The Worlds’ First Ever Cooling Tower Acceptance Test Using Process Data Reconciliation

The cooling capacity of cooling towers is influenced by multiple constructive and atmospheric parameters in a very complex way. This leads to strong variations of the measured cold-water temperature and causes unacceptable unreliability of conventional acceptance tests, which are based on single point measurements. In order to overcome this lack of accuracy a new approach to acceptance test based on process data reconciliation has been developed by BTB Jansky and applied at a nuclear power plant. This approach uses process data reconciliation according to VDI 2048 [1, 2] to evaluate datasets over a long period covering different operating conditions of the cooling tower. Data reconciliation is a statistical method to determine the true process parameters with a statistical probability of 95% by considering closed material-, mass- and energy balances. Datasets which are not suitable for the evaluation due to strong transient gradients are excluded beforehand, according to well-defined criteria. The reconciled cold-water temperature is then compared, within a wet bulb temperature range of 5°C to 20°C to the manufacturer’s guaranteed temperature. Finally, if the average deviation between reconciled and guaranteed value over the evaluated period is below zero, the cooling tower guarantee is fulfilled.