Method for determining the forward/reverse action of a regulator

Thread Content

1. Method for determining the positive/negative action of a regulator: When switching the control system to automatic mode, once the control strategy has been determined and the positive/negative action of the regulator is known, the most crucial aspect is how to tune the P, I, and D parameters. Based on years of field experience, I will discuss how to adjust these parameters in a control system; please use this as a reference in your own projects. Before setting the P, I, and D parameters of the control system, it is necessary to ensure that the closed-loop control system operates with negative feedback, that is, Ko*Kv*Kc > 0. Regulation target Ko: When the valve or actuator is opened and the measured PV increases, then Ko > 0 ； Conversely, Ko＜0 ； Adjusting valve Kv: If the valve is of direct-acting type (operates when air or electricity is applied), then Kv > 0 ； If the valve has a reverse action (air-operated, electric-operated), then Kv＜0 ； The sign of Ko and Kv is determined by the process object and production safety. Based on the signs of Ko and Kv and the condition that Ko*Kv*Kc > 0, we can determine the sign of Kc ； Regulator Kc: If Kc > 0, the regulator is reactive ； If Kc < 0, the regulator is positive acting ； The software configuration must be set correctly; before device commissioning, startup, and the tuning of P, I, D parameters, it is essential to check that the regulator’s forward/reverse action is proper and accurate. 1. Before setting the P, I, and D parameters of the control system, it is necessary to ensure accurate measurements and smooth valve operation ； 2. When tuning the P, I, and D parameters of the control system, it is necessary to inform the users in advance and ask them to pay close attention to the production conditions in order to ensure safe operation ； 3. When setting the P, I, and D parameters of the control system, start with automatic mode before switching to cascade control; start with the secondary loop before the primary loop, with the secondary loop having a coarser tuning and the primary loop a finer one. On the CRT at the operation station, open the tuning screen for the regulator, change the setpoint SP or the output value OP to generate a step signal that is acceptable within the process parameters, observe the changes in the measured value PV as well as the trend chart, and continuously adjust the PID parameters – often repeating this process several times until stable control is achieved. In practice, it is generally sufficient to achieve a first-order characteristic that satisfies the process requirements. II. Pre-set values for empirical PID tuning parameters For media that are fluids (gases, liquids), the reference values for empirical PID tuning parameters are as follows. (It is best to set these values in the software configuration before leaving the facility; fine-tuning can be done on-site or the values can remain unchanged): 1. For flow rate control (F): Generally, P=120～200%, I=50～100S, D=0S ； For the anti-surge system: generally P=120～200%, I=20～40S, D=15～40S ； 2. For pressure adjustment (P): generally P=120～180%, I=50～100S, D=0S ； For the venting system: generally P=80～160%, I=20～60S, D=15～40S ； 3. For level control (L): 1] For large containers (tanks with a diameter of 4 meters and a height of over 2 meters): generally P=80～120%, I=200～900S, D=0S ； 2] Medium-sized containers (towers with a diameter of 2–4 meters and a height of 1.5–2 meters): generally, P=100–160%, I=80–400S, D=0S ； 3] Small containers (towers and tanks with a diameter of 2 meters and a height of less than 1.5 meters): generally P=120～300%, I=60～200S, D=0S ； 4. Temperature regulation (T): Generally, P=120～260%, I=50～200S, D=20～60S ； The above parameters are empirical and not absolute. Furthermore, in practice, sometimes problems with the process object or valve (positioner) of a control system can be overcome by adjusting the PID parameters, thereby enabling automatic operation. Automating requires patient observation and continuous adjustment. The key to enabling automation in practice is that the valves (positioners) and actuators are reliable and can operate flexibly. In a cascade control system (for example, one with 2 controllers), the entire inner loop (the secondary controller, whose Ko1*Kv1*Kc1 > 0) is equivalent to Kv of the main loop, and it is always positive. Result of PID parameter tuning: By observing the curve, a first-order characteristic is generally sufficient (although theoretically, a second-order damping characteristic is desired). III. Precautions for activating automatic circuits 1. Basic principles: When the device is in operation, activating the automatic circuits must ensure the stable operation of each section; major parameters should not experience significant fluctuations, and parameters such as pressure, level, and temperature of other auxiliary equipment should also not change to such an extent as to affect the normal operation of the device. 