Dissolved Oxygen Model Predictive Control for Activated Sludge Process Model Based on the Fuzzy C-means Cluster Algorith
- PDF / 604,170 Bytes
- 10 Pages / 594.77 x 793.026 pts Page_size
- 37 Downloads / 178 Views
ISSN:1598-6446 eISSN:2005-4092 http://www.springer.com/12555
Dissolved Oxygen Model Predictive Control for Activated Sludge Process Model Based on the Fuzzy C-means Cluster Algorithm Minghe Li, Saifei Hu, Jianwei Xia, Jing Wang*, Xiaona Song, and Hao Shen Abstract: In this work, the problem of predictive control of dissolved oxygen for the activated sludge process model with high nonlinearity and strong coupling is addressed. Firstly, the determination of the structure of fuzzy rules is displayed established upon Activated sludge model 1 (ASM1). Besides, the fuzzy space is divided through the clustering algorithm of fuzzy C-means. The corresponding parameters are estimated by means of the wellknown least squares method. Subsequently, a fuzzy predictive model of dissolved oxygen is established by using the historical data. The aim is to design a predictive controller that is capable of performing the online track of dissolved oxygen attributed to better dynamic response and steadier output in different weather. Ultimately, the availability and validity of the developed technique are verified by a comparison example. Keywords: Activated sludge model 1, dissolved oxygen control, fuzzy model, predictive control.
1.
INTRODUCTION
The problem of water pollution attracts more and more attention, and wastewater treatment has become particularly important [1, 2]. At present, activated sludge processes (ASPs) are widely used in complex and non-linear wastewater treatment process [3–6]. The method of process is mainly to charge the aeration tank with an appropriate amount of oxygen [7]. Therefore, the dissolved oxygen (DO) concentration becomes a very important parameter. If the DO concentration is too high, it will increase the operating energy consumption. On the contrary, it will cause the sludge to swell and reduce the decomposition effect. Hence, the dissolved oxygen concentration must be controlled within a certain range to meet effluent standards [8, 9]. In recent years, many scholars have conducted extensive research to handle the issue of DO control. For example, in [10], the issue of DO control was achieved by optimizing the parameters of the proportional-integralderivative (PID) controller. However, it is not satisfactory to control the nonlinear system by PID controller. So many advanced control strategies such as fuzzy control [11, 12], adaptive control [13–16] and sliding mode
control [17, 18] are constantly being applied to DO control. A powerful tool is fuzzy modeling approach [19], which can well describe the uncertainty of objects and become more and more popular in nonlinear systems. Based upon the Takagi-Sugeno (T-S) fuzzy model, an approach that can authenticate the dynamical behavior of the DO concentration was put forward [20]. The T-S fuzzy model [21–28] has the most widely been applied to fuzzy systems in recent years. It only needs a small number of rules to describe the complex dynamic characteristics of systems well, so it is concerned in the field of nonlinear modeling [29]. As noted in [30, 3
Data Loading...
