22-09-2016, 10:27 AM
1455617013-desktop.pdf (Size: 401.18 KB / Downloads: 18)
Abstract—This paper proposes the design of PID-like fuzzy
logic controller (PIDFLC), on Field Programmable Gate Array
(FPGA) device. The Fuzzy Inference System (FIS) used in the
controller is aided with Active Rules Selection Mechanism.
Developments were made to this FIS to make it able to
manipulate signed numbers, (which is important issue in
control system), then, it was blended with integral and
derivative control components of tunable gains. These new
features enable the controller to function as a PDFLC, a
PIFLC, and a PIDFLC efficiently. The design utilizes 1394
slices of the target FPGA, and is able to produce an output at
0.421 µsec with maximum frequency of 40.295 MHz.
Mathematical model of linear plants were used to test the
controller. The simulation results using the proposed controller
connected to these plants in unity feedback system were
compared with simulation results of a similar system that uses
a software-based controller. The plant responses controlled by
the proposed controller were smooth and much similar to the
plant responses when using software based controller.
I. INTRODUCTION
Fuzzy logic has rapidly become one of the most successful
of today's technologies for developing sophisticated control
systems. Fuzzy controllers are more robust than PID
controllers because they can cover a much wider range of
operating conditions than PID can, and can operate with
noise and disturbances of different nature. Given the
dominance of conventional PID control in industrial
applications, it is significant both in theory and in practice if
a controller can be found that is capable of outperforming
the PID controller with comparable ease of use. Some of
PID fuzzy controllers are quite close to this dream [1]. The
simplest and most usual way to implement a fuzzy controller
is to realize it as a computer program on a general purpose
computer. However, a large number of fuzzy control
applications require a real-time operation to interface highspeed
constraints. Software implementation of fuzzy logic
on general purpose computers
can not be considered as a suitable design solution for this
type of application, in such cases, design specifications can
be matched by specialized fuzzy processors.
Higher density programmable logic devices such as FPGAs
can be used to integrate large amounts of logic in a single
IC. Semi-custom and full-custom application specific
integrated circuit (ASIC) devices are also used for this
purpose but FPGAs provide additional flexibility: they can
be used with tighter time-to-market schedules. The FieldProgrammable
Gate Array (FPGA) places fixed logic cells
on the wafer, and the FPGA designer constructs more
complex functions from these cells. The term field
programmable highlights the customizing of the IC by the
user, rather than by the foundry manufacturing the FPGA.
Several researchers discussed the design of hardware fuzzy
logic controller. Number of these works were specialized in
control application [2,]-[3], and were aim to get better
control responses. Others were concerned in developing
general fuzzy logic processors [4]-[5]-[6]-[7]. Their searches
were concern using new techniques in fuzzy algorithm, to
get higher processing speed versus low utilization of chip
resource. As a result, the proposed design in this paper is
aim to employ the new techniques of fuzzy algorithm in
controlling industrial application with the aid of
conventional PID control to serve these applications
efficiently.
II. THE PROPOSED PID-LIKE CONTROLLER
The general layout of the controller chip in a unity
feedback control system is shown in Fig. 1. Generally, the
proposed controller accept the output of the plant (yp) and
the desired output (yd), both as digital signals, and deliver
digital control action signal as an output. The design accepts
also four 8-bit digital signals that represent the gain
coefficients needed by the controller (proportional gain Kp,
derivative gain Kd, integral gain Ki
, and output gain Ko), and
two one-bit signals to select the type of the controller
(PDFLC, PIFLC, or PIDFLC).
III. STRUCTURE OF THE PROPOSED PIDFLC
In order to build a PIDFLC, it is required to design a
fuzzy inference system with three inputs that represent the
proportional, derivative, and integral components.
A fuzzy controller with three inputs may not be
preferred, because it needs large number of rules, instead,
the PID fuzzy controller can be constructed as a parallel
structure of a PD fuzzy controller and a PI fuzzy controller
and the output of the PIDFLC is formed by algebraically
adding the outputs of the two fuzzy control blocks.
However, it is difficult to formulate control rules with the
input variable sum of error (∑e), as its steady-state value is
unknown for most control problems. To overcome this
problem, a PD controller may be employed to serve as PI
controller in incremental form. Equation