ARTICLES

Download PDF

1    2    3    4

In today’s competitive environment every company has to maximise the utilisation of its facilities and equipment. A key component of that strategy is the use of process control technology in its various forms – pneumatic, analogue and digital control. These three control technologies perform what is known as conventional or regulatory control and as such form the basis from which more advanced algorithms and optimisation can be conducted. Figure 1 shows the ‘hierarchy of control’ illustrating how each level of a control system and associated benefits depend on the reliable operation of the level below.

The remaining method to operate a facility is with manual or ‘open loop’ control where the operators make changes to the process by physically changing the position of a valve or operating conditions of a pump. Despite the fact that almost no facilities currently operate this way, the economic benefits of control are based on this most basic method of operating a facility.

Figure 1: Hierachy of Control

Control Hierarchy

As illustrated in Figure 1, the ability to control is entirely based on the ability to measure and influence the pressure, temperature, level, composition or flowrates of the process. These field devices must be repeatable so that the same response can be assumed. If the device is not accurate, it is still possible to control because the engineer can introduce a bias to ‘calibrate’ the device and provide the reading or response desired.

The devices most commonly used to alter process conditions are control valves and/or variable speed pumps that change the rate of flow of the process. Variable speed pumps change the rate of flow by altering the rate of rotation while control valves alter flowrates by changing the resistance in the process flow when they open or close. The selection of the control valve is a function of the process conditions and type of change to be made to the process though the two must be matched to ensure reliable operations. A common item overlooked in process automation is that all valves are mechanical devices, which means they are subject to wear and some of the linkages of the device can also become ‘loose’ over time. The same is true to some degree for pumps.

Regulatory Control

Regulatory control forms the basis for all other closed loop control and is responsible for maintaining or returning a process to a steady state operation under the influence of either set point changes (changes to the process made by the process operator) or perturbations to the process (a change to the process operating conditions due to a change in composition or equipment). The most common regulatory algorithms are proportional integral derivative (PID) (alternately called gain reset rate), proportional integral (PI) and proportional derivative (PD) or straight proportional control. Selection of the regulatory control algorithm to be used is a function of the process and its associated dynamics. For example, processes with a large lag time such as temperature are often controlled with a PI controller since the addition of derivative as in PID will result in unstable oscillations.

Significant levels of research have been conducted on regulatory control for many years, though more recently most investigations have been on the'higher' levels of the control hierarchy. Most of the work in this area is based on findings of the 1930s and 1940s when the first closed loop control equipment became widely available.

Typical applications for regulatory control will be the single or cascade loops such as the ones used to maintain level in a vessel or flow in a process line or temperature.

1    2    3    4