As mentioned at the outset, the main purpose of this research is to develop a Fault Detection and Diagnosis (FDD) method that accommodates nonlinear characteristics of the system, variable operating conditions, signals from all possible sources and a diagnostic result with some imprecision targeted at emulating operators way of describing faulty conditions. To achieve the objective, the whole work was divided into sequence of steps. The first step was about reviewing the signals available for measurement and the faults likely to occur in a CCP. The conclusion from this stage has led us to choose appropriate model identification method. The second step was focused on model identification for normal operating conditions.