We operate Hitachi gas turbines and perform online water washing every two weeks. The procedure requires the turbine to reach baseload, with the inlet guide vanes (IGVs) above 77% for the wash to start or continue.

However, when we initiate the water wash command, the turbine reduces its load by 5% per the system’s logic, and there is a 40-second delay before the water wash valve opens. This unexpected load reduction has led to low-frequency events, which have, on multiple occasions, triggered plant equipment trips due to the low-frequency load-shedding scheme.

In my experience with other sites and GE gas turbines, I haven't encountered a similar logic requiring a load reduction before an online water wash. Can anyone provide insight into why this load reduction is built into the Hitachi turbine logic? Also, is there any way to override or modify this function?

 

@Zia Naqvi,

1) How many shafts do the Hitachi machines have—one or two?

2) Are the machines synchronized to a large grid or supplying power to loads independent of a large grid?

3) What kind of combustion system do the Hitachi machines have?

4) What control system do the Hitachi machines have?

5) Is there some kind of external Power Management System (PMS) that sends commands to the Hitachi machines to control the frequency and/or load of the Hitachi machines?

6) Are the Hitachi machines being operated in Isochronous Speed Control Mode or Droop Speed Control mode—or some kind of Isochronous load-sharing mode?

Lots of questions, I know, and the scheme you describe seems a little unusual; I can’t recall encountering anything similar in my forty-plus years of experience. But there’s a lot we don’t know about the specifics of your site and how the machines are being used and if there’s some kind of concern that we don’t know about.

Has anyone asked Hitachi or the control system provider about the reason for the logic? And recommendations for resolving your particular issues?

One division of GE is notorious for inventing new control schemes to replace tested and proven control schemes that usually result in knock-on effects like you are describing. Many of the control system programmers of that division anticipate problems that don’t actually exist or because of “similar” problems or complaints from other sites the concoct unusual solutions that aren’t fully developed or tested.

If these Hitachi machines were recently installed it’s likely there is some kind of overall plant operations manual that may offer some insight into why this scheme was implemented.

Help us to understand the site and the process and we might be able to offer some insight or advice. But my strong suspicion is that the Hitachi machine control system provider-programmer is going to need to get involved to come up with a permanent solution to your issue.