Reset action

Proportional-action governor with reset is a governor with inherent regiilation so that the momentary output is proportional to input change, and subsequently a reset action initiated by the output acts on the speed changer or its equivalent to make the settled regulation less than the inherent regulation. 

By adding the reset action to the proportional action the controller produces a larger output for the given error signal and causes a greater adjustment of the control valve. This causes the process to come back to the setpoint more quickly. Additionally, the reset action acts to eliminate the offset error after a period of time. 

Figure 31 demonstrates the combined controller response to a demand disturbance. The proportional action of the controller stabilizes the process. The reset action combined with the proportional action causes the measured variable to return to the setpoint. The rate action combined with the proportional action reduces the initial overshoot and cyclic period. 

Clock resets When a transition is taken, a clock reset action d = 0 attached to it simply assigns the value 0 to the clock c,. Note that the clock is not stopped, i.e., the clock rate c= 1 remains unchanged. 

Where ESD and DCS systems are provided, they should be functionally segregated such that failure of the DCS does not prevent the ESD from shutting down and isolating the facilities. Alternatively, failure of the ESD system should not prevent an operator from using the DCS to shut down and isolate the facility. There should be no executable commands over the ESD-DCS communications links. Communication links should only be used for bypasses, status information and the transmission of reports. Confirmation of ESD reset actions can be incorporated into the DCS but actual reset capability should not. 

As shown by Bristol (72) for controllers with heavy reset action, this measure has very interesting properties. Input/output pairs are selected for those Pij approaching 1. A negative element in Pij indicates instability or non-minimum phase behavior, so a glimpse of the dynamics can be obtained from Bristol s method. 

Initially use proportional-only controllers in all loops except flow7 controllers, where the normal tight tuning can be used K = 0.5 and T = 0.3 minutes). Set the gains in all level controllers (except reactors) equal to 2. Adjust the temperature, pressure, and composition controller gains by trial and error to see if you can line out the system with the proposed control structure. If P-only control cannot be made to work, PI will not w7ork either. When stable operation is achieved, add a little reset action to each PI controller (one at a time) to pull the process into the setpoint values. 

The reset time, which is user-adjustable, can range from 0.05 seconds to 80 minutes or more, depending on controller design. The reset time constant, when converted to frequency l/2(Tjj) Hz (where Tjj is the reset time in seconds), determines the frequency where the reset and proportional response characteristics of the controller merge (see Fig. 8-64h). Tuning the reset achustment on the controller moves the reset frequency to the left or right along the frequency axis and thereby affects the reset action of the controller. 

The Integral mode is sometimes referred to as reset because it continues to take action over time until the error between measurement and setpoint is eliminated. The parameter to specify this action is Integral time, which can be thought of as the length of time for the controller to repeat the initial proportional response if the error remained constant. Note that as this parameter is made smaller, the reset increases as the control action is repeated in a shorter period of time. Some controllers use an alternate parameter, Reset, that is the reciprocal of Integral time and is referred to as repeats/unit time. This latter approach is perhaps more intuitive in that as the Reset parameter is increased, there is more reset action being applied. 

We have just described proportional (P) control and proportional plus derivative (PD) control. Integration can be added to a controller, which not only gives it reset action, but also can exacerbate instability. There are proportional plus integral (PI) and proportional-integral-derivative (PID) controllers. These classical types are used where the system dynamics (the Plant) are well defined. 

In the second example discussed in the next section, a more complex process is considered in which reset action is needed in both the normal and the override controller, and the integral tuning constants are fairly large. 

The integral (reset) action in the controller is primarily for eliminating the offset error at steady state (Equation 8.2) ... 

Some reset screws are calibrated in repeats per minute. For these controllers, one turns up the screw to obtain more reset action. Level control loops are inherently sensitive to reset and can become unstable with small amounts of reset. 

Set reset to the maximum practical setting (minimum reset action). 

The main purpose of integral action is to eliminate offset. Sometimes called reset action it continues to change the controller output for as long as an error exists. It does this by making the rate of change of output proportional to the error, i.e. 

There will be occasions when the controller saturates. For example a flow controller may encounter a hydraulic limit so that, even with valve fully open, the SP cannot be reached. The integral (or reset) action will respond to this by continually increasing the output but, because the valve is fully open, will have no affect on the flow. This is reset windup. Windup should be avoided because, if the process constraint is removed - for example by starting a booster pump, there will be a delay while the controller removes the windup and can begin actually closing the valve. This is resolved by keeping the permitted output range as narrow as possible, typically —5 to 105 %. 

The loop phase was found by first selecting 6.5. sec as the natural iieriod reset action must then contribute 2S° to bring the total to 180°.. i different value of reset time would change r .) The gain contribution of reset at that phase angle is 1.11. Notice that all elements contribute some idiase lag, but only those whose phase lag ai)proaches 45° affect the loop w i n noticeably. 

When a proeess is shut down by the closing of hand valves, reset action begins to force the proportional band of the controller upward to its... 

A logical solution to the reset-windup problem is to add enough intelligence to the controller to make it aware of a shutdown condition. This is done by placing in the controller s reset circuit a switch energized by the output. Whenever the output exceeds 100 percent, the switch disables the reset circuit, leaving a proportional (or proportional-plus-deriva-tive) controller. In the absence of automatic reset action, a bias must take its place. Because this bias equals the output of the controller at zero deviation, it is ordinarily adjusted in relation to the expected process load. For this reason it is sometimes called the preload setting. 

The question often arises whether proportional, reset, and derivative are really the best control modes for every application. For the easier-to-control processes, their use can be justified. A single-capacity process and some two-capacity processes need only narrow-band proportional action. Derivative is of great value in processes with two or three capacities. But for the more difficult processes, it has been found that reset action is essential. 

Proportional control is obviously insufficient, as it was with pure dead time, so reset action is necessary. To facilitate identification of the influence of a sampling element, a process consisting of pure dead time and a gain of unity will be selected. The controlled variable will then follow the manipulated variable one dead time later. Figure 4.20 shows the first case, where = 0. 

Intervals that are too short also pose a problem. The reset component In the Incremental control algorithm Is At/R. If At is very small with respect to reset time, this component Is subject to truncation or round ing-off Suppose that the word length In a given DDC computer is limited to 5 decimals. A deviation that when multiplied by At/R results In a product less than 5 X 10" will not be acted upon. As an example, let At = 1 sec and R = 50 min. The minimum error causing reset action will be... 

If limit cycling and offset are both unacceptable, a two-mode controller can be added to the loop. This controller would actuate the three-state device which in turn drives the valve, as shown in Fig. 5.12. Reset action will keep driving the output of the controller out of the dead zone until the error is reduced to zero. Only then will the loop reach a steady state. Proportional action is necessary for stability, for without it, the double integration of reset and motor would cause an undamped cycle. The availability of a proportional band adjustment eliminates the need for an adjustable dead zone, since the two effects are similar. 

Zero offset requires a controller with infinite gain, in the steady state. An integrator supplying "reset action is sufficient to satisfy this criterion. 

A sensible approach is to use a linear controller in this region, with gain adjustable by the proportional band. Because the gain is expected to be low, reset action is also required. Derivative is also recognized as... 

---