2019 5th International Conference for Convergence in Technology (I2CT)

Pune, India. Mar 29-31, 2019

Real-time Implementation of PID Control

Algorithm using Power Pmac Controller for
Antenna Position Control System.
Nupoor Patil Durgesh Behere
IMTPL,Group Intelmac, IMTPL,Group Intelmac,
Ambegaon,Pune Ambegaon,Pune
[email protected] [email protected]

Abstract—Ground station antennas find their application in potentiometer changes the input angle, θ(s) to a voltage,
communication with satellites. An antenna communicates with Vi(s).
the spacecraft by sending command signals and receiving the The purpose of the preamplifier is to take the input
data from the spacecraft. Antenna dish rotates with respect to signal voltage and output a voltage that the power amplifier
elevation axis and whole antenna structure rotates on circular
track which is an azimuth axis [1]. System identification is an
can use. The Preamplifier is also modeled by a gain that can
important task of modelling of a process. A model merely be specified by the design engineer to achieve a desired
represents the actual process in much simpler format. output[2]. The Preamplifier is a system in which the input
Modelling is required to design and simulate new control voltage is amplified by some gain K and output as a voltage.
technique before real time application .This paper discusses takes the output voltage from the Preamplifier and converts
modeling of antenna control systems and implementation of it to a Voltage that is useable by the motor. This requires the
PID algorithm into power Pmac controller. Simulation results Power Amplifier to output a significant amount of power,
have been evaluated for open loop as well as closed loop something that the Preamplifier is not capable of.
system. Controller has been applied to real-time antenna
system and hardware results for step and ramp signals for
azimuth position have been evaluated. Paper concludes with
inferences drawn from the designed control algorithm.

Keywords—System Identification, Step test, Ramp

test,Antenna Control systems,Power Pmac,Matlab,Simulink.

I. INTRODUCTION
Under ISTRAC’s (ISRO) Network stations new 11m-
antenna second terminal has been installed in ground station
at Bhopal, Madhya Pradesh . This is a co-located antenna
along with antenna ground station at Port-Blaire(Andaman)
; with the first terminal as a full backup terminal for Fig. 1: Antenna System without controller
supporting ISRO’s upcoming launch and satellite programs.
A position control system basically consists of position The horizontal control axis is called elevation and
sensing module and error correction module. The aim to; set vertical axis is azimuth. Antenna servo control systems
the actual value from the position encoder (22 bit) to match (ACSS ) consists of two parts are is situated in pedestal
with desired position value, thus reducing the error. For room. The antenna control unit (ACU )is a part of ACSS
positioning control standard servo PID algorithm in Power which consists of controller responsible for position control
Pmac controller has been used .Antenna servo control .Antenna drive unit (ADU ) is a part of ACSS which
system with monopulse tracking has been described briefly consists of electrical driving components like drives
in section II. Section III explains System identification ,motors, gears etc.[3]
technique .Section IV simulation of PID algorithm on Figure (2) shows antenna control systems. The primary
antenna position control system. Section V is about operator interface for the Antenna control servo system is
introduction to power Pmac controller. .Section VI describes the Remote Antenna Console (RAC) located in the TTC
real- time results of developed PID algorithm. Section VII Control Room at a distance from the Antenna Control Unit
concludes the paper. References and acknowledgements are (ACU) which provides remote control of ACU. The Remote
provided in last sections. Antenna Console (RAC) communicates with the Antenna
Control Unit (ACU) over OFC (Optical Fiber Cable)
II. ANTENNA CONTROL SYSTEM Interface for 100 to 300 m distance.
For assurance of continuous tracking during earth’s ACU is responsible for closing the position loop, reading
rotation antenna dish rotates with respect to two control the position sensors and commanding the Antenna
axis. Figure (1) displays antenna system without controller. Azimuth/Elevation Drives (ADU ). In order to remove the
The input and feedback potentiometer each have an effect of backlash in gears ADU has been provided with
associated transfer function, in the form of a gain. The counter torque arrangement. The position loop is built with

978-1-5386-8075-9/19/$31.00 ©2019 IEEE 1

appropriate inner loops (rate loop and current loops) such position is measured using encoder .The controller tries to
that the equal-and-opposite counter-torque-bias is added minimize the error between actual and reference value so
appropriately at the rate loop input and both motors that the target is achieved. Empirical or data-based System
feedback taken in the loops. ACU provides antenna control identification is a task of determining the relationship
functions for remote operations from motion and control between the input and output data based upon the collected
system through RAC. and observed data. It builds a mathematical map between
the applied value and the measured value. All the blocks
that include electrical drives , electrical motor ,antenna load
and gear box are considered as a single transfer function
model with applied input as target torque ( applied at drive)
and actual position as output ( recorded at the encoder).
Figure(4) exemplify the data of measured input and output
value used for system identification. System identification
toolbox is used for obtaining the process model as well as
the transfer function model of the system.

