Modelling and Controller of Liquid Level system using PID controller
Deign Gloria Jose1, Shalu George K.2 1PG student, Dept. of EEE, MBCET, Trivandrum, Kerala 2 Assistant Professor, Dept. of EEE, MBCET, Trivandrum, Kerala
---------------------------------------------------------------------***--------------------------------------------------------------------- Abstract— The control of liquid level in tanks is 2. MODELLING OF LIQUID LEVEL SYSTEM one of the basic problem in the process industries. The thesis work deals with design of PID controller for the Liquid Level The diagram of plant (the Liquid Level system) under System. System identification of LLS is done and modeled consideration is shown in the fig 1. The plant level through Empirical Zeigler-Nichols Tuning method. A process controller VLPA-101-CE is truly versatile and transfer function is obtained with first order system plus highly reliable standard computer based level delay. The system is controlled by PID controller and tuned process Controller. It is a self contained process and by using Zeigler-Nichols tuning method. The PID controller control equipment. The VLPA- 101-CE has miniature is implemented in MATLAB and then simulated in Simulink pump, level sensor for sensing and personal to test the output of the system with respect to input. The computer for controlling. output is obtained with less steady state error.
1. INTRODUCTION Development of Liquid Level System has become an unavoidable part in many industries due to the wide use of boilers in nuclear power plants and other liquid based production techniques. . In process control, level control is a common method. Hence the level control system must be properly controlled by the suitable controller. Fig 1: The Liquid Level System VLPA-101-CE Table 1 liquid level system specifications Pump Process tank Reservoir tank To analyze the systems involving fluid flow, it essential to divide flow regimes into turbulent flow Model Tullu 80 Material Acrylic Material Mild and laminar flow, according to the value of Steel magnitude of Reynolds number. If the Reynolds Speed 6500RPM 2 Capacity 7 number is greater than or about 3000 to 4000, then Capacity liters liters it is turbulent flow. And if the flow is laminar then the Reynolds number is less than or about 2000. And when Reynolds number is between 2000 and 3000 is called as transitional flow. In laminar flow, fluid flow PID controller is one of the most easiest and mainly occurs in streamlines with no turbulence. simplest controller that always been used in Systems involving turbulent flow are represented by industrial. There are several methods to obtain the nonlinear differential equations, while systems parameters for PID controllers such as trial and error involving laminar flow are represented in linear method, Cohen-Coon (C-C) method and Ziegler- differential equations. (The flow of liquids in Nichols (Z-N). The values of the parameters in the Industrial processes is often through pipes and tanks. controller determine the performance of system. In Such flow is often turbulent and not laminar.) this paper Ziegler-Nichols (Z-N) tuning method is In order to identify the behavior of a process, used. a mathematical description of the process has to be
developed. But usually, the mathematical model of 2.2 CAPACITANCE OF LIQUID-LEVEL SYSTEMS most of the physical processes is nonlinear in nature. On the other hand, most of the analysis like in The capacitance of a tank is defined to be the simulation and design of the controllers, assumes change in quantity of stored liquid necessary to cause that the process is linear in nature. In order to build a unity change in the potential (head). The potential this gap, the linearization of the nonlinear model is (head) is the quantity that includes the energy level needed. This linearization is always done with of the system. respect to a particular operating point of the system. This section illustrates the nonlinear mathematical behavior of a process and the linearization of the model. Consider a specific example of a simple process described in Fig.2. Capacitance (C) is nothing but is cross sectional area (A) of the tank.
Rate of change of fluid volume in tank = flow in – flow
out dV qi qo dt (4)
Since volume is (area x height)
Fig 2: Example of a physical process
From the above figure it is understood that qi is the d ( A h)
qi qo (5) inflow rate and qo is the outflow rate (in m3/sec) of dt the tank, and h is the height of the liquid level of the tank at any time instant. And also assume that the cross sectional area of the tank be A. In steady state dh A qi qo (6) condition, both qi and qo are same, and the height h dt of the liquid level of the tank will be constant.
2.1 RESISTANCE OF LIQUID LEVEL SYSTEM And cross sectional area can be replaced by capacitance The resistance for liquid flow in such a pipe or restriction is defined as the change in the level dh C qi qo (7) difference to a unit change in flow rate; that is, dt Where the resistance R may be written as change in level difference(m) Resistance (1) change in flow rate (m3 / s) dH h R dQ q0 (8) dH Then rearranging the equation (8) we get R (2) dQ h q0 (9) R
Substitute equation (9) in equation (7), we get processes the process reaction curve is an S-shaped curve.
