摘要
本文在综述神经网络和模糊逻辑的基础上，结合火灾报警系统
自身的特性，提出一种火灾探测特征模型及一种能够给出火灾预报
概率的智能算法。这种算法结合了神经网络和模糊逻辑各自的特点。
它判断的依据是一段时间内探测信号整体的变化趋势。因此，这种
算法能够较为准确地给出火警预报概率，仿真研究证明了方法的可
行性。
在火灾智能报警算法的基础上，本文设计并调试了一块硬件接
口板。接口板主要完成工控机与外界数据交换的功能。接口板为ISA
总线标准，以485通讯协议与外界接口。本文实现了接口板的锁存
工作方式。
本文还编制了火灾报警系统的上位机软件。上位机软件通过接
口板获取各回路探测器的信息，并将其反应在软件界面上；同时，
用户可以将不同的指令通过接口板发送出去，控制每一个探测器的
运行状态。在上位机软件整体框架的基础上，本文完成了部分关键
模块程序的编制，如通讯模块、数据库操作模块等。实验证明了设
计的合理性。
Combining with the character of fire alarm system, this paper puts
forward a fire detection model and an intelligent arithmetic that can
present a fire prediction probability on the base of summarizing neural
network and fuzzy logic. The arithmetic combines with the character of
neural network and fuzzy logic. It judges on the detection signal changing
trend in a period. So, the arithmetic could present fire prediction
probability precisely, and simulation research proves the method feasibility.
The paper designs and debugs a hardware interface board on the base
of the fire intelligent alarm arithmetic. The board fulfills the exchanging
data function between industrial controlling computer and outside. The
board bus standard is ISA, communicating according to 485 protocol. The
paper fulfills the board locking work mode.
The paper programs the software of fire alarm system. Through interface
board, the software could get each ioop detector information and response
information to software interface. At the same time, through interface
board, user could send out various commands to control any detector
running status. The paper fulfills part key module programs on the base
of software whole frame, i. e. communication module, database operation
module and so on. Experiments prove the rationality of design.
引文
[1]易继锴，侯媛彬．智能控制技术．北京：北京工业大学出版社，1999．
[2]易继锴．现代控制系统设计．北京：北京工业大学出版社，1991．
[3]傅京孙．人工智能及其应用．北京：清华大学出版社，1987．
[4]李士勇．模糊控制、神经控制和智能控制论．哈尔滨：哈尔滨工业大学出版社，1998．
[5]沈美明，温冬婵．IBM-PC汇编语言程序设计．北京：清华大学出版社，1991．
[6]田盛丰．人工智能原理与应用．北京：北京理工大学出版社，1993．
[7]李华．MCS-51系列单片机实用接口技术．北京：北京航空航天大学出版社，1993．
[8]赵振宇．模糊理论和神经网络的基础与应用．北京：清华大学出版社，1996．
[9]L.C.埃杰布赖奇特．IBMPC 接口技术．北京：海洋出版社，1988．
[10]蔡自兴．智能控制-基础与应用．北京：国防工业出版社，1998．
[11]李晶、李毓麟．智能大厦．上海：复旦大学出版社，1997．
[12]而师玛乃·花铁森．火灾报警器．北京：原子能出版社，1995．
[13]王殊、杨宗凯、何建华．神经网络模糊推理系统在火灾探测中的应用．数据采集与处理，1998，13(2)：6．
[14]何振亚主编．神经智能．湖南：湖南科学技术出版社，1997．
[15]刘君华．智能传感器系统．西安：西安电子科技大学出版社，1999．
[16]Nonlinear neural network:Principles,mechanism,and architectures.1988.Neural Networks:vol.1 17-61.
[17]C.C. Lee: Fuzzy logic in control systems. IEEE trans. Syst. Man Cybern., vol. 20, no. 2, pp404-435,1990.
[18]LiMin Fu. Rule Generation from Neural Networks. IEEE TRANSACTIONS ON SYSTEMS,MAN,AND CYBERNETICS,VOL 24 NO.8,AUGUST 1994.
[19]D.Park,A.Kandel and G.Langholz. "Genetic_Based New Fuzzy Reasoning Models with Application to Fuzzy Control". IEEE Transactions on Systems,Man,and Cybernetics,Vol.24,1994.1.
[20]Michio Sugeno and G. T. Kang: Structure identification of fuzzy model. Fuzzy Sets and Systems, vol.28, pp15-33,1988.
[21]James A. Villarreal: Fuzzy logic and network technologies. NASA CP-2491, pp 111-119, 1990.
[22]B.S.Zhang:Self-organizing fuzzy logic controller.IEE Proceedings-D, vol. 139.no.3, pp460-464, 1992.
[23]Timon Chin-Ting and Philip M.Wolfe:Amalgamation of neural networks and logic systems-a survey. Computers & Industrial Engineering, v 29 n 1-4 Sep 1 995, P 193-197.
[24]D.E.Golderg. Genetic Algorithm in Search, Optimization and Machine Learning. Addision-Wesley Pulishing Company, 1989.
[25]Volkinar Uebele, Shigeo Abe and Ming-Shong Lan: Neuralnetwork-based fuzzy classifier. IEEE Transactions on Systems, Man and Cybernetics, v 25 n 2 Feb 1 995, P 353-361.