Determination of design and operation parameters of a surface condenser using an adaptive neuro-fuzzy inference system Maida Bárbara Reyes Rodríguez ⇑ , Jorge Laureano Moya Rodríguez ⇑ , Cristiano Hora Fontes Industrial Engineering, Polytechnic Institute, Federal University of Bahia, Post Code: 40210-63 Brazil article info Article history: Received 13 February 2019 Received in revised form 23 March 2019 Accepted 7 April 2019 Keywords: Surface condensers ANFIS Geometrical parameters Operating conditions abstract Surface condensers are among the widest equipment used in various industrial sectors as well as in power plants. One of the most common types of surface condensers is shell and tube heat exchangers which are robust and easy to use in multifold environments. On the other hand, adaptive neuro-fuzzy inference systems (ANFIS) are widely used in many engineering applications such as healthcare services, production management, quality control, emergency responses, traffic control and so on. In the current study, a shell-and-tube-surface condenser is represented by a phenomenological model and a parametric analysis of the operational and geometric variables is performed. These results are adjusted by an ANFIS in order to enable a systematic tool that allows the design and evaluation of the equipment under other operating conditions. The use of ANFIS to evaluate the operation of a surface condenser or to make a pre- vious design based on certain operating conditions demonstrates advantages over the use of traditional approaches such as the Engineering Equation Solver (EES) because the use of fuzzy rules offers a high combinatorial option in the analysis that joint with the possibility of learning of neural networks allow to obtain a reliable and easy application tool to quickly achieve the desired results within a planned range of operation and design, being possible, in addition, to change the ranges of the input variables without needing to do calculations, only reusing the tool obtained with different data, depending on the condi- tions of the process in question. The results obtained were compared with a conventional evaluation tool (EES) and the percentage error when comparing both approaches is lesser than 10%, which demonstrates the validity of the neuro-fuzzy system proposed. Ó 2019 Published by Elsevier Ltd. 1. Introduction Heat transfer process between two fluids at different tempera- tures is imperative for most industrial processes and heat exchang- ers are the devices that specifically designed for this purpose. Heat exchangers are broadly used in many applications, such as thermal power plants, petroleum refineries, chemical plants, food indus- tries, air conditioning, refrigeration etc. [18]. According to [25], shell and tube heat exchangers are the most widely used (37% of total heat exchanger market) among different types of heat exchangers, because of their simple manufacturing and adaptabil- ity to different ranges of operating conditions, pressures and tem- peratures. A shell and tube heat exchanger contains a tube bundle mounted inside the shell. One fluid flows over the tubes (shell side) and the other fluid flows through the tubes, and heat transfer occurs between these two streams of fluids. The heat exchanger can operate in a single-phase mode to heat up or cool down fluids or it can operate in a two-phase mode as an evaporator or condenser. A surface condenser is a type of shell and tube heat exchanger usually used in conjunction with condensing steam turbines. It is the equipment responsible for the ‘‘production” of the vacuum in the exhaust of these turbines - in order to maximize the available energy jump in the steam - and the return of condensate to the sys- tem. The vacuum is produced by the drastic reduction of specific volume resulting from the condensation of the steam. When the steam comes in contact with the thousands of tubes cooled by the cooling water, turns into liquid and returns to the system to again turn into steam in the boiler in a closed cycle, efficient and without waste. In order to evaluate the operation of a surface condenser, as well as in any other type of condenser, the most important output variables are the condensate mass flow rate and the mass flow rate of cooling water. The first one determines (such as in a power https://doi.org/10.1016/j.ijheatmasstransfer.2019.04.031 0017-9310/Ó 2019 Published by Elsevier Ltd. ⇑ Corresponding authors. E-mail addresses: maidabrr@gmail.com (M.B.R. Rodríguez), jorgemoyar@gmail. com (J.L.M. Rodríguez), crfontes@ufba.br (C.H. Fontes). International Journal of Heat and Mass Transfer 138 (2019) 17–24 Contents lists available at ScienceDirect International Journal of Heat and Mass Transfer journal homepage: www.elsevier.com/locate/ijhmt