The procedure to obtain the reaction curve of a plant
dh h in an open loop experiment as follows: C qi (10) dt R 1) When the plant is in open loop condition, take the plant manually to a normal operating condition. After simplifying above equation the equation (10) 2) Say the plant output settles at y(t) = y0 for a becomes constant plant input u(t) = u0. 3) At an initial time t0, apply a step change to the dh plant input from u0 to u∞ (this should be in RC h Rqi (11) dt the range of 10 to 20% of full scale). 4) Record the plant output until it settles to the new operating point. Taking Laplace transform considering initial 5) Thus the curve obtained is called process conditions to zero reaction curve. Which is shown in figure 3.8, m.s.t. stands for maximum slope tangent. RCsH ( s) H ( s) RQi ( s) (12)
The transfer function can be obtained as
H ( s) R Qi ( s ) ( RCs 1) (13)
Where,
C=cross sectional area of the tank =πr2 cm2 =35πcm2
Radius of the tank is 6cm.
In the middle of the process tank consists of a tube
with radius 1cm. Fig 3 Plant Step Response 3. SYSTEM IDENTIFICATION
The determination of the dynamic behaviour
6) Where the process gain value K can be of a process by experiment is called process calculated as identification. The method used in modelling of liquid level system is the system identification is yo y Empirical method in which the experimental input- K (14) output data is used. In this section, the transfer u0 u function model using process reaction curve method is discussed for which the input and output data are generated from the real time system response.
Open-loop identification is widely used in Where,
the industry. In open loop step testing, a step change in input is applied to the process which will produce y∞ and y0 are outputs a corresponding response. It is called process u∞ and u0 are inputs reaction curve. In the chemical industry, for many
A very useful empirical tuning formula was proposed by Ziegler and Nichols in early 1942. The tuning formula is obtained when the plant model is given by a first-order transfer function model with a pure time-delay. In real-time process control systems, a large variety of plants can be modeled approximately. If the system model cannot be physically derived, experiments can be made to extract the parameters for the approximate model. Many industrial processes show step responses with pure a periodic behaviour according to Figure 3.9. Fig: 4. Sketches of the responses of a first–order plus This S-shape curve is characteristic of many high- delay model order systems and such plant transfer functions may be approximated by the mathematical model can be The transfer function model using process reaction expressed as curve method is shown here:
K (15) G( s) e Ls H ( s) 1 e s (Ts 1) Qi ( s ) (5s 1) (16)
This contains a 1st order delay element and a dead
time Where, Where, L K = 1, L = 1 and T = 5 and aK T K = process gain 4. EMPIRICAL ZIEGLER-NICHOLS TUNING T = process time constant , FORMULA
L = dead time of the process For instance, if the step response of the plant model can be measured through an experiment, the output 3.2 TANGENT METHOD signal can be recorded as sketched in Fig.3, from 1. Obtain the step response experimentally. which the parameters of K, L and T can be extracted.
2. Draw a tangent at the inflection point. The S-shaped reaction curve can be characterized by two constants, delay time L and time constant T, 3. Find gain value as ratio of steady - state change in which are determined by drawing a tangent line at output y to amplitude of input step A. the inflection point of the curve and finding the 4. Dead time L= from time of step input to the intersections of the tangent line with the time axis intersection of the tangent line with the time axis. and the steady-state level line. Using the parameters L and T, we can set the values of KP, Ki and Kd 5. T+L= time interval between the step input and the intersection of the tangent line with the final steady- according to the formula shown in the table below. state output level.
6. The output signal can be recorded as sketched in
Fig.3.10, from which the parameters of K, L and T can be extracted
Table 2: Ziegler–Nichols tuning formula 6. CONCLUSION
From step From frequency The control of liquid level in tanks is one of controller the basic problem in the process industries. To response type achieve this, PID controller for the plant is simulated Kp Ti Td Kp Ti Td and implemented. The work mainly includes the P 1\a 0.5Kc design and modeling of liquid level system. System PI 0.9\a 3L 0.4Kc 0.8Tc identification of LLS is done and modeled through PID 1.2\a 2L L\2 0.6Kc 0.5Tc 0.12Tc Empirical Zeigler-Nichols Tuning method. A transfer function is obtained with first order system plus 5. TUNING OF PID CONTROLLER PARAMETERS delay. Then the PID controller is designed by using Simulation of liquid level system using the controller Zeigler-Nichols tuning method. PID REFERENCES [1] Tatsuo Nakagawa, Akihiko Hyodo, and Kenji Kogo ‘‘Contactless Liquid-Level Measurement With Frequency Modulated Millimeter Wave Through Opaque Container’’ IEEE sensors journal, vol. 13, no. 3, march 2013. Fig 5: Simulation of single tank using pid controller [2] Book, Dr. A. Aziz. Bazoune “Chapter 7 – Fluid Systems and Thermal Systems.” ME 413 Systems Dynamics & Control. Table:3: PID Parameters from Empirical Ziegler- Nichols tuning formula [3] Pid Controller Design From "Linear Feedback Kp Ki Kd Control" by Dingyu Xue, YangQuan Chen, and Derek 1.2 0.6 0.6 P. Atherton. constant