Fig. 2: Antenna Control systems

An integral or external color display allows the operator

to continuously and simultaneously view all information of
interest in clear alphanumeric font. The Major subsystems
like RAC, ACU or Servo-Controller, Servo Motors and the
associated Drive Amplifiers etc. have after sales service
support for a minimum period of 10 years.
III. SYSTEM MODELS FROM IDENTIFICATION
Obtaining a model of plant is a tedious task for control
theory applications. Models can be obtained by two ways
either through rigorous calculations or through recording
data. Models obtained by calculations provide good insight Fig. 4:Data set for system identification
of the plant .However it requires an expertise to get an
accurate model .Another method of obtaining a model is Figure (5) point up the percentage fit for obtained models
through empirical system identification or data based system with respect to validation data. For identification and
identification . It is the most accurate and fastest method validation two different data sets taken from same plant
because the models are obtained by real time plant data that have been used. The transfer function obtained is
inherently includes all the measured and unmeasured
parameters that directly or indirectly affect the system. ʹ͹ʹǤͻ
However, the technique does not give actual insight of the ‫ܩ‬ሺ‫ݏ‬ሻ ൌ ǥ ǥ ǥ Ǥ ሺͳሻ
ሺ‫ݏ כ ݏ‬ሻ ൅ ͹͸Ǥ͹ͷ‫ ݏ‬൅ ͷǤͷ͵ͳ
system under operation. The system identification toolbox
of Matlab being the most powerful tool is used for
identifying the plant model[4].
State space schematic of antenna positioning system is
shown in figure (3).

Fig. 3:State space schematic of antenna positioning system

Controller takes two inputs; firstly the input which is

reference value and secondly a value from the feedback.
Input of the process is the reference position value. Actual Fig. 5:Percentage fit for antenna models

2
 state response. Figure (8) shows the PID simulation for the
The obtained transfer function model was validated obtained transfer function. A continuous PID block is
using cross correlation for noise spectrum analysis. It can be applied to continuous transfer function that represents the
seen from figure (6) that data stays within confidence level load system (drive, motor, antenna and gearbox).Simulation
proving that there is no cross correlation between output and results have been obtained for step input and ramp input .
input value. So the model obtained is a noble model.

Fig. 8:Simulation of PID control algorithm

Fig. 6:Cross-corelation of data set

Fig. 9:Step Response of PID signal

Figure (9) displays the simulation results for step

response of system Various step data information like rise
time. settling time for open loop as well as closed loop
system with controller have been obtained for understanding
the improvement in performance Tuning values taken are
Kp=0.1; Ki=0.001 and Kd=8. Table(1) displays the
performance criteria for designed PID data.
Fig. 7:Bode plot of system model TABLE I:STEP RESPONSE OF OPEN LOOP AND CLOSED LOOP
SYSTEM
Parameters Open loop PID performance
Figure (7) depicts the bode analysis of the model. It can response evaluation
be said that model replicates the graph that of a low pass
Rise Time 30.53 secs 1.03secs
filter having good gain at lower frequencies and disturbing
gain at higher frequencies. Thus model attenuates any gain Settling Time 54.38 secs 6.88secs
changes at higher frequencies and amplifies any minute Overshoot 0% 11.5
changes at lower frequencies[5]. Gain Margin Inf 44.7
Phase Margin 88.5 75.9
IV. SIMULATION OF PID ALGORITHM ON ANTENNA POSITION
CONTROL SYSTEM
The open loop performance depicts the rise time as 30
PID controller is the most basic form of controller for secs and settling time as 54 secs .However inclusion of PID
any application. It has three parameters that can be tuned to controller has brought the rise time to 1 secs and settling
obtain a desired response. The proportional constant is time to 6 secs. The system response with controller displays
responsible for system gain but creates an steady state an overshoot of 11% with gain margin of 44.7 and phase
offset. The integral constant nullifies the steady state offset margin of 75.9.Since values of both gain margin and phase
created due to proportional action but makes the system margin are positive even after implication of controller the
response sluggish. The derivative constant is used for closed loop system is BIBO stable.
anticipatory action improving transient as well as steady

3
 generation .The generated code needs to deployed and then
compiled in user code generation setup of power Pmac IDE
software .Once the ccode gets downloaded successfully the
PPmac controller now works on user generated PID
algorithm.[8]
VI. REAL- TIME RESULTS OF DEVELOPED PID ALGORITHM
PID algorithm which was generated has been deployed
in power Pmac .The controller was then checked for step
response and ramp response. Figure(12) exhibits the step
response of the system. For step response, a positive step
was first applied to the system followed by a negative step.
It was observed that system responds for target step value
with a maximum following error of 694 counts for 22 bit
encoder in essence 0.05 deg.[9]

