| 引用本文: | 丁宝,李庆超,张永明,张进,齐维贵.适于双子电梯群控系统的交通模式预测方法[J].哈尔滨工业大学学报,2013,45(8):79.DOI:10.11918/j.issn.0367-6234.2013.08.012 |
| DING Bao,LI Qingchao,ZHANG Yongming,ZHANG Jin,QI Weigui.Traffic mode prediction method for twin elevator group control system[J].Journal of Harbin Institute of Technology,2013,45(8):79.DOI:10.11918/j.issn.0367-6234.2013.08.012 |
|
| 摘要: |
| 为了避免传统电梯交通模式识别存在模式滞后的缺陷,适应新型的双子电梯群控系统性能要求,提出了基于预测交通流的电梯交通模式预测方法．采用支持向量机(Support vector machine,SVM)进行电梯交通流预测;采用BP神经网络进行电梯交通流模式识别,并用遗传算法(Genetic algorithm,GA)对BP神经网络进行优化;将电梯交通流预测与交通模式识别相结合,再次利用神经网络对所预测的交通流进行模式识别,实现交通模式预测．研究结果表明,预测交通流的交通模式与实际交通流的交通模式一致,验证了交通模式预测的准确性．交通模式预测方法可避免模式滞后的缺陷,为双子电梯群控系统工程应用提供理论依据． |
| 关键词: 双子电梯 交通模式识别 交通流预测 模式预测 |
| DOI：10.11918/j.issn.0367-6234.2013.08.012 |
| 分类号: |
| 基金项目:国家自然科学基金资助项目(51207035); 黑龙江省自然科学基金资助项目(E201116, F201207); 中国博士后科学基金资助项目(2012M510953); 中央高校基本科研业务费专项资金资助项目(HIT.NSRIF.2012012). |
|
| Traffic mode prediction method for twