Maximum likelihood method for energy balance error correction in heat exchanger data

Heat exchanger performance data commonly contain redundant heat transfer rate measurements. Due to measurement uncertainties involved in the experiments, these redundant heat transfer rates have some discrepancies. While it is a common practice and adopted by engineering standards to use the arithmetic mean of heat transfer measurements, the resulting performance indicators of heat exchangers do not result in a minimum uncertainty possible. Also, this approach fails to resolve discrepancies in resulting transport performance parameters depending on the use of UA-LMTD method or effectiveness-NTU method. In this paper, heat exchanger performance data with two heat transfer measurements from hot and cold fluid streams are combined to produce a least uncertainty of the performance indicators. Individual measurements of mass flow rates and temperatures are corrected by most likely errors based on their respective uncertainties. The validity of this method has been demonstrated by Monte-Carlo simulations. Using air conditioning heat exchanger performance data under dry and wet surface conditions, it is demonstrated that the proposed method leads to a minimum uncertainty of the calculated variables.