航站楼碳浓度预测控制方法研究

发布时间：2018-03-11 15:11

本文选题：航站楼　切入点：碳浓度　出处：《中国民航大学》2017年硕士论文　论文类型：学位论文

【摘要】：航站楼作为旅客空陆交通的集散中心,容易在旅客流密度大区域出现碳浓度过高并影响旅客身体健康现象,故建立高精度的碳浓度预测模型来预测碳浓度未来变化趋势,并事先制定出碳浓度控制方案有其实际意义。由于航站楼内部人员流动性大、环境参数变化复杂,难以用机理模型精确的反映出碳浓度变化规律。为此,本文利用数据驱动方法对碳浓度时间序列建立组合预测模型,实现了对碳浓度变化趋势的精确预测,并提出将组合预测模型和模糊控制器相结合的航站楼碳浓度控制方案。本文研究的主要工作如下:首先,进行碳浓度数据采集系统设计,通过反复测试表明系统稳定性好、可靠性高、功耗低。其次,对从天津滨海国际机场T2航站楼H值机区采集的碳浓度数据进行数据特征分析,分析结果表明碳浓度数据序列具有非平稳性、非线性、低性噪比特征。然后,利用小波分析技术同基于PSO优化算法的SVR模型和ARMA模型预测方法相结合建立组合预测模型对碳浓度时间序列进行建模预测,并将其预测结果同单一SVR模型以及单一ARMA模型预测结果进行对比分析,其结果表明组合预测模型具有更高的预测精度。最后,设计模糊控制器并对其进行碳浓度控制仿真实验,并提出了将碳浓度组合预测模型和模糊控制器相结合的航站楼碳浓度控制方案。
[Abstract]:As the center of passenger air and land transportation, the terminal is prone to high carbon concentration in the area with large passenger flow density, which affects the health of passengers. Therefore, a high precision carbon concentration prediction model is established to predict the trend of carbon concentration change in the future. It is of practical significance to draw up the control scheme of carbon concentration in advance. Because of the large mobility of personnel inside the terminal building and the complex change of environmental parameters, it is difficult to accurately reflect the changing law of carbon concentration with the mechanism model. In this paper, a combined prediction model of carbon concentration time series is established by using data-driven method, which can accurately predict the trend of carbon concentration change. The main work of this paper is as follows: firstly, the design of carbon concentration data acquisition system is carried out, and the stability of the system is proved by repeated tests. Secondly, the data of carbon concentration collected from T 2 terminal H check-in area of Tianjin Binhai International Airport are analyzed. The results show that the sequence of carbon concentration data is nonstationary and nonlinear. Secondly, combining wavelet analysis with SVR model based on PSO optimization algorithm and ARMA model prediction method, a combined prediction model is established to model and predict the time series of carbon concentration. The prediction results are compared with those of single SVR model and single ARMA model. The results show that the combined prediction model has higher prediction accuracy. Finally, a fuzzy controller is designed and simulated for carbon concentration control. A control scheme of carbon concentration in terminal building is proposed, which combines the combined prediction model of carbon concentration with fuzzy controller.
【学位授予单位】：中国民航大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：V351.1;TU834.8

【参考文献】