New approximation algorithms for the state functions of water and steam for the application of transient processes and fast on-line applications

Operational optimisation for transient processes to exploit full potential of industrial plants has become indispensable today. To overcome this problem, accurate and fast determination and prediction of water-steam transients require precise algorithms with easy structure of the approximation, to allow algebraic transformation of equations and equation systems. This is beneficial especially for solving partial differential equations in the area of thermodynamics for water/steam systems. Our contribution focuses on the definition of new approximation algorithms for the determination of state functions of water/steam for dynamic simulation. Two main aspects of this work are to reduce computational time (by saving more than 50% just for the calculation of a single property), using a direct method with a given accuracy to enable extensive dynamic process simulation calculations also for real-time applications like demand-side management and optimal control, and to keep the whole set of equations as linear equations which can be directly solved for the unknown parameter. Additionally, occurring errors by parameter variation of the respective approximation were analysed. The practicability is demonstrated by a simple highly dynamic application example of a steam drum in load cycling behavior comparing the developed new linear and quadratic approximation functions with the industry standard IAPWS-IF97.