Abstract: A method using an algorithm-controlled monitoring of disturbances apt to occur at random or in bursts. Counting values are used for counting the disturbances. An abnormal event regarded to be a disturbance is first definded. Then, a base against which disturbances are to be counted is defined, followed by defining a unit to be used as a measure of a disturbance frequency. Values of the disturbance frequency are determined in a variety of circumstances, the values including a critical value fC of the disturbance frequency where the monitoring nominally issues an alarm. At the critical value there is determined a peakedness factor F that is a measure of how bursty the disturbances are. An inertia value J is chosen that is a measure of how fast or slowly the algorithm is desired to react to changes in the disturbance frequency, so as to achieve an acceptable compromise between speed and reliability of the monitoring.

Abstract: A method for reacting to disturbances that occur in industrial processes includes the steps of extracting information regarding occurrence of disturbances from an industrial process, transmitting the extracted information to a disturbance filter which includes a leaky bucket having a defined disturbance step and floor and ceiling height for the disturbances, and triggering an alarm when occurrence of disturbances in the industrial process falls without the defined disturbance step and floor and ceiling height. The present invention also provides a rigorous method of setting parameters within a leaky bucket algorithm.