Method and apparatus for alarm monitoring
First Claim
1. A computer-implemented method for alarm monitoring of a process operation by an alarm monitoring apparatus comprising a sampling unit, a determining unit, an acquisition unit and a monitoring unit, the method comprising:
- collecting sample data representing one or more parameters of a process by the sampling unit;
determining a distribution type of the sample data by the determining unit;
obtaining an alarm monitoring strategy corresponding to the distribution type by the monitoring unit; and
monitoring the process using the alarm monitoring strategy by the monitoring unit,wherein determining the distribution type of sample data comprises;
determining whether a value level of the sample data is greater than a first threshold value;
if the value level is not greater than the first threshold value, determining whether the distribution type is a single constant-type or multi-level discrete type;
otherwisedetermining whether the sample data is normally distributed;
if the sample data is normally distributed, setting the upper control limit to a maximum value of the sample data and the lower control limit to a minimum value of the sample data, andif the sample data is not normally distributed, determining whether the distribution type is a continuous non-normal distribution type, cyclical trend-up/down-type, or a drift after a periodic maintenance (PM) type.
1 Assignment
0 Petitions
Accused Products
Abstract
A method for alarm monitoring abnormal conditions associated with a process operation includes collecting N sample data representing one or more parameters of a process, determining a distribution type of the sample data, obtaining an alarm monitoring strategy associated with the distribution type, and monitoring the process using the alarm monitoring strategy. The distribution type may be single-constant, multi-level discrete, normally distributed, continuous non-normally distributed, cyclical trend-up/down, and drifting away after a period maintenance. The alarm monitoring strategy includes a unilateral or bilateral control chart. The one-side control chart has either an upper control limit (UCL) of (100−p) or a lower control limit (LCL) of (p), and the two-sided control chart has a UCL of (100−p/2) and an LCL of (p/2), where p is a predetermined false alarm rate which determines the number N. The control limit may be a difference between two adjacent sample values of the sample data.
41 Citations
20 Claims
-
1. A computer-implemented method for alarm monitoring of a process operation by an alarm monitoring apparatus comprising a sampling unit, a determining unit, an acquisition unit and a monitoring unit, the method comprising:
-
collecting sample data representing one or more parameters of a process by the sampling unit; determining a distribution type of the sample data by the determining unit; obtaining an alarm monitoring strategy corresponding to the distribution type by the monitoring unit; and monitoring the process using the alarm monitoring strategy by the monitoring unit, wherein determining the distribution type of sample data comprises; determining whether a value level of the sample data is greater than a first threshold value; if the value level is not greater than the first threshold value, determining whether the distribution type is a single constant-type or multi-level discrete type;
otherwisedetermining whether the sample data is normally distributed; if the sample data is normally distributed, setting the upper control limit to a maximum value of the sample data and the lower control limit to a minimum value of the sample data, and if the sample data is not normally distributed, determining whether the distribution type is a continuous non-normal distribution type, cyclical trend-up/down-type, or a drift after a periodic maintenance (PM) type. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
9. An apparatus for alarm monitoring of a process operation, the apparatus comprising:
-
a sampling unit configured to obtain a number N of sample data representing one or more parameters in a production process; a determining unit configured to determine a distribution type of the sample data; an acquisition unit configured to acquire an alarm monitoring strategy for the distribution type; and a monitoring unit configured to monitor the production process using the alarm monitoring strategy, wherein the determining unit determines the distribution type of the sample data by performing following operations; determining whether a value level of the sample data is greater than a first threshold value; if the value level is not greater than the first threshold value, determining whether the distribution type is a single constant-type or multi-level discrete type;
otherwisedetermining whether the sample data is normally distributed; if the sample data is normally distributed, setting the upper control limit to a maximum value of the sample data and the lower control limit to a minimum value of the sample data, and if the sample data is not normally distributed, determining whether the distribution type is a continuous non-normal distribution type, cyclical trend-up/down-type, or a drift after a periodic maintenance type. - View Dependent Claims (10, 11, 12, 13, 14)
-
-
15. A method for alarm monitoring abnormal conditions associated with a process operation by a server, the method comprising:
-
collecting at least N sample data representing one or more parameters of a process by the server; determining a distribution type of the sample data by the server; obtaining a control chart in response to the determined distribution type by the server; and monitoring the process using the control chart by the server, wherein N is a function of a predetermined false alarm rate p, and wherein determining the distribution type of the sample data comprises; determining whether a value level of the sample data is greater than a first threshold value; if the value level is not greater than the first threshold value, determining whether the distribution type is a single constant-type or multi-level discrete type;
otherwisedetermining whether the sample data is normally distributed; if the sample data is normally distributed, setting the upper control limit to a maximum value of the sample data and the lower control limit to a minimum value of the sample data, and if the sample data is not normally distributed, determining whether the distribution type is a continuous non-normal distribution type, cyclical trend-up/down-type, or a drift after a periodic maintenance type. - View Dependent Claims (16, 17, 18, 19, 20)
-
Specification