Scribd Logo
Upload

Welcome to Scribd!

  • 183 views

    Coupled Tank

    Uploaded by

    Devraj Mandal
    This document provides instructions for an experiment to control the liquid level in a coupled tank system using a proportional-integral (PI) controller. The system consists of two connected tanks that can fill and drain water. The goal is to design a PI controller to regulate the water level in the second tank. The document describes: 1) Modeling the coupled tank system as a second order system. 2) Tuning the PI controller initially using Ziegler-Nichols methods. 3) Implementing the controller in Simulink and fine-tuning the parameters to meet specifications for overshoot and settling time.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd

    Coupled Tank

    Uploaded by

    Devraj Mandal
    183 views5 pages
    This document provides instructions for an experiment to control the liquid level in a coupled tank system using a proportional-integral (PI) controller. The system consists of two connected tanks that can fill and drain water. The goal is to design a PI controller to regulate the water level in the second tank. The document describes: 1) Modeling the coupled tank system as a second order system. 2) Tuning the PI controller initially using Ziegler-Nichols methods. 3) Implementing the controller in Simulink and fine-tuning the parameters to meet specifications for overshoot and settling time.

    Original Description:

    copuled tank

    Copyright

    © Attribution Non-Commercial (BY-NC)

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document provides instructions for an experiment to control the liquid level in a coupled tank system using a proportional-integral (PI) controller. The system consists of two connected tanks that can fill and drain water. The goal is to design a PI controller to regulate the water level in the second tank. The document describes: 1) Modeling the coupled tank system as a second order system. 2) Tuning the PI controller initially using Ziegler-Nichols methods. 3) Implementing the controller in Simulink and fine-tuning the parameters to meet specifications for overshoot and settling time.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    183 views5 pages

    Coupled Tank

    Uploaded by

    Devraj Mandal
    This document provides instructions for an experiment to control the liquid level in a coupled tank system using a proportional-integral (PI) controller. The system consists of two connected tanks that can fill and drain water. The goal is to design a PI controller to regulate the water level in the second tank. The document describes: 1) Modeling the coupled tank system as a second order system. 2) Tuning the PI controller initially using Ziegler-Nichols methods. 3) Implementing the controller in Simulink and fine-tuning the parameters to meet specifications for overshoot and settling time.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 5

    Indian Institute of Technology, Bombay Department of Mechanical Engineering ME 413:Control Systems Laboratory

    (At Syscon Lab, Room 202, ACRE Building)

    Expt. 3: Liquid Level Control

    1. Aim: To design a Proportional and Integral Controller for the Coupled-Tank Plant set-up. 2. Apparatus:
    The coupled tank apparatus consists of a transparent plexi glass container measuring (20 cm long, 10 cm wide and 30 cm high). A centre partition is used to divide the container into two tanks. Flow between the two tanks is by means of a series of holes drilled at the base of the partition. Refer fig1. Water is pumped from one reservoir into the first tank by a variable speed pump, which is driven by an electric motor. The actual flow rate (qi) is measured by a flow meter, which is in the flow line between the pump and tank 1. qi

    h1

    q1

    h2

    q2

    h3 Tank-1 Cross sectional area a1 Tank-2

    Cross sectional area a2

    Fig 1: Schematic of the Coupled Tank Apparatus The level of the fluid is measured using parallel track level sensors, which are stationed in tank1 and tank2. This device performs as an electrical resistance, which varies with the water level.

    The changes in resistance are detected and provide an electrical signal, which is proportional to the height of water. The water, which flows into tank 2, is allowed to drain out via an adjustable tap, and the entire assembly is mounted in a large tray, which forms the supply reservoir for the pump. The motor drive and level sensor signal processing is performed inside the instrumentation box, such that the pump motor may be driven by a voltage between zero and ten volts applied to the pump socket. Correspondingly the level sensor outputs are provided as a voltage in the range zero to -10 volts. The level sensing system is prone to extraneous disturbance; filters are provided which may be switched in and out of the circuit as required. The pump motor drive is normally derived from an analog or digital computer, however for demonstration purposes an internal drive is supplied. This is marked on the instrumentation box front panel and should normally be switched out of the circuit.