Fig. 10:Ramp response of the system simulation

Figure (10) illustrates ramp response of the system. It

can be evidenced that ramp of slope 1 has been applied as
input and system tracks the ramp signal with minimum
error. Therefore, for simulation purpose two types of input
signals namely step and ramp have been used for analyzing
the performance of designed controller.[6]
V. POWER PMAC CONTROLLER
Power Pmac is a motion controller and computer rolled
up in one. The Power PMAC family of controllers is the
latest generation of motion and machine controllers from Fig. 12:Step response of real time antenna control system
Delta Tau Data Systems, Inc. It is available in a large and
increasing number of configurations, permitting the user to Figure (13) reveals ramp response of the system for the
configure controller hardware and software to particular implemented controller .A ramp of positive slope was
application needs. Power PMAC is a general-purpose applied initially followed by a ramp of negative slope ,It can
embedded computer with a built-in motion and machine- be observed that system tracking is achieved for applied
control application. It also provides a wide variety of ramp signal with maximum following error of +/- 27 counts
hardware machine interface circuitry that permits in essence 0.002 deg (+ for positive ramp test and – negative
connection to common servo and stepper drives, feedback ramp test) . The actual position maps the desired position
sensors, and analog and digital I/O points[7]. Power Pmac is which concludes that controller designed is a noble
compatible with Matlab-Simulink facilitating user program controller[10].
ccode generation. The objective is to fully automate code
generation for Power PMAC’s servo loop closure routines
(tasks) utilizing Simulink control blocks. Figure (11) shows
the Power Pmac simulation diagram for PID algorithm.

Fig. 13:Ramp response of real time antenna control system

VII. CONCLUSION
Fig. 11:Power Pmac simulation for PID algorithm
Paper is about antenna position control systems used for
The plant transfer function blocks have been replaced both azimuth and elevation positioning control. First the
with power Pmac components so as to separate the antenna hardware has been explained in the system.
controller from the plant. The ccode for the controller can Modeling is done using black box technique of system
then be generated using Mathworks’ Embedded code identification which takes input as target torque and output

4
as actual position of antenna. Simulation of obtained ACKNOWLEDGEMENT
transfer function model for PID controller has been done. Authors would like to thank ISRO and allied industries
Step and ramp responses for simulation have been for cooperating and providing all required technical as well
implemented for open loop and closed loop system. PID as non-technical facilities to complete the work. Authors
algorithm was implemented in power Pmac controller would also want to express a deep sense of gratitude to
The real time step response and ramp response have been Group Intelmac for their guidance and support without
evaluated in the hardware. It can thus been concluded that which the project would not be a success.
designed PID algorithm shows appropriate response for the
system with minimum following error. Thus, power Pmac
controller is suitable for antenna position control systems AUTHORS’ BIOGRAPHIES
application as it also provides facilities for customizable
code generation using Matlab Miss. Nupoor Patil.

REFERENCES Qualification: MTech (Electrical Control

[1] W. Gawronski, Dynamics and Control of Strictures: A Modern Systems) ,BE,Diploma(Electrical)
Approach, New York, Springcr-Verlag, 1998. Work : Application Engineer at Intelligent
[2] Okumus H.I., Sahin E., Akyazi O., Antenna Azimuth Position Control Motion Technology,(P)Limited,Pune.
with Classical PID and Fuzzy Logic Controllers. IEEE Transaction on Area of Research: Control Systems, System
Education, 2012.
[3] Xuan L., Estrada J., Digiacomandrea J., Antenna Azimuth Position
Identification and Control Theory Applications.
Control System Analysis and Controller Implementation. Term
Project, 2009
[4] Lennart Ljung, “System Identification: Theory for the User”, P T R Mr.Durgesh Behere
Prentice Hall, Englewood Cliffs, New Jersey 0763
[5] A. K. Tangirala, “Principles of system identification: theory and Qualification: BE (Electrical)
practice,” 2014. Work: Application Engineer at
[6] W. Gawronski, “Modelling and Control of Antennas and Intelligent Motion
Telescopes”,Springer.
[7] Power Pmac controller manual available at
Technology,(P)Limited,Pune.
http://www.deltatau.com/manuals/ Area of Research: Motion Controller,
[8] Evans W.R., Graphical Analysis of Control Systems. Transactions of Industrial Automation.
the AIEE, 67, 547– 551, 1948.
[9] Kalman R.E., On the General Theory of Control Systems. IRE
Transactions on Automatic Control, 4, 3, 110, 1959. Abstract. Full
Paper Published in Proceedings of the 1st IFAC Congress, Moscow,
1960.
[10] Grimholt, C. and Skogestad, S. (2013). Optimal PID control on first
order plus time delay systems & verification of the Sims rules. In 10th
IFAC International Symposium on Dynamics and Control of Process
Systems.