Effect of Operation Conditions on Exit Water Temperature of Condenser (Atmospheric) by Using Neural Network

Abstract

The goal of this research is the determination exit water temperature of a condenser (atmospheric) use in steam power plant by artificial neural network with various operation conditions. Input of neural network include surface area, inlet water temperature, water flow rate, steam temperature, enthalpy difference and steam flow rate. Output of the neural network consists of the exit water temperature. For the subject of the neural network, training or learning algorithm are applied the most famous of which is back propagation algorithm. This algorithm is a systematic method for training multi layer artificial neural network. The real exit water temperature first using experimental work and is defined as a goal function for neural network (NN) , so that all outputs of the network can be compared to this function and the error can be calculated. Then another a set of input from experimental work was used to test the NN, the performance of the NN is optimum. Compared with a validated first model, the standard deviations of neural network models are less than 0.12%, and all errors fall into .