Staff publications (SoE)

Permanent URI for this collection

https://dspace.lib.cranfield.ac.uk/handle/1826/19

Browse

Browsing Staff publications (SoE) by Author "Al Seyab, Rihab Khalid Shakir"

Now showing 1 - 3 of 3
  • Thumbnail Image
    ItemOpen Access
    Differential recurrent neural network based predictive control.
    (Elsevier Science B.V., Amsterdam., 2008-07-24T00:00:00Z) Al Seyab, Rihab Khalid Shakir; Cao, Yi
    In this paper an efficient algorithm to train general differential recurrent neural network (DRNN) is developed. The trained network can be directly used in the nonlinear model predictive control (NMPC) context. The neural network is represented in a general nonlinear state-space form and used to predict the future dynamic behavior of the nonlinear process in real time. In the new training algorithms, the ODEs of the model and the dynamic sensitivity are solved simultaneously using Taylor series expansion and automatic differentiation (AD) techniques. The same approach is also used to solve the online optimization problem in the predictive controller. The efficiency and effectiveness of the DRNN training algorithm and the NMPC approach are demonstrated through a two-CSTR case study. A good model fitting for the nonlinear plant at different sampling rates is obtained using the new method. A comparison with other approaches shows that the new algorithm can considerably reduce network training time and improve solution accuracy. The DRNN based NMPC approach results in good control performance under different operating conditions.
  • Thumbnail Image
    ItemOpen Access
    Nonlinear model predictive control for the ALSTOM gasifier.
    (Elsevier Science B.V., Amsterdam., 2006-09-01T00:00:00Z) Al Seyab, Rihab Khalid Shakir; Cao, Yi
    In this work a nonlinear model predictive control based on Wiener model has been developed and used to control the ALSTOM gasifier. The 0% load condition was identified as the most difficult case to control among three operating conditions. A linear model of the plant at 0% load is adopted as a base model for prediction. A nonlinear static gain represented by a feedforward neural network was identified for a particular output channel—namely, fuel gas pressure, to compensate its strong nonlinear behaviour observed in open-loop simulations. By linearising the neural network at each sampling time, the static nonlinear model provides certain adaptation to the linear base model at all other load conditions. The resulting controller showed noticeable performance improvement when compared with pure linear model based predictive contro
  • Thumbnail Image
    ItemOpen Access
    Nonlinear system identification for predictive control using continuous time recurrent neural networks and automatic differentiation.
    (Elsevier Science B.V., Amsterdam., 2008-07-01T00:00:00Z) Al Seyab, Rihab Khalid Shakir; Cao, Yi
    In this paper, a continuous time recurrent neural network (CTRNN) is developed to be used in nonlinear model predictive control (NMPC) context. The neural network represented in a general nonlinear state-space form is used to predict the future dynamic behavior of the nonlinear process in real time. An efficient training algorithm for the proposed network is developed using automatic differentiation (AD) techniques. By automatically generating Taylor coefficients, the algorithm not only solves the differentiation equations of the network but also produces the sensitivity for the training problem. The same approach is also used to solve the online optimization problem in the predictive controller. The proposed neural network and the nonlinear predictive controller were tested on an evaporation case study. A good model fitting for the nonlinear plant is obtained using the new method. A comparison with other approaches shows that the new algorithm can considerably reduce network training time and improve solution accuracy. The CTRNN trained is used as an internal model in a predictive controller and results in good performance under different operating conditions.