Quotient algorithm in monitoring disturbance processes
AssigneeTelefonaktiebolaget LM Ericsson (publ)
InventorDavid LeCorney
Life of this patent
Estimating the term…
Abstract
The Quotient algorithm is used for controlling the monitoring of disturbances in a physical process and disturbance-counting values acquired in the monitoring are utilized to determine a disturbance frequency and a peakedness value for the disturbance process. The peakedness value is used as a measure of the bursty behavior of the disturbances. The disturbance frequency and the peakedness value determined from the acquired disturbance-counting values are utilized in calculating more refined thresholds for the algorithm, and the thresholds initially set for the Quotient algorithm are replaced by these more refined thresholds, thus improving the performance and quality of the disturbance monitoring.
Claims22
A method for performing monitoring, controlled by a Quotient algorithm, of disturbances apt to occur at random or in bursts in a physical process, said method comprising the steps of:
- defining an abnormal event regarded to be a disturbance;
- defining a base against which disturbances are to be counted;
- designing the Quotient algorithm by setting a base-event threshold H and a disturbance threshold T;
- monitoring the disturbances of the physical process by counting base events, counting disturbances and using the Quotient algorithm to selectively raise an alarm;
- determining for the physical process, a value F of a peakedness factor defined as the ratio of the variance to mean of occurrences of disturbances in the physical process based on the disturbance-counting values acquired in the monitoring peakedness being used as a measure of how bursty the disturbances are;
- and
- adjusting at least one of said base-event threshold H and said disturbance threshold T of said Quotient algorithm based on at least said value F of the peakedness factor.
A, system for monitoring disturbances in a physical process, said system comprising:
- a disturbance counter for counting abnormal events regarded to be disturbances;
- a base-event counter for counting occurrences of a base against which the disturbances are counted;
- a control unit implementing a Quotient algorithm for controlling the monitoring of the disturbances of the physical process, the Quotient algorithm using a base-event threshold, a disturbance threshold and counter values obtained from said disturbance counter and said base-event counter to selectively raise an alarm,
- means for determining at least one value of the disturbance frequency of the physical process based on disturbance-counting values acquired in the monitoring;
- means for determining for the physical process, a value of a peakedness factor defined as the ratio of the variance to mean of occurrences of disturbances in the physical process based on the disturbance-counting values acquired in the monitoring, peakedness being used as a measure of how bursty the disturbances are;
- and
- means for adjusting said base-event threshold of said Quotient algorithm based on at least said value of the disturbance frequency and said value of the peakedness factor, and for adjusting said disturbance threshold of said Quotient algorithm based on at least said value of the peakedness factor.
A method for performing an algorithm-controlled monitoring of disturbances apt to occur at random or in bursts in a physical process, said method comprising the steps of:
- defining an abnormal event regarded to be a disturbance;
- defining a base against which disturbances are to be counted;
- configuring a first monitoring algorithm, a Quotient algorithm, by setting a base-event threshold and a disturbance threshold;
- counting base events and counting disturbances in accordance with the Quotient algorithm,
- determining at least one value of the disturbance frequency of the physical process based on disturbance-counting values acquired in the monitoring;
- determining for the physical process, a value of a peakedness factor defined as the ratio of the variance to mean of occurrences of disturbances in the physical process based on the disturbance-counting values acquired in the monitoring peakedness being used as a measure of the bursty behavior of the disturbances of the physical process;
- determining threshold values for a second monitoring algorithm based on said value of the disturbance frequency and said value of the peakedness factor;
- and
- continuing the monitoring of the disturbances of the physical process using the second algorithm.
Dependencies are inferred from the claim text.