Bathaee - K.N.Toosi University of technology Dept. of Electrical Eng. Tehran-Iran
Alireza Raei - K.N.Toosi University of technology Dept. of Electrical Eng. Tehran-Iran
Teshnehlab - K.N.Toosi University of technology Dept. of Electrical Eng. Tehran-Iran

In this paper, a program with artificial neural network (ANN) based on non-linear optimization method in energy management of Combined Heat and Cold and Power (CHCP) is proposed. It is include introduction to cogeneration small power plants and artificial neural network with back propagation and flexible algorithms and modeling and presentation of a program for optimum management. Training data are obtained via non linear programming and running with LINGO software. The range of small power plants that have been considered in the prepared software is expanding (according to table-1). This is considered for residential, industrial and commercial consumptions. For all the cases discussed in the paper, three layer feed forward neural network are used. Back propagation algorithm, flexible method and gain rate training are adopted to solve the problem of cogeneration energy management. Finally, case study is presented and three above mentioned methods are compared.

cogeneration, model, ANN, training, BP algorithm, classic, flexible, gain rate