the backpressure calculated by traditional air‐cooled condenser models and the actual value. To address the issue, a research study was conducted on a 600‐MW direct air‐cooled unit. Numerical simulation methods were used to obtain the corresponding air flow rates and fan inlet air temperatures for each air‐cooled heat exchanger, which were then combined to establish a backpressure calculation model. From the above model, the backpressure prediction model and unit net output of full conditions were established using a backpropagation neural network. Therefore, taking the net output as the optimization objective, a genetic algorithm was used to compute the optimal backpressure and optimal fan speed in off‐design situations. Compared with traditional calculation approaches, the model produces backpressure predictions that were closer to the actual situation under the effect of ambient wind. The results indicate that both the optimal backpressure and fan speed were positively correlated with the exhaust flow and ambient temperature. It has been observed that when a unit was affected by different wind directions, the effect of the forwarding wind on the backpressure was smaller than that of other wind directions, especially under high‐load conditions. Moreover, the fan group operates close to full capacity under high‐temperature and high‐load conditions. Therefore, considering the influence of ambient wind, the obtained optimal backpressure and fan speed under variable working conditions were more realistic.