Abstract
Computational fluid dynamics (CFD) techniques are widely adopted to simulate the behavior of fire. However, CFD suffers from the shortcoming of requiring extensive computer storage and a lengthy computational time. In practical applications, although comprehensive field information on velocities, temperatures, pressure, and the fractions of different constitutes can be obtained from CFD simulations, the user may only be interested in few important parameters that index the performance of a compartment design in the event of a fire. The height of the thermal interface (HTI) is one such key index, and refers to the average height above floor level inside a fire compartment at which the temperature gradient is highest. In practice, a fire compartment is considered untenable when the HTI drops below the respiratory level of the occupants, and in optimizing the design of a fire system, another set of design parameters (e.g., the width of the door opening) must be examined if the HTI of a fire compartment design is evaluated by CFD as being too low. This trial and error exercise then continues until a close to optimum set of design parameters is achieved. This approach is theoretically feasible, but requires lengthy computational time. This paper proposes the application of an Artificial Neural Network (ANN) approach as a fast alternative to CFD models to simulate the behavior of a compartment fire. A novel ANN model named GRNNFA has been specially developed for fire studies. It is a hybrid ANN model that combines the General Regression Neural Network (GRNN) and Fuzzy ART (FA). The GRNNFA model features a network structure that grows incrementally, stable learning, and the absence of the noise embedded in experimental fire data. It has been employed to establish a system response surface based on the training samples collected from a full-scale experiment on compartment fire. However, as the available training samples may not be sufficient to describe the behavior of all systems, and especially those involving fire data, this paper proposes that extra knowledge be acquired from human experts. Human expert network training has thus been developed to remedy established system response surface problems. After transforming the remedied system response surface to the problem domain, a Genetic Algorithm (GA) is applied to evaluate the close to optimum set of design parameters.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Balagangadhar D, Roy S (2001) Design sensitivity analysis and optimization of steady fluid-thermal systems, Comput Method Appl Mech Eng 190:5465–5479
Bezdek JC (1980) A convergence theorem for the Fuzzy ISODATA clustering algorithms. IEEE Trans PAMI 2:1–8
Burrascano P (1991) Learning vector quantization for the probabilistic neural network. IEEE Trans Neural Netw 2:458–462
Carpenter GA, Grossberg S, David BR (1991) Fuzzy ART: fast stable learning and categorization of analog patterns by an adaptive resonance system. Neural Netw 4:759–771
Cavalieri S, Russo M (1998) Improving Hopfield neural network performance by fuzzy logic-based coefficient tuning. Neurocomputing 18:107–126
Chen Y, Sathyamoorthy S, Serio MA (1982) New fire detection system using FT-IR spectroscopy and artificial neural network. NISTIR6242, NIST annual conference on fire research, Gaithersburg, MD
Darwin C (1859) On the origin of species by means of natural selection or the preservations of favoured races in the struggle for life. Murray, London
Díaz G, Sen M, Yang KT, McClain RL (1999) Simulation of heat exchanger performance by artificial neural networks. HVAC Res 5(3):195–208
Díaz G, Yanes J, Sen M, Yang KT, McClain RL (1996) Analysis of data from single-row heat exchanger experiments using an artificial neural network. Proc ASME Fluid Eng Div FED 242:45–52
Drysdale D (1988) An introduction to fire dynamics. Wiley
Efron B (1979) Bootstrap methods: another look at the jackknife. Ann Stat 7:1–26
Efron B (1981) Nonparametric standard errors and confidence intervals. Can J Stat 9:139–172
Floreano D, Mondada F (1998) Evolutionary neurocontrollers for autonomous mobile robots. Neural Netw 11:1461–1478
Fogel DB, Chellapilla K (2002) Verifying Anaconda’s expert rating by competing against Chinook: experiments in co-evolving a neural checkers. Neurocomputing 42:69–86
Hansen JV, Meservy RD (1996) Learning experiments with genetic optimization of a generalized regression neural network. Decis Support Syst 18:317–325
Holland JH (1975) Adaptation in natural and artificial system, MIT Press
Holland JH (1992) Genetic Algorithm. Sci Am 267(1):66–72
Hou D (1998) Thermal design and optimization by integration of simulation and response surface method—methodologies and applications. PhD Thesis, University of Texas at Austin
Iba H, Mimura A (2002) Inference of a gene regulatory network by means of interactive evolutionary computing. Info Sci 145:225–236
Ishii H, Ono T, Yamauchi Y, Ohtani S (1994) Fire detection system by multi-layered neural network with delay circuit. In: Proceedings of the 4th international symposium on fire safety science, pp 761–772
