Feedback Control

Experimental studies of feedback control schemes for laboratory and bench scale crystallizers have been based on the measurement of the suspension density in the fines removal line by manipulating the fines removal flow. The major drawback of this method is that the manipulation of the fines stream influences not only the number of crystals in the fines flow but also the cut-size of the baffle zone. Therefore the controller will in essence control the fines density removed from the system and not the fines density in the crystallizer.

An in-line Lasentec probe has been used to control a 1000 L DTB crystallizer producing potassium chloride using the fines removal flow as a process actuator. Problems identifying the optimal signal from the sensor and process disturbances affecting the CSD measurements decreased the efficiency of the controller.

The most complete study was one in which the fines removal rate was used as the process actuator to control the process. Figure 4 shows the general control scheme in which u is the process input on which the controller acts, in this case the fines removal rate t and y the process output. d and m are possible process disturbances and the measurement noise respectively.

An on-line particle counter, measuring the number of crystals in a predefined size range (60-100 |im), was used. The control equation is then shown in eqn [17]:

= Nf-Nf f,setpoint

Figure 4 Feedback single-input/single-output control scheme.

1. Stabilization of the process. A considerable reduction in the oscillations after start-up was obtained in closed loop compared with the open loop behaviour (see Figure 5). The severe oscillations resulting from the onset of a product classifier (wet screen) could also be suppressed by the controller (see Figure 6)

2. Set-point tracking. The set-point in the number of fines could be followed by the controller. The changes in the set-point for the number of fines also resulted in changes in the median crystal size of the crystal produced.

3. Disturbance suppression. A process disturbance introduced in the process by closing the product flow for 1 h was analysed. It was shown that open loop response on the disturbance was almost completely suppressed in closed loop operation of the process (see Figure 7).

The choice of the size range for the particle counter, which could be affected by the detector threshold and the settling velocity in the funnel used to discharge the particles from the crystallizer, is essential for the performance of the controller. It has been shown that when the detection size range is moved to particles below 40 |im the controller becomes unstable.

As an alternative to the particle counter, the reduced signal of the laser diffraction instrument was used. Similar results were be obtained (see Figure 6).

The error sk is the difference between the counter output and the set-point value. T is the sample time and Ti the integral or reset time. The following functions of the controller were tested.

Figure 5 (A) Open and (B) closed loop start up trend of the median crystal size. (Eek, 1995)
Figure 6 Open and closed loop trends in the yr, the X50 and the total crystal mass after the onset of a product classifier at a. Points b, c and d represent changes in controller set point and product flow rate respectively. (Eek, 1995.)
Solar Panel Basics

Solar Panel Basics

Global warming is a huge problem which will significantly affect every country in the world. Many people all over the world are trying to do whatever they can to help combat the effects of global warming. One of the ways that people can fight global warming is to reduce their dependence on non-renewable energy sources like oil and petroleum based products.

Get My Free Ebook


Post a comment