scholarly journals Input-Delay Model Predictive Control of Inland Waterways Considering the Backwater Effect

Author(s):  
P. Segovia ◽  
L. Rajaoarisoa ◽  
F. Nejjari ◽  
E. Duviella ◽  
V. Puig
10.29007/p2cx ◽  
2018 ◽  
Author(s):  
Pau Segovia ◽  
Lala Rajaoarisoa ◽  
Fatiha Nejjari ◽  
Eric Duviella ◽  
Vicenç Puig

Inland waterways are large, complex systems composed of interconnected navigation reaches dedicated mainly to navigation. These reaches are generally characterized by negligible bottom slopes and large time delays. The latter requires ensuring the coordination of the current control actions and their delayed effects in the network. Centralized control strategies are often impractical to implement due to the size of the system. To overcome this issue, a distributed Model Predictive Control (MPC) approach is proposed. The system partitioning is based on a reordering of the optimality conditions matrix, and the control actions are coordinated by means of the Optimality Condition Decomposition (OCD) methodology. The case study is inspired by a real inland waterways system and shows the performance of the approach.


Author(s):  
Kailei Liu ◽  
Hongbin Qiang ◽  
Shaopeng Kang

Hydraulic drive unit (HDU) is a typical actuator, but the characteristic of input delay hinders the application of many advanced control methods in HDU. Firstly, in this paper, a mathematical model of HDU with input delay is established and the parameters are identified. Then, aiming at the input delay problem in HDU, a Smith estimated compensation model predictive control (SECMPC) strategy is proposed. On the one hand, the input delay state equation is employed to be a mathematical pattern for the state observation and predictive model. However, the combination between model predictive control (MPC) and Smith estimated compensation (SEC) is realized, the system state at k+d (d is the time delay coefficient) time is estimated in advance at k time to compensate the delay of the state. And then the prediction model based on input delay state equation is used for model prediction and rolling optimization. Thus the delay system which is unstable is promoted to a stable system without delay, and the effectiveness of SECMPC is proved with the HDU experiment and simulation. The SECMPC have some guiding significance for the control of systems with input delay.


Sign in / Sign up

Export Citation Format

Share Document