May I ask what control laws are currently used in DCS control?

Most DCS systems have basic control functions; advanced controls such as those mentioned by the original poster, like optimal control, require investment. If you’re skilled enough, you can develop your own programs to create the control algorithms you need

I am currently working on improving the algorithm, so I need to first understand the control algorithms that are commonly used in DCS systems before I can make any changes. A control algorithm that has not been used in DCS must be employed in order to compare it with the original algorithm and demonstrate its advantages.

I understand what you’re thinking. I’ve worked with advanced control for a few years, so I have some personal insights on this topic. Most advanced control methods can be divided into two categories: 1) PID optimization. Such systems can evolve into more complex control mechanisms, and self-tuning functions can also be implemented. 2) Modeling – either by creating fixed models or, through self-tuning, dynamic models. Of course, there seems to be another method that isn’t based on PID; I forgot what it was, haha. It’s a rather primitive approach. Oh, right, there’s also an advanced method based on a software company from abroad – SixNet. I haven’t worked with this stuff for a long time, so I’m not sure if it will be helpful to you. I typed all this because it’s rare to see you online at such a late hour.

HONEYWELL’s TPS features the following control algorithms: PID (Pid) – conventional PID; PID with Feedforward (PidFf) – PID with feedforward; PID with External Reset Feedback (PidErfb) – PID with external reset feedback; Position Proportional (PosProp) – position proportional control; PID Position Proportional (PIDPosPr); Ratio Control (RatioCtl) – ratio control; Ramp Soak (RampSoaK) – ramp control; Override Selector (ORSel) – override selection control; Auto Manual (AutoMan) – automatic/manual switching; Incremental Summer (IncrSum) – incremental control; Switch; Null. It also includes the following calculation algorithms: Data Acquisition (DataAcq); Flow Compensation (FlowComp) – flow compensation; Middle of 3 Selector (Mid0f3) – selection from three values; High Low Average Selector (HiLoAvg) – selection of high, low, and average values; Summer (Summer) – adder; Variable Dead Time with Lead Lag (VDTLDLag) – variable dead time with lead and lag elements; Totalizer (Totalizr) – accumulator; General Linearization (GenLin) – general linearization; Calculator (Calcultr) – calculator. This post was last edited by liuei on 2009-3-14 at 23:49

Well, PID is also a model-free control method. I’m now planning to use a different model-free control law for control purposes. Since there are no actual devices available, I still need to create my own models and generate output data; then, using these output data, I will conduct research on model-free control. Well, haha, it’s still a bit tough.

I really don’t understand this. Without the equipment and actual hands-on experience, it is indeed quite difficult to understand this. I wonder if the moderators or liuei can explain it in simpler terms.

Those who haven’t studied HONEYWELL may not be able to understand it. I’ve sent a document to your email; please read parts 7 and 8 carefully!

Thank you. I am indeed engaged in pure theoretical research at the moment, but there’s nothing I can do about it. Of course, I still prefer practical things.