1.0 Experiment 1: Effect of Controller Gain (P-value)

Figure 1: Simulink setup for Experiment 1

Graph 1: Graph from value P = 0.005, I = 0.01 and D = 0

Graph 2: Graph from increasing value of P = 0.01 and fixed value of I = 0.01 and D = 0

Graph 3: Graph from increasing value of P = 0.02 and fixed value of I = 0.01 and D = 0

Graph 4: From combining Graph 1-3 to compare the result. From the simulation, the value of P in PID causes the graph change. Increasing the value of P makes the graph become more stable. (Graph 1 – Blue, Graph 2 – Green and Graph 3 – Red)

2.0 Experiment 2 – Effect of Integral Time (I-value)

Figure 2: Simulink setup for Experiment 2

Graph 5: Graph from value P = 0.005, I = 0.01 and D = 0

Graph 6: Graph from decreasing value of I = 0.001 and fixed value of P = 0.05 and D = 0

Graph 7: Graph from decreasing value of I = 0.0005 and fixed value of P = 0.05 and D = 0

Graph 8: From combining Graph 5-7 to compare the result. From the simulation, the value of I in PID causes the graph change. Decreasing the value of I makes the graph become more stable. (Graph 5 – Blue, Graph 6 – Green and Graph 7 – Red)

3.0 Experiment 3 – Effect of Derivative Time (D-value)

Figure 3: Simulink setup for Experiment 3

Graph 9: From combining the graph from the experiment to compare the result. From the simulation, the value of D in PID causes the graph change. Increasing the value of D makes the graph become more stable. (Blue – D = 0, Pink – D = 0.5 and Red – D = 1)
4.0 Experiment 4 – Effect of Deadtime

Figure 4: Simulink setup for Experiment 4

Graph 10: Graph from value P = 0.02, I = 0.01, D = 0 and time delay 5

Graph 11: Graph from value P = 0.02, I = 0.01, D = 0 and time delay 7

Graph 12: Graph from value P = 0.02, I = 0.01, D = 0 and time delay 9

Graph 13: From combining Graph 10-12 to compare the result. From the simulation, the value of time delay in PID causes the graph change. Increasing the value of time delay makes the graph become more unstable. (Graph 10 – Blue, Graph 11 – Green and Graph 12 – Red)

5.0 Discussion
The PID control algorithm is made of three basic responses, Proportional (or gain), integral (or reset), and derivative. Proportional control applies an effort in proportion to how far you are from the set point. Integral control tries to even out the difference of the time spend on both sides of the line .Derivative acts as a brake or dampener on the control effort. The more the controller tries to change the value, the more it counteracts the effort.
PID controller also has all necessary dynamics which is fast reaction on change of controller input (D mode), increase in control signal to lead error towards zero (I mode) and suitable action inside control area to eliminate oscillation (P mode). In this experiment, the system use digital control system. In digital control system in simulink, adjust the sample time equal to 1 whereas in analog control system the sample time is equal to -1 and the save format for both systems is array. In a digital control system, an integral control systems computes the error from measured output and user input to a program and integrates the error using some standard integration algorithm, then generates an output control signal from the integration. For analog control system, it is mean by an integral control system integrates the error signal to generate the control signal. If the error signal is a voltage and the control signal is also a voltage, then a proportional control is just an analog integrator. Controllability is the ability of a process to achieve acceptable performance. Derivative in PID controller used for control the process in order to know how far PV, (measured process variable) from any instant in time. Other than that, the derivative action will stabilize the controlled response as shown in plotted graph, allowing higher gains and shorter the integral time constant to be used.
From Experiment 1, the integral and derivative value is remain fixed and by increasing the value of P which starting from 0.05, 0.1 and 0.2, the controller action become slower, which in turn slows down the process response. The graph also more stable when increasing the P value. Hence, it can be conclude that from Experiment 1, increasing the value for P in PID controller makes the system more stable.
From Experiment 2, the value for proportional and derivative time is remain fixed and only by decreasing the value of I from 0.01 to 0.001 and 0.0005, the controller action also become slower, which in turn slow down the process thus makes the graph more stable. Hence, it can be conclude that from Experiment 2, decreasing value for I in the PID controller makes the system more stable.

One of the effect adding and adjust the derivate action is that the overshoots on controlled responses can be restricted. As we know, the derivative action used to provide improvement in controlled responses not commonly used in the industrial process. Based on Experiment 3, we can conclude when there are increased of PV, so there are some changing in time. Hence, increasing the value of derivative time from 0 to 0.5 and 1 will make the graph more stable.
From Experiment 4, the PID controller must wait until the dead time has passed before it gets any feedback from the process. Dead time is the time taken for the PID to start getting any feedback from the process. Dead time will delay the time to take reading for PID controller. Hence, the increasing of dead time from 5 to 7 and 9 will increase the feedback from passed. The best thing to do for controlling this type of process is by reducing the dead time. Increasing the dead will produce the unstable graph as shown in Experiment 4. As suggestion, simply moving a probe closer to the valve may reduce the dead time.