Jambunathan K, Hartle SL, Ashforth-Frost S, Fontama VN (1996) Evaluating convective heat transfer coefficients using neural networks. Int J Heat Mass Transf 39(11):2329–2332
Kohonen T (1990) The self-organizing map. Proc IEEE 78(9):464–1480
Lawrence Berkeley Laboratory (1989) DOE-2 supplement version 2.1D. Report LBL-8706, Rev 5 Supplement. Lawrence Berkeley Laboratory, Berkeley, CA
Lee EWM, Lim CP (2002) Discussion on possibility of integrating human intervention in genetic algorithms and artificial neural networks, Private Communication, City University of Hong Kong, April 2002
Lee EWM, Yuen KK, Lo SM, Lam KC (2000) Prediction of sprinkler actuation time using the artificial neural networks. J Build Surv 2(1):10–13
Lee EWM, Yuen RKK, Lo SM, Lam KC (2001) Application of Fuzzy ARTMAP for prediction of flashover in compartmental fire. In: Proceedings of the international conference on Construction Hong Kong, 19–21 June 2001, pp 301–311
Lee EWM, Yuen RKK, Lo SM, Lam KC (2002) Probabilistic inference with maximum entropy for prediction of flashover in single compartment fire. Adv Eng Info 16:179–191
Lee EWM, Lim CP, Yuen RKK, Lo SM (2004a) A hybrid neural network for noisy data regression. IEEE Trans Syst Man Cybern B 34(2):951–960
Lee EWM, Yuen RKK, Lo SM, Lam KC (2004b) A novel artificial neural network fire model for prediction of thermal interface location in single compartment fire. Fire Saf J 39:67–87
Lim CP, Harrison RF (1997) Modified Fuzzy ARTMAP approaches Bayes optimal classification rates: an empirical demonstration. Neural Netw 10(4):755–774
McGranttan KB, Baum HR, Rehm RG, Hamins A, Forney GP (2000) Fire dynamics simulator—technical reference guide. NISTIR 6467 2000, National Institute of Standards and Technology, Gaithersburg, MD
Milke JA, Mcavoy TJ (1995) Analysis of signature patterns for discriminating fire detection with multiple sensors. Fire Technol 31(2):120–136
Moore B (1989) ART 1 and pattern clustering. In: Touretzky D, Hinton G, Sejnowski T (eds) Proceedings of the 1988 connectionist models summer school, San Mateo, CA, Morgan Kaufmann Publishers, pp 174–185
Okayama Y (1991) A primitive study of a fire detection method controlled by artificial neural net. Fire Saf J 17(6):535–553
Pacheco-Vega AJ (2002) Simulation of compact heat exchangers using global regression and soft computing. PhD dissertation, Department of Aerospace and Mechanical Engineering, University of Notre Dame
Pacheco-Vega A, Sen M, Yang KT, McClain RL (1998) Genetic-algorithm-based predictions of fin-tube heat exchanger performance. In: Proceedings of the 11th international heat transfer conference, pp 137–142
Pacheco-Vega A, Sen M, Yang KT (2003) Simultaneous determination of in- and out-tube heat transfer correlations in heat exchangers by global regression. Int J Heat Mass Transf 46:1029–1040
Penny WD, Roberts SJ (1999) Bayesian neural networks for classification: how useful is the evidence framework? Neural Netw 12:877–892
Pfister G (1997) Multisensor/multicriteria fire detection: a new trend rapidly becomes state of art. Fire Technol 33(2):115–139
Sen M, Yang KT (2000) Applications of artificial neural networks and genetic algorithms in thermal engineering. In: Kreith F (ed) CRC handbook of thermal engineering, Sect 4, vol 24, pp 620–661
Specht DF (1991) A general regression neural network. IEEE Trans Neural Netw 2(6):568–576
Steckler KD, Quintiere JD, Rinkinen WJ (1982) Flow induced by fire in a compartment. NBSIR 82–2520, National Bureau of Standards, Washington DC
Tomandl D, Schober A (2001) A modified general regression neural network (MGRNN) with new, efficient training algorithms as a robust ‘black box’ tool for data analysis. Neural Netw 14:1023–1034
Traven HGC (1991) A neural network approach to statistical pattern classification by ‘semiparametric’ estimation of probability density functions. IEEE Trans Neural Netw 2(3):366–377
Tseng ML (1991) Integrating neural networks with influence diagrams for multiple sensor diagnostic systems. PhD dissertation, University of California at Berkeley
Wang HY, Coutin M, Most JM (2002) Large-eddy-simulation of buoyancy-driven fire propagation behind a pyrolysis zone along the vertical wall. Fire Saf J 37:259–285
Yamada S, Kage H, Nakashima M, Shiono S, Maeda M (1992) Data processing for multi-channel optical recording: action potential detection by neural network. J Neurosci Method 43(1):23–33
Acknowledgments
The work described in this paper was fully supported by a grant from the Research Grant Council of the Hong Kong Special Administrative Region, China [Project No. CityU 115506].
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Lee, E.W.M. Application of an intelligent model developed from experimental data to building design for fire safety. Stoch Environ Res Risk Assess 23, 493–506 (2009). https://doi.org/10.1007/s00477-008-0236-4
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00477-008-0236-4