The simulation results of the proposed control structure show significant improvement in the system performance and stability compared to the fixed PI controller.
IV. DISCUSSIONS
All the gain scheduling control approaches are based on this assumption that all states can be measured and a generalized observability holds [15]. In this study, we also need to clarify if this assumption is valid. The parameters that we need to measure or estimate are room temperature and radiator flow rate. Measuring the first state is mandatory when the goal is seeking a reference for this temperature. However, radiator flow is not easily measurable. To have an estimation of the radiator flow rate, one possibility is using a new generation of TRVs which drive the valve with a step motor. It is claimed that this TRV can give an estimation of the valve opening. Knowing this fact and assuming a constant pressure drop across the radiator valve, we would be able to estimate the flow rate.
Fig.10. (Top) ambient temperature, (bottom) room temperature for three controllers. The results of the simulation with flow adaptive controller together with two fixed PI controllers are shown. The PI controller designed for the low demand condition is very slow for the high demand situation.
We have shown through the paper that using the new generation of TRVs, gain scheduling control would guarantee the efficiency of the radiator system.
V. CONCLUSION
The dynamical behavior of a TRV controlled radiator is investigated. A dilemma between stability and performance for radiator control is presented. We dealt with the dilemma using a new generation of thermostatic radiator valves. With the new TRV, flow estimation and control would be possible. Based on the estimated flow, we have developed a gain schedule controller which guarantees both performance and stability for the radiator system. To this end, we derived low-order models of the room-radiator system. The model is parameterized based on the estimated operating point which is radiator flow rate. Gain schedule controller is designed for the resulted time varying model.
REFERENCES
[1] S. Svendsen, “Energy project villa,” Danfoss and DTU,” Main Report,mar 2005.
[2] ASHRAE, ASHRAE Handbook 1990, fundamentals. Atlanta: ASHRAE Inc., 1990.
[3] F. Tahersima, J. Stoustrup, and H. Rasmussen, “Stability-performance dilemma in TRV-based hydronic radiators,” Submitted for publication,2011.
[4] P. Andersen, T. S. Pedersen, J. Stoustrup, and N. Bidstrup, “Elimination of oscillations in a central heating system using pump control,” vol. 39, pp. 31–38, Jun 2004.
[5] P. Rathje, “Modeling and simulation of a room temperature controlsystem,” Proceedings of SIMS87 Symposium, pp. 58–69, 1987.
[6] K. Astrom and T. Hagglund, PID Controllers. North Carolina: ISA, 1995.
[7] A. K. Athienitis, M. Stylianou, and J. Shou, “Methodology for building thermal dynamics studies and control applications,” ASHRAE Trans., vol. 96, no. 2, pp. 839–848, 1990.
[8] M. L. Kurkarni and F. Hong, “Energy optimal control of residential space-conditioning system based on sensible heat transfer modeling,” Building and Environment, vol. 39, pp. 31–38, 2004.
[9] G. Hudson and C. P. Underwood, “A simple building modeling procedure for matlab/simulink,” vol. 99, no. 2, pp. 777–783, 1999.
[10] H. Madsen and J. Holst, “Estimation of a continuous time model for the heat dynamics of a building,” Energy and Buildings, vol. 22, pp. 67–79, 1995.
[11] K. M. Letherman, C. J. Paling, and P. M. Park, “The measurement of dynamic thermal response in rooms using pseudorandom binary sequences,” Building and Environment, vol. 17, no. 1, pp. 11–16, 1982. 散热器恒温阀英文文献和中文翻译(6):http://www.751com.cn/fanyi/lunwen_13992.html