International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 07 Issue: 03 | Mar 2020 www.irjet.net p-ISSN: 2395-0072 © 2020, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 1675 Liquid Level Control Strategy using Fractional Order PID Controller Based on Artificial Intelligence Adnan Mukhtar 1 , Farhan Mukhtar 2 1 Electrical and Electronics Engineering Department, Amity University Uttar Pradesh, Noida, 201303, India 2 Automobile Engineering Department, Manav Rachna International Institute of Research and Studies Faridabad, Haryana, 121004, India ----------------------------------------------------------------------***--------------------------------------------------------------------- Abstract - At present, process control industries face serious problems related to the liquid level. These industries find it hard to maintain and regulate the liquid level so to attain efficient product quality because the level of the liquid needs to be maintained at a pre-defined rate. Without any interruption, the level of the liquid needs continuous monitoring and control that is only possible by designing a suitable controller or control system. The Proportional- Integral-Derivative (PID), Tilted-Integral-Derivative (TID) and Fractional Order PID (FOPID) controllers are designed based on artificial intelligence techniques to regulate the liquid level. This paper analyses and compares the liquid level control for a process by classical PID, TID and FOPID controllers. The simulation results show far better peak value (PV), settling time (ST), rise time (RT), steady-state error(Ess), Integral of Time Multiplied by Absolute Error (ITAE), Integral of Absolute Error ( IAE) and Integral of Square Error (ISE) with FOPID controller, thus ensuring overall higher efficiency. Key Words: AI, FOPID Controller, GA, ITAE, PID Controller, PSO, TID Controller 1. INTRODUCTION The Control system finds numerous applications in the field of process control. Regulating the height and temperature of the liquid in a liquid tank are some of the main problems encountered by these industries [5-6]. These industries find it very hard to regulate and maintain the height and temperature of the liquid. So to overcome these problems, process control industries need to adopt a suitable controller or control system that will regulate the height and temperature of the liquid [3], [4]. This paper mainly emphasizes one such problem, the liquid level control [1], [2], [4], [6]. There are numerous methods designed in Control systems particularly to resolve such problems [5]. This paper proposes three methods: the Proportional-Integral-Derivative controller (PID), Tilted- Integral-Derivative controller (TID), and Fractional Order PID controller (FOPID). PID is feasible, robust and its concept is simple. TID resembles with PID but the tilted gain replaces the proportional gain It is easy to tune than PID [7]. FOPID being an advanced PID controller is more robust and flexible than PID and TID because it has five gain parameters [9]. Moreover, it yields better output responses. These controllers are optimized using artificial intelligence (AI) techniques, Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). PSO is easy to understand, it can be easily incorporated, and it has robust parameters [2]. GA has a simple concept; better efficiency and its response are fast. In this work, the tuning of the parameters of the three controllers is done using GA and PSO. A comparison between the three controllers is done according to the peak value, settling time, rise time, steady-state error, Integral of Time Multiplied by Absolute Error (ITAE), Integral of Absolute Error (IAE) and Integral of Square Error (ISE). The entire work is carried out in MATLAB/SIMULINK. This work is structured as- section 2 defines the process modeling that comprises of the control valve, liquid tank, and model equation, section 3 explains the simulation results and the conclusion is discussed in section 5. 2. SYSTEM DESCRIPTION 2.1 Process Modeling The modeling process comprises the control scheme, control valve, and liquid tank. Fig. 1 shows the basic control scheme for the process liquid level. 2.2 Control Valve Fig -2: Simulink diagram of Control Valve