How Can We Reach SIL 2 With Manual Intervention?
Readers help a reader solve this control problem. Next month's problem: How Can We Control Fly Ash Resistivity?
Gary Law, Product Manager
Emerson Process Management, www.easydeltav.com
Factor in the Basic Process Control System
The first step would be to identify the SIL requirement for the natural gas compression station. There are several means to do this including cause and effect matrices, standards such as EN 1050 (Safety of Machinery-Principles for Risk Assessment), and ANSI B11.TR3 (Risk Assessment and Reduction). This involves analyzing the hazard using risk parameters that include extent of damage or consequences, which ranges from minor injury to fatalities. Then the probability of the hazard occurring is evaluated. This could involve elements such as exposure time, occurrence probability, and hazard avoidance/mitigation. Once the appropriate SIL level is determined, the appropriate SIL reduction measures can be applied.
...Many companies perform a Levels of Protection Analysis (LOPA) to determine what abatement measures are possible. IEC 61511-1, paragraph 9.4, provides examples of typical risk reduction measures and methods. These protection levels include the basic process control system (BPCS), prevention measures such as alarms with operator interaction, and SISs. An approach could be to claim a risk reduction for the BPCS, another for the SIS and operator corrective action.
...Each of these measures needs to be independent. The SIS, for example, would have its own logic solver, sensors, and valves and would be programmed to automatically perform an emergency shutdown when required. Rockwell Automation's ControlLogix has been certified for use up to SIL 2 as an SIS for emergency shutdown.
Art Pietryk, Program Manager, Safety & Automotive
Rockwell Automation, www.ra.rockwell.com
How Can We Control Fly Ash Resistivity?
The electrostatic precipitators in our coal-fired power plant flues cannot operate at peak efficiency unless fly ash resistivity is controlled to an optimum level. Unfortunately, there are considerable lags between fly ash resistivity changes and changes in precipitator efficiency. To compound problems, the relationship between fly ash resistivity and precipitator efficiency is nonlinear. Traditional PID control does not seem to work, and we would like to consider using artificial intelligence. What other control schemes might be effective?
Please send us your comments, suggestions, or solutions to this problem. Solutions will be published in our November issue, but we need your responses by October 6 to meet our editorial production deadlines. If you have a solution to this problem, or a problem of your own that you would like to pose to other colleagues, e-mail, fax, or mail it to: CONTROL Problem Solving Ideas, 555 W. Pierce Rd., Suite 301, Itasca, IL 60143; Fax 630/467-1124; firstname.lastname@example.org