Semigroup based neural network architecture for extrapolation of enthalpy in a power plant

This paper presents a semi-group based neural network architecture for extrapolating steam enthalpies along the temperature axis in the water-steam cycle in a power plant. The assumption is that steam pressures and temperatures are known (available) in a limited high-pressure, high-temperature region, which allows enthalpy calculations to proceed in this limited region but are not available in an adjoining high-temperature region. This method of extrapolation provides for similar calculations in the extended region