Intelligent Active Control Simulation System of Building Structures by Fuzzy Optimal Control

为了开发有效的建筑物控制系统，必须考虑建筑物的复杂性、未来荷载的不确定性等特点。本研究基于模糊理论、神经网络、遗传算法等智能系统，开发了建筑物抗地震荷载最优自适应预测控制仿真系统。这些系统具有以下特点；(1)用隶属函数描述目标响应和控制变量；(2)对地震输入进行实时预测和结构识别；(3)采用模糊最大化决策进行优化。本研究开发了以下两个系统，即；(1)基于智能系统的建筑结构模糊最优主动控制仿真系统；(2)基于遗传算法的建筑主动控制系统优化设计方法。这两个系统是在windows环境下用面向对象编程语言分别开发的。在此基础上，讨论了分布式环境下基于智能主动控制的综合仿真系统的开发方法。这里讨论以下两种方法，即；(1)模糊最优控制测试系统在分布式仿真系统中的应用；(2)基于internet浏览器的分布式仿真系统。采用前一种方法，便于在真实测试系统的基础上开发分布式仿真系统。然而，本测试系统是在DOS环境下开发的，难以通过图形化的方式开发出用户友好的系统。采用后一种方法，易于开发独立的用户友好型系统。然而，组件对象很难控制，因此作者无法将独立系统转换为可以在internet浏览器上执行的系统。通过本研究，作者可以得到分布式环境下建筑智能主动控制仿真系统的模糊优化逻辑的大量基础数据和系统。少

To develop the effective control system of buildings, it should be necessary to take account of special features such as complexity, uncertainty of future loading. In this research, optimal adaptive and predictive control simulation systems of buildings against earthquake loading are developed based on intelligent systems such as fuzzy theory, neural network, genetic algorithm and so on. These systems have following features ; (1) Target responses and control variables are described with membership function, (2) Real time prediction of earthquake input and structural identification are performed, and (3) Optimization is performed by fuzzy maximizing decision.In this research, following two systems are developed, i.e., ; (1) Simulation system of fuzzy optimal active control of building structures by intelligent systems, and (2) Optimal design method of active control system of buildings by genetic algorithm. These two systems are developed individually by object oriented programming lan … More guage under windows environment.In the next step, there are discussed developing methods of an integrated simulation system based on imtelligent active control under distributed environment. Here, following two methods are discussed, i.e. ; (1) Application of testing system of fuzzy optimal control to distributed simulation system and (2) Distributed simulation systems by internet browser. In the former method, it is easy to develop distributed simulation systems based on a real testing system. However, this testing system is developed under DOS environment, so it is difficult to develop user-friendly systems by graphical presentation. In the latter method, it is easy to develop user-friendly system as stand-alone one. However, it is difficult to control component objects, so authors can not translate stand-alone systems into ones which can execute on internet browser. Through this research, the authors can get a lot of basic data and systems on the intelligent active control simulation system of buildings by fuzzy optimal logic under distributed environments. Less