Seeking advice on the PID automatic control of the denitration system
Thread Content
I have a question I’d like to ask. Our factory has a denitration control system that adjusts the frequency of the pump to change the amount of ammonia used, thereby removing nitrogen oxides from the flue gas. I now want to go for an automated system, but the nitrogen oxide measurement points are located far away in the flue gas reaction zone; as a result, the nitrogen oxide levels have already changed by the time a measurement is taken, resulting in a delay of about 2 minutes. Is pure PID control useful?
The Smith predictor seems to not work either
It probably won’t work. What does two minutes mean? The flow rate of gas is very high; it’s possible to model this in software, but it still won’t work. It’s best to conduct actual measurements on-site. A two-minute delay won’t change the results of your measurements, unless you can use algorithms to estimate the values from two minutes ago
It is recommended to take a photo of both the trend of nitrogen oxide measurement values and the trend of ammonia flow rate to see the specific response; there are experts on the forum who can help you determine it.
It is theoretically possible to eliminate pure lag through d
The differential term d cannot eliminate pure lag.
Feedforward at the boiler outlet, feedback at the reactor outlet, and then add trend analysis? It shouldn’t be delayed by 2 minutes, right? Sampling time + data analysis + total pump response time: 2 minutes? The main contradiction should lie in the sampling time