An improved model and its applied adaptive control for power turbine generator

• Nobumasa Matsui ^[1] ; Fujio Kurokawa ^[1] ; Keiichi Shiraishi ^[2]
  1. [1] Nagasaki University

     Nagasaki University

     Japón

     
  2. [2] Mitsubishi Heavy Industries
• Localización: Compel: International journal for computation and mathematics in electrical and electronic engineering, ISSN 0332-1649, Vol. 30, Nº 1 (Special Issue: Selected papers from EVER’09), 2011, págs. 201-221
• Idioma: inglés
• Enlaces
  • Texto completo
• Resumen

  • Purpose – The purpose of this paper is to present an improved model and its applied adaptive controller for a waste heat recovery generation system using a power turbine generator (PTG) with an accurate model on shipboard that is employed by an identification method on the basis of an overall system model.

    Design/methodology/approach – The PTG system has been developed as a waste heat recovery type generation system making use of exhaust gas from the main shipboard diesel engines. Conventionally, control of a plant is exercised using the proportional‐integral‐derivative (PID)‐based controller. The PID controller, however, is difficult to keep in place because of fouling conditions and variations across time. Thus, the load bank controller is proposed using a PID‐based controller. The controller should take into account both the fouling conditions and variations across time because the exhaust gas contains considerable amounts of ash and soot. Hence, an accurate model needs to improve the dynamic characteristics of the PTG system. The identification method clarifies the PTG system. The unknown parameters of the PTG speed model can be estimated using the prediction error method after the mathematical model is transferred to the state‐space model.

    Findings – Simulation results are verified with measured data of a prototype. In the transient response of the PTG speed, all the errors are within 0.23 percent. The proposed model using the identification method shows the error between the accurate model and the standard to be less than 10 percent. The proposed controller is evaluated by comparing it with the conventional controller. As a result of using the proposed controller, limit speed overshooting is improved by more than 25 percent. Hence, the proposed model is confirmed to have excellent property.

    Originality/value – The PTG is an extremely effective system for fuel cost reduction in the face of rising fuel prices, and systems capable of providing several thousand kilowatts are being considered.