Optimal operation of boiler plant with thermal storage

Industrial boiler plants play an important role in energy conservation, for there is huge amount of energy consumed by a great number of industrial boilers. Optimal operation of an industrial boiler plant through optimal load assignment to boilers in the boiler plant is one of the major effective methods to improve the real efficiency of the plant. It is especially convenient for industrial boiler plants to carry out optimal load assignment in practice when the boiler plants are assisted by thermal energy storage (TES) that stabilizes the load levels of boiler plants. To accomplish this, accurate load forecast is a prerequisite. Based on the analysis of recorded load data the error back propagation (BP) neural networks are used to perform the forecast of daily load curves. Combined with the segmentation of the daily load curve, the minium-departure model (MDM) that is newly developed is used to optimally assign the plant load, which is invariant due to TES during a time segment, to various boilers in the plant, so that the boiler plant can achieve the minimum daily fuel cost. A computer program was completed and used in an example. The results show that the fuel costs of the boiler plant are reduced.