17-03-2014, 07:13 PM
Abstract
We present a platform for implementing low power selftuning
analog proportional-integral-derivative controllers.
By using a model-free tuning method, the platform overcomes
problems typically associated with reconfigurable
analog arrays. Unlike a self-tuning digital PID controller,
our prototype controller combines the advantages of low
power, no quantization noise, high bandwidth and high
speed. The prototype hardware uses a commercially available
field programmable analog array and Particle Swarm
Optimization as the tuning method. We show that a selftuned
analog PID controller can outperform a hand-tuned
solution and demonstrate adaptability to plant drift.
1 Introduction
Proportional-Integral-Derivative (PID) control has been
used successfully for regulating processes in industry for
more than 60 years. Today, digital self-tuning PID controllers
are ubiquitous in the industry. These controllers automatically
set gain values (i.e. parameters) according to
the process (alternatively, plant) and may optionally require
inputs from a human designer.
The parameter settings of a PID controller for optimal
control of a plant depend on the plant’s behavior. Therefore,
information about the plant is required to tune the PID
controller. The tuning methods fall into two broad categories:
online model-free methods and methods that build
a model of the plant. The former tune the PID controller in
loop with the given plant using an optimization algorithm
such as steepest descent or Newton’s method to minimize
some cost function [12, 14], for instance, error between the
input and output . The second approach builds a model of
the plant and accordingly decides the parameters of the controller
by using a deterministic approach or an optimization
method [16, 3, 5]. A comprehensive review of these methods
is given in [14].
The hardware used for PID controllers has evolved from
pneumatic configurations in the 1940’s, electrical devices
in the 1960’s to the current microprocessor-based controller
technology. Observation indicates that analog electronic
PID controllers have never been in widespread use. This
is despite the fact that most processes that are controlled are
analog. Instead of a strictly analog controller, digital signal
processor (DSP) based PID controllers are used. These employ
analog-to-digital converters (ADCs) at their input and
digital-to-analog converters (DACs) at their output to interface
with the plant. An analog controller combines several
advantages over a digital controller which can be summarized
as follows.
We present a platform for implementing low power selftuning
analog proportional-integral-derivative controllers.
By using a model-free tuning method, the platform overcomes
problems typically associated with reconfigurable
analog arrays. Unlike a self-tuning digital PID controller,
our prototype controller combines the advantages of low
power, no quantization noise, high bandwidth and high
speed. The prototype hardware uses a commercially available
field programmable analog array and Particle Swarm
Optimization as the tuning method. We show that a selftuned
analog PID controller can outperform a hand-tuned
solution and demonstrate adaptability to plant drift.
1 Introduction
Proportional-Integral-Derivative (PID) control has been
used successfully for regulating processes in industry for
more than 60 years. Today, digital self-tuning PID controllers
are ubiquitous in the industry. These controllers automatically
set gain values (i.e. parameters) according to
the process (alternatively, plant) and may optionally require
inputs from a human designer.
The parameter settings of a PID controller for optimal
control of a plant depend on the plant’s behavior. Therefore,
information about the plant is required to tune the PID
controller. The tuning methods fall into two broad categories:
online model-free methods and methods that build
a model of the plant. The former tune the PID controller in
loop with the given plant using an optimization algorithm
such as steepest descent or Newton’s method to minimize
some cost function [12, 14], for instance, error between the
input and output . The second approach builds a model of
the plant and accordingly decides the parameters of the controller
by using a deterministic approach or an optimization
method [16, 3, 5]. A comprehensive review of these methods
is given in [14].
The hardware used for PID controllers has evolved from
pneumatic configurations in the 1940’s, electrical devices
in the 1960’s to the current microprocessor-based controller
technology. Observation indicates that analog electronic
PID controllers have never been in widespread use. This
is despite the fact that most processes that are controlled are
analog. Instead of a strictly analog controller, digital signal
processor (DSP) based PID controllers are used. These employ
analog-to-digital converters (ADCs) at their input and
digital-to-analog converters (DACs) at their output to interface
with the plant. An analog controller combines several
advantages over a digital controller which can be summarized
as follows.