Page 90 - IJEEE-2022-Vol18-ISSUE-1

86 | Ali & Rashid

The estimation algorithm uses ???? sin(??????) and 2) Study the response of the system with PI control (manual
???? cos(??????) as regressors to estimate ?????????????? and?????????????? . tuning PI parameters) under factors effect (inlet feed and
For N frequencies, the algorithm uses 2N regresses. The opening valve).
computation assumes that the perturbation signal u(t) is
applied to a plant with zero nominal input and output. To 3) Study the system with PI control (auto-tuning PI
achieve this condition, the block subtracts from the measured parameters) under factors effect (inlet feed and opening
plant input and output signals their values measured at the valve).
start of the experiment.
Fig. 4: graphical presented for model.
IV. EXPERIMENTS AND RESULTS First Scenario: Without Controller

This section will concentrate on building systems in Simulink models of the tank are illustrated in Fig. 5. The
system has been provided with a constant reference level
MATLAB Simlink, and work will be based on simulations with a predefined peak value, as well as an input feed for the
separator tank, which we specified with a variable value
of real industrial systems and the results of numerous (m2). The performance of the system is presented in Fig. 6
(the curves illustrates both the desired set point level (blue
experiments, so we will first evaluate the system's closed- trace) and the simulated actual level (red trace) as functions
of time) and its characteristics are listed in Table 1. Clearly,
loop performance without any control, then change the results (rise time and overshoot values) show that the
system is not controlled or acceptable to manage the plant.
configuration with a new PI control and finally offered an Due to the system's primary attribute being non-linearity, it
was important to analyze the change in level using both the
auto-tuner method by adding new blocks to the system to integrator and the other system components. All of these
features have the potential to cause disruptions if they are lost
opportunistically optimize the system. or diminished for whatever reason.

initiated many experiments that tested control on tank

D5204 illustrated in Fig. 4 with its below specification:

Height: 5,200 mm tang to tang.

Dimeter: 1,600 mm

Total volume: 11.5 m3

Operation pressure: 25 bar
45 ?
Operation temperature:
Liquid density at 45 ?: 759 kg/m3

Liquid operation density for inlet stream: 650,52 kg/m3

Liquid operation density for output stream: 722,8 kg/m3

The Coefficient of level control valve: cvi=44.6

The pressure drop through the valve acting: 10 bar

Inlet liquid flow rate to tank m2: 600 Kmole/h

Valve opening in % yi: 50%

Desired level set point: 40%

From equations (3 and 5), the design of the system model
will be implemented by MATLAB software, in which
experiments will have divided into several scenarios:
1) Study the response of system without controller under

factors effect (inlet feed and opening output valve)

Fig. 5: System diagram without a controller.