    3. Modeling:
    Consider the basic coupled tank apparatus, as described in fig 1. Taking flow balances about each tank may derive the dynamical equations of the system. For the first tank we have dV1 dH Qi Q1 = rate of change of fluid volume in tank 1 = =A 1 (1.1) dt dt where V1 = the volume of fluid in tank 1 H1 = height of the fluid in tank 1 A = Cross sectional area of tank 1 and tank 2 Q1 = flow rate of fluid from tank 1 to tank 2 Qi = pump flow rate For the second tank: Q1 Q0 = rate of change of fluid volume in tank 2 = Where V2 = the volume of the fluid in tank 2 H2 = height of the fluid in tank 2 Q0 = flow rate of fluid out of tank 2 If the inter-tank holes and the drain tap are assumed to behave like orifices, then the following equations follow from the characteristic relations for orifices: Q1 = C d1 a1 2 g ( H 1 H 2 ) (1.3) Q0 = C d 2 a 2 2 g ( H 2 H 3 ) where : a1 = cross sectional area of orifice 1 a2 = cross sectional area of orifice 2 C d1 , C d 2 = discharge coefficient ( = 0.6 for a sharp edged orifice ) (1.4)
    dV2 dH 2 =A dt dt

    (1.2)

    H3 = height of the drain tap g = gravitational constant Equations 1.1 to 1.4 describe the system dynamics in its true nonlinear form. For control system studies it will be necessary to linearize the equations by considering small variations qi in Qi, q1 in Q1, q0 in Q0, h1 in H1 and h2 in H2, all measured about some mean level H2

    Assuming that in the steady state we have: Qi = Q0 = Q1 then,


    qi q1 = A q1 q 0 = A dh1 dt dh2 dt

    (1.5)

    (1.6)

    q1 =

    h h2 Q1 Q1 1 h1 + h2 = C d1 a1 2 g 1 2 H 1 H 2 H1 H 2

    (1.7)

    And h2 H2 H3 Substituting equations 1.7 and 1.8 into 1.5 and 1.6 gives q0 =
    qi K1 (h1 h2 ) = A dh1 dt dh2 dt
    Q0 1 h2 = CD2 a2 2 g 2 H 2

    (1.8)

    (1.9)

    k1 (h1 h2 ) k2 h2 = A
    Cd1 a1 2 g 2 H1 H 2 Cd1 a2 2 g 2 H 2 H3

    (1.10)

    where

    k1 =

    (1.11)

    k2 =

    (1.12)

    Rearranging equation sets 1.9and 1.10 produces the following state space model of the coupled tanks: k1 k1 & h1 A A h1 1 A + q & = (k1 + k2 ) h2 0 i h2 k1 A A

    (1.13)

    By taking the Laplace transform of eq 1.13 the following transfer function is obtained:

    1 1 h2 ( s ) k2 k2 = = ( sT1 + 1)( sT2 + 1) qi ( s ) A2 2 A(2k1 + k2 ) s + s +1 k1k2 k1k2

    (1.14)

    where the time constants T1 and T2 are related to k1, k2 and A by:
    2 T1T2 = A

    k1k2

    , T1 + T2 =

    A(2k1 + k2 ) k1k2

    (1.15)

    The transfer function of the coupled tank including the dynamics of the motor and the sensor was identified used System Identification Techniques. In this method, a Pseudo Random Binary Signal (PRBS) is generated externally and given as input to the system and the corresponding output is recorded. PRBS is used, since it excites all the modes of the system and hence assists in estimating a more accurate Using the input and output data, the second order transfer function is computed using System identification toolbox in Matlab. The input to the system is Input voltage to the motor and output is the level of water in the second tank. G ( s) = 0.00082816 s + 0.1141727 s + 0.00055707
    2

    4. Tuning of PI Controller using the Ziegler Nichols method:


    The selection of controller coefficients to meet steady state error and stability requirements requires a transfer function model of the system. In many industrial process applications however, a reliable mathematical model of the system is not available. In such cases the rules developed by Ziegler and Nichols may be used to set up controllers using only basic information from the observed transient behavior of the system. The rule for the controller setting is as follows (Refer: Modern Control System by K.Ogatta) K K ( s) = K p + i s where, Kp = Proportional Gain Ki = Integral Gain The method employed is as follows, the proportional gain Kp is increased until the system achieves sustained oscillations. The value of proportional gain at which the oscillation occurs, Kc and the period T of the oscillations is noted. The settings recommended by Ziegler and Nichols are:

    Proportional Control: Kp = 0.5 Kc Proportional plus integral control: Kp = 0.45 Kc Ki = 0.55 Kc / T

    5. Closed Loop Specifications:


    When the PI controller is placed in closed loop to the plant and when a step input is given to the motor the output should follow the specifications given below.
    (a) (b) First peak overshoot: Less than 5% of steady state value Settling Time: Between 300 and 350sec

    6. Procedure for conduct of Experiment:


    (a) The transfer function model of the coupled tank apparatus got by the method enumerated in modeling section is as follows: 0.00082816 G ( s) = 2 s + 0.1141727 s + 0.00055707 (b) The PI controller is to be designed for this plant model. The initial tuning of the PI controller can be done by Ziegler Nichols tuning method explained earlier. This shall give us an approximate value of Kp and Ki to start with. (c) Develop the Simulink model and feed the value of Kp and Ki arrived at in the previous step. Check the closed loop response and compare with the desired specifications. Fine-tune the values of Kp and Ki until the closed loop specs are met. (d) Implement the Simulink file on the apparatus using the ADAM module. Compare the practical and theoretical results.

    You might also like