I am currently working on a glass feeder channel firing control system. The goal is to maintain a set temp for each of 6 zones. Each zone's glass temp is affected by the manifold temp of the previous zone the most. The manifold temp is affected by the flow of mixed gas measured with delta pressure across an orifice plate. The gas flow is controlled by an actuator that opens and closes via discrete outputs and gives position feedback. I have a 1769-L35E processor already in place doing some monitoring functions that is only using about 5% of the memory and plan to add this to that. My first thought is to use a cascaded pid loop for the control. I have tuned PID loops in the past, however have no experience with a cascaded loop. The idea would be to have the first pid set a target manifold temp for zone #2 based on the glass temp of zone #3. The second pid would set the target gas flow based on the target manifold temp. The third pid would set the target valve position based on target gas flow. The final pid is a position proportional that will set the discrete outputs to open or close the valve to maintain the target gas flow.
Am I overcomplicating this? The goal is to maintain the temp as tight as possible. Would it be worth it to try this with out the second and third pid's or will I get tighter control with the current setup?
Am I overcomplicating this? The goal is to maintain the temp as tight as possible. Would it be worth it to try this with out the second and third pid's or will I get tighter control with the current setup?