Applying Data-driven Techniques to Online Updated PID
Controller for Calciner Outlet Temperature
ZHANG Qiang
1
, Yu Hongliang
1
, Xu Dezhi
2
1. School of Electrical Engineering, University of Jinan, Jinan 250022, P. R. China
E-mail: ave
zhangq@ujn.edu.cn, cse yuhl@ujn.edu.cn
2. College of Internet of Thing Engineering, Jiangnan University, Wuxi 214122, P. R. China
E-mail: xudezhi@jiangnan.edu.cn
Abstract: An online updated proportion integration differentiation (PID) controller is proposed for calciner outlet temperature
using input/output (I/O) data. Firstly, the adaptive observer is used to estimate the pseudo-partial derivative (PPD) parameter of
compact form dynamic linearization, which is used to dynamically linearize a nonlinear system. Secondly, The proposed PID
parameter setting principles is only based on the PPD parameter estimation derived online from the I/O data of the controlled
system. Finally, the simulation results show the good tracking control performance of the proposed method.
Key Words: Calciner outlet temperature; Data-driven control; Adaptive observer; PID controller
1 Introduction
In recent years, there has been significant progress in the
area of designing controllers for actual systems [1-3]. This
is motivated to a large extent by the fact that actual systems
are difficult to control. However, these techniques are usu-
ally implemented under the assumption of superior identi-
fication of process dynamics and their operational circum-
stances. So, they cannot give satisfactory results when suf-
fering poorly modeling.
The term data-driven is firstly proposed in computer sci-
ence and has only recently entered the vocabulary of the con-
trol society. Because only the input/output (I/O) measure-
ment data is used in data-driven controller design procedure,
the modeling process, the unmodeled dynamics and the the-
oretical assumptions all disappear. So it has caught consid-
erable attention in recent years [4-10]. There are a few data-
driven control methods as following: model-free adaptive
control (MFAC)[4-6], iterative feedback tuning (IFT)[7], it-
erative learning control (ILC)[8], virtual reference feedback
tuning (VRFT)[9],lazy learning control (LLC), unfalsified
control (UC) methodology and others.
On the other hand, proportion integration differentiation
(PID) control technique has been applied to industrial pro-
cesses, which is a most promising control algorithm. The
PID controller achieves computational simplicity by using
simpler but more intuitive design guidelines. The advantages
of fewer online calculations, a simpler algorithm and higher
control precision are attributes of the PID control, which
contribute to its industrial use.
Based on the aforementioned works,we study the prob-
lem of designing a stabilizing online updated PID controller
for calciner outlet temperature systems under input/output
(I/O) data. First, under our published result[11,12], we
present adaptive observer to estimate unknown parameter in
the compact form dynamic linearization. Then, to guarantee
the stability of the tracking temperature error, a stabilizing
online updated PID controller is designed by the PPD pa-
This work is supported by National Natural Science Foundation
(NNSF) of China under Grant 61403161 and Natural Science Foundation
of Shandong Province under Grant ZR2012FQ030).
rameter estimation. To the authors’s knowledge, few online
updated PID controller and the PPD parameter identification
observer forcalciner outlet temperature systems have been
studied, which partly motivates our present work.
The rest of this paper is organized as follows. Following
the introduction, calciner description and the problem for-
mulation are described, some assumptions and several lem-
mas which will play a basic role in our analysis, are intro-
duced in Section 2. In Section 3, a systematic adaptive ob-
server is presented for PPD parameter identification. Then,
we propose online updated PID control algorithm based on
the adaptive observer in section 4. The simulation results are
presented to demonstrate the effectiveness of the proposed
scheme in Section 5. Finally, conclusions are drawn in Sec-
tion 6.
2 Calciner Description and Problem Formulation
Outlet temperature stabilization is the basic requirement
for calciner tasks. The calciner process is shown in Fig.1.
C
1
C2
C3
C4
C5
Raw material
feeding
Calci
ner
Kiln
Cold smoke
chamber
Coal feeding
Tertiary
air
Exhaust gas
Gooseneck
Fig. 1: Cement precalcining process
In fig.1, the raw materials are putted into the connection
pipeline of C1-C2 cyclone, and the hot wind from C2 cy-
clone puts the material into C1 cyclone for separating. Then
the materials enter the pipeline of C2-C3 cyclone from the
air lock at the bottom of C1 cyclone. Subsequently, the hot
wind from C3 cyclone will put the material into C2 cyclone
Proceedings of the 34th Chinese Control Conference
Jul
28-30, 2015, Han
zhou, China
94