2. Coordination with operators of the same process: The activation of the automatic circuit is part of the commissioning work for automation upgrade projects. If our team is responsible for setting up the automatic circuit, we should first explain to the user’s operators what our tasks will be, how the process operators need to cooperate, what the implications will be, and how to handle any unexpected situations. After the debugging is complete, we should inform the process operators. Before the automatic loop is put into operation for the first time, process operators should be asked to adjust the operating conditions of that section to a relatively stable state as much as possible. 3. Specific precautions for the control system: (1) The configuration of all automatic circuits should be thoroughly tested before being put into use. If the configuration is altered on-site, the signal flow and logic of the configuration should be carefully checked before switching to automatic mode; attention should also be paid to the timing aspects of the switching logic in the signal switching section. The configuration should include a simple logical element at the outlet of the automatic circuit to the field that allows for manual intervention, so that in the event of a configuration error, the automatic circuit can be manually stopped from operating in the field. (2) When starting automation, the proportional band and integral time values of the PID module can be increased first; the upper and lower limits of the PID module’s output should be set within an allowable range near the module’s current tracking output value, while the rate of change of the PID module’s output should be reduced. After enabling automation, check whether the operating direction of the PID module is correct and whether the changes in the input deviation of the PID module remain within the normal range. Once this is confirmed, gradually lift the various output restrictions on the PID module to restore its normal function, and then adjust the parameters of the PID module according to the desired control quality. 4 PID Parameter Tuning Methods 1. Basic Knowledge In automatic control systems, E=SP-PV. Here, E represents the deviation, SP represents the given value, and PV represents the measured value. A positive deviation occurs when SP is greater than PV; otherwise, it is a negative deviation. 1) The action of proportional control is proportional to the magnitude of the deviation ； When the proportionality is set at 100, the output of the proportional action responds to the error in a 1:1 ratio based on their respective range limits. When the proportionality is 10, it operates at a ratio of 10:1. That is, the smaller the proportionality, the stronger the proportional action. Excessive proportional action can cause oscillations. Being too weak leads to proportional understeer, resulting in too many fluctuation cycles during the system’s convergence process and a too small damping ratio. Its function is to stabilize the parameter being adjusted. 2) The action of the integral control mechanism is proportional to the time integral of the error. That is, as long as there is an integral effect of the deviation, there will be an output. It serves to eliminate residual errors. An overly strong integral action can also cause oscillations, while a too weak one will result in residual errors in the system. 3) The action of the differential control is proportional to the rate of change of the deviation. Its effect is to prevent any changes in the parameter being adjusted, serving as a form of lead regulation. It works very well for objects with large latency. But it cannot overcome pure lag. Suitable for temperature regulation. Using differential adjustment can shorten the convergence time of the system. An excessively long differential time can also cause oscillations. 2. The empirical method is the most widely used tuning method for simply adjusting systems; it is a trial-and-error approach. It is achieved through preset parameters and repeated trial and error. The default values of the parameters should be determined based on the characteristics of the object and the range of the instrument. The PID parameters with a large measurement range should have their effect enhanced appropriately. The general ranges for the four types of adjustable parameters are as follows: The critical proportional gain method involves putting the system into automatic mode using pure proportionality, with the integration time set to its maximum value and the differentiation time set to 0. Gradually reduce the gain until the system just begins to exhibit amplitude-modulated oscillations; note down the gain value Pbc and the oscillation period Tc at this point, and then calculate the PID gain and integral time using the following formula: P=2.2Pbc ； T=0.85Tc. The actual situation may go beyond this range. For systems with large pure time lags and time constants, the critical proportional gain method is not suitable for using MACS PID, as it is difficult to determine the Pbc value.