IMPROVED ALGORITHM FOR SOLVING MULTIPLE-EFFECT EVAPORATOR PROBLEMS

Multiple-effect evaporator problems are solved in three steps. The first step is construction of equations to solve a problem, the second step is to solve the equations other than by a trial-and-error method, and the third step is to solve the equations by use of an iteration method. A new algorithm for the third step is proposed in this paper. It consists of construction of an equation-variable matrix, rearrangement of the matrix, and solving of these equations according to a procedure indicated by the rearranged matrix. The features of the algorithm are a setting up of equations to converge well, value of elements in the matrix, and simple rearrangement of the matrix. A number of multiple-effect evaporator problems are tested. The majority of problems could be solved by use of 0 or 1 for elements of matrix. However, several problems must be solved by use of a revised algorithm which uses 0, 1 or m (m > 1) for elements of the matrix. The revised algorithm suits problems that have poor convergence on using 0 or 1 for elements of the matrix.