Assessment H16A2
Instrumentation & Control
SUBMITTED BY:
__________
Part 1a - Control concepts used within a process
i) Control Terminologies and Concepts
Capacity and Stability
Capacity of a control system is termed as the fundamental limit (in bits) of the rate of dissipation
of uncertainty in that system, i.e. before becoming saturated or ineffective (Sahai, 2017). An
example of such system is a power transformer feeding an electric arc furnace. Assuming the
furnace is rated at a fixed current value, this sets a hard limit on the current that can be delivered
to the arc. Even if the electrode regulator or commands more power to increase arc intensity, it
cannot exceed the capacity of the transformer. System saturation may occur if attempted to push
the current beyond the limit.
Stability is termed as the ability of a system to gradually recover to its equilibrium state after it
is subjected to an initial excitation or disturbance (circuitbread, 2021). Oscillations are dampened
in a stable system over time, whereas an unstable system will amplify these oscillations. An
example of stability is the chemical process plant, where exceeding capacity may result in an
inability to control temperature, leading to thermal runaway. Engineers prevent such instability
by tuning the system, ensuring that temperature returns smoothly to setpoint after disturbances
like reactant addition.
Feedback and Error Handling
A Feedback Control System includes the measurement of input variable and calculation of its
difference from the output to attain a desired value (Daswani, 2025). The difference between
input and output is known as Error. The level sensor in a water tank is an example which uses a
feedback control system. It involves input from a sensor which measures the tank level. If the
level is too low, a pump increases flow until the desired level is restored.
Error Handling refers to the response of the system to the error signal, which is the deviation
between desired and actual outputs. This error is used to adjust the control effort. A PLC
compares the actual water level measured by the sensor to a predefined setpoint and calculates
the error. Based on this error, the PLC adjusts the opening of an inlet valve to either increase or
decrease the inflow.
PID Control and Tuning
PID (Proportional-Integral-Derivative) is a control method which is widely used to attain
desired system performance in the shortest possible interval. The Proportional (P) term
responds to the present error, generating an output proportional to its magnitude. The Integral
(I) term addresses any persistent errors or long-term deviations from the setpoint by
accumulating the error over time. The Derivative (D) term anticipates future changes in the error
by evaluating its rate of change (Smith, 2024).
Tuning PID parameters (Kp, Ki, Kd) is performed to balance responsiveness and stability. Poor
tuning may cause overshoots, oscillations, or slow response. In HVAC systems, PID controllers
maintain room temperature by adjusting heating/cooling based on temperature deviation.
A block diagram of PID is given below:
Figure 1: Block Diagram of a Typical PID Controller (Smith, 2024)
Transfer Functions
In control systems, the transfer function is defined as the ratio of the output's Laplace transform
to the input's Laplace transform, with all initial conditions set to zero. It simplifies the analysis of
linear time-invariant (LTI) systems by converting differential equations to algebraic equations in
the s-domain. Mathematically it is given as,
Ly ( t )
T ( S )=
Lu (t)
Where y(t) is the output of the system and u(t) is the input.
The transfer functions are useful in robotic arms, where the DC servo motor driving a joint is
modeled by the transfer function. This function considers the motor's electrical time constant,
mechanical inertia, and damping.
ii) System Tuning Methods and Techniques
Ziegler–Nichols Method
The Ziegler–Nichols tuning method is widely used for tuning PID controllers in industrial
systems. It includes two approaches:
Open-Loop (Reaction Curve) Method
In this method, a step input is applied and output is monitored to find Dead Time (L), Time
Constant (T), and Process Gain (Kp). These are then used to determine PID settings using
empirical formulas. It is simple to implement but it is not ideal for systems with large delays or
non-linearities.
Closed-Loop Method (Ultimate Gain Method)
This method involves disabling the integral and derivative actions and increasing proportional
gain (Kp) until the system output exhibits sustained oscillations. At this point, Ultimate Gain
(Ku) and Ultimate Period (Pu) are recorded. These values are used to calculate PID settings as:
Table 1: Finding PID Parameters using Closed Loop Method
Kp 0.6 Ku
Ti (sec) Pu/2
Td (sec) Pu/8
Part 2a - Control concepts used within a process
i) Scilab Simulation Results
A PID control system was developed using XCOS in Scilab software, based on the given
instructions. The image of the setup with an initial response is shown below:
Figure 2: XCOS setup for PID Control
The objective of this simulation is to design a feedback control system for the plant with the
transfer function:
4
G ( s )= 2
s +s
A unit step input was added to observe transient response characteristics such as overshoot,
settling time, and overall system stability. The XCOS model was constructed using essential
control blocks: a step function as the reference input, a PID controller, a transfer function block
to represent the plant, a summation block to calculate the control error, and a scope to visualize
the system output. The initial PID parameters were considered as, Kp = 1, Ki = 0, Kd = 0,
resulting in a system response with a noticeable overshoot. The system response was also not
stabilized in 10 seconds. This behavior indicated an underdamped system with insufficient tuning
to achieve a fast and stable response.
To improve the system performance, the PID controller was tuned using Ultimate Gain Method,
targeting an overshoot below 10% and a settling time of under 10 seconds. The optimized
parameters were set to, Kp = 7.0, Ki = 0.05, and Kd = 1.2. After applying these values in the
simulation, the system's response improved significantly. The maximum overshoot was reduced
to around 6.5%, and the output settled within 3 seconds. Additionally, the integral component
effectively eliminated steady-state error, while the derivative action helped suppress residual
oscillations, contributing to a smoother transient response.
The final simulation confirmed that the tuned PID controller successfully achieved the desired
control objectives: minimal overshoot, acceptable rise and settling times, and stable output
behavior under step input conditions, the resultant output of which is shown in the figure below:
Figure 3: Tuned response of the PID system
References
circuitbread. (2021, April 06). The Concept of Stability | Control Systems 3.1. Retrieved
from circuitbread.com: https://www.circuitbread.com/tutorials/the-concept-of-stability-
3.1
Daswani, K. (2025). Feedback in Control Systems | Overview, Types & Examples.
Retrieved from study.com: https://study.com/academy/lesson/feedback-control-system-
overview-types-examples.html#:~:text=A%20feedback%20control%20system
%20involves,to%20drive%20the%20correct%20output.
Sahai, G. R. (2017). Control Capacity. ∗Microsoft Research, Redmond.
Smith, G. M. (2024, June 19). What is a PID Controller? Retrieved from dewesoft.com:
https://dewesoft.com/blog/what-is-pid-controller
