A new boiler master concept combines the benefits of throttle pressure firing (return to the throttle pressure setpoint) and drum pressure firing (dynamic stability). It consists of a throttle pressure controller (reset action plus feed-forward signal) working in parallel with a drum pressure controller (proportional and derivative action). Because they are separated in the time domain, the controllers do not interact. This approach has proven particularly effective at stabilizing the operation of boilers equipped with low-NO
x burners.
The new arrangement is much more stable than traditional throttle pressure firing. It has been successfully deployed on several units and should be of great interest to owners of subcritical coal-fired units because it offers a quick and inexpensive solution to pressure stability problems. Following a discussion of the new boiler control strategy, this article presents three studies detailing its installation at four coal-fired units owned and operated by the Kentucky Utilities (KU) subsidiary of E.ON U.S.: The 495-MW Unit 3 of E.W. Brown Generating Station; the 75-MW Unit 3 of Tyrone Generating Station; and the 75-MW Unit 3 and 100-MW Unit 4 of Green River Generating Station. Coal-fired plants produce about 95% of Kentucky's total generation.
Time for a change
The strategy for the new boiler master combines several well-known approaches to control developed by Burns and Roe and others over the past 20 years. In 1985 we thought we had found the perfect solution with the Direct Energy Balance (DEB)-400 control strategy. However, problems quickly developed. After Leeds & Northrup went out of business, DEB-400 could not be replicated easily or legally on other distributed control system (DCS) platforms.
Burns and Roe began searching for a simpler firing rate strategy that would:
- Use drum pressure as a process variable, as a way of capturing most of the DEB benefits.
- Be portable to any DCS platform.
- Be easy to tune, reliable, and inexpensive.
Something old, something new
The idea of using the boiler drum as a calorimeter has been practiced commercially since at least 1957. Basically, this approach uses the derivative of the drum pressure as an excellent real-time indicator of the required heat input to the boiler (see box).
Going even further back, prior to the development of superheaters, all boilers were controlled by drum pressure. The boiler master of old was basically a proportional plus integral (P+I) drum pressure controller, and it worked very well.
Marine boiler control systems built in the 1950s are a good example of robustness. Most were pneumatic systems with a P+I drum pressure controller assisted by a steam flow feed-forward signal. The boiler master drove the fuel oil control valve directly, without the benefit of a fuel flow measurement. There was no gain compensation for the number of burners in service. All firing rate errors had to go through the process (that is, drum pressure upset). It was a fairly crude setup, but, nevertheless—because of the robustness of the drum pressure signal—this arrangement supported an amazing ramp rate of over 200% of the unit's maximum continuous rating (MCR)/minute.
There's a good reason why most boilers are still fired by throttle pressure: Throttle pressure is the final, visible product. It is what impacts steam turbine performance. Drum pressure is an intermediate result; the control system has to stabilize and maintain throttle pressure as constant as possible.
Throttle pressure equals drum pressure minus piping losses (losses vary as a square of the steam flow). Ideally, dynamic stability is maintained using drum pressure in real time while applying steady-state correction for throttle pressure. Dynamic stability is not hard to achieve when using drum pressure control. The problem is how to implement the necessary correction for throttle pressure without introducing dynamic instability.
Email
Comments
Print
Reuse
