Ensemble dual recursive learning algorithms for identifying flow with leakage

In industrial process, pipes and tank may leak and sensors may have biased since corrosion, measuring noise and instrument faults exist. In order to maintain production and to prevent accident from happen it is crucial to develop reliable method of analyses of flammable gas release and dispersion. Relative mass release of the leakage is introduced as the input for the simulation model and the data from the simulation model is taken at real time (on-line) to feed into the recursive algorithms. The objective of this paper is to introduce a combination of advantages of different algorithm scheme into one learning algorithm. For this purpose, three models is developed, first using recursive least square algorithm (RLS), second using recursive instrument variable (RIV) algorithm and lastly using combination of this two algorithm. This paper proposed that, combination of two algorithms into one learning algorithm for predicting mass flow rate of a flow with leakage resulting in a better mass prediction error compare to a model with single learning algorithm.