article_407_kestrel3

Alarm overload

June 15, 2005
Faced with the problem of porting across a large number of alarms from one system to the next, the control room at Kestrel Coal (Australia) required an upgrade to its HMI system.
KESTREL COAL IS an underground mine in the Bowen Basin of Australia that produces high-quality coking coal and thermal coal for export. Owned in part by Rio Tinto Coal Australia, formerly Pacific Coal, the open cut and underground operation uses longwall technology. An upgrade was being considered for the Human Machine Interface (HMI) in the company's control room, and the Kestrel personnel were faced with the problem of porting across the large number of alarms from one system to the next. To complicate matters, the number of alarms was well above the guidelines set out by the Engineering Equipment and Materials Users Association (EEMUA).Additionally, the alarm management system in MacroView (the HMI) had been essentially disabled - the audible alarm disconnected and the alarm page unused by the operators.Reducing and Prioritizing AlarmsTo transition to a new HMI -- while at the same time increase its productivity and improve safety at the plant -- an alarm management application was needed. Increased productivity requirements centered on the ability of operators and supervisors to respond quickly to plant upsets in order to ensure constant production. As well, continuous monitoring and equipment maintenance upgrades were needed by the company in order to reduce down time, and help operators understand process upsets so as to prevent future occurrences. Improving site safety requires that the operators can respond immediately to high priority alarms and thus bring the plant back under control.To achieve these ends, Matrikon was brought in to provide the mechanism to reduce the total number of alarms to an acceptable level, and prioritize alarms and events correctly.The Kestrel Coal project involved collecting alarm data at a centralized Windows 2000 Server machine from two HMIs running under UNIX found at different physical locations.A Line Printer Daemon was installed on the Windows Server Machine to receive the alarms across the network via TCP/IP from the MacroView HMIs (for both the underground and the prep plant). This data was subsequently rerouted to a ProcessGuard Alarms & Events Collector before archiving in the centralized Alarms and Events Archive.Clients could connect from both the business and control LAN to see the alarms in real time and also perform offline alarm and event analysis on the historical data.Significant Plant improvements
According to the company, the installation of the alarm and events collector at Kestrel, and the intelligent use of the newly installed data mining tool, reportedly resulted in a 97% reduction in mean daily alarm count and the suppression of frequent alarm storms.By reducing the total alarm count to a manageable level, attention was subsequently placed on correctly prioritizing alarms as low, high, emergency and critical: thus achieving the recommendations set out in the EEMUA 191 Alarm Management Standard.The benefits of achieving such standards include:
  • High priority alarms are clearly defined and can be responded to immediately, improving plant safety.
  • Improves operator response times through reduction of total number of alarms, reducing plant
    downtime.
  • Helps identify the root cause of alarms, providing the ability to focus maintenance, reducing plant
    downtime. In fact, a more comprehensive understanding of the plant and process can be acquired from root cause analysis of the alarm and events just prior to downtime events. An understanding of what caused the downtime can enable the engineers to prevent future re-occurrences.

By installing the alarm and events collector, Kestrel coal reportedly has been able to increase its plant profitability and safety.

The business benefits Kestrel Coal achieved included:

  • Improved safety
  • Reduced production losses
  • Reduced potential involving equipment damage during upsets, shutdown and start-up
  • Minimized environmental upsets
  • Reduced equipment wear and failure
  • Fewer incidents