论文部分内容阅读
为构建能针对建筑安全系统事故特征的事故预测模型,从气候、地域性、传统节日等方面分析我国房屋建筑及市政工程生产安全事故数据。基于灰色系统理论及季节变动预测方法,建立建筑安全事故灰色季节指数预测模型。用GM(1,1)模型作为季节指数预测的趋势方程,使不满1个整数周期的事故量统计数据参与建模,实现对模型的即时更新。结果表明:建筑安全事故时间序列有以指数变化为特征的趋势性和以年为周期的周期性;在建筑安全事故预测应用中,灰色季节指数预测模型平均预测准确度达到89%,比常规趋势方程季节指数预测方法预测精度提高5%以上。
In order to build an accident prediction model that can deal with the characteristics of accident of building safety system, this paper analyzes the data of production safety accident of building construction and municipal engineering from the aspects of climate, region and traditional festival. Based on gray system theory and seasonal variation forecasting method, a gray seasonal index forecasting model of building safety accident is established. Using the GM (1,1) model as the trend equation of the seasonal index forecast, the statistical data of accidents with less than one integer period are involved in modeling, and the model is updated immediately. The results show that the time sequence of building safety accidents has the trend of exponential change and the periodicity of annual cycle. In the application of building safety accident prediction, the average forecast accuracy of gray seasonal index forecasting model reaches 89%, which is higher than the conventional trend The prediction accuracy of the seasonal index forecasting method is improved by more than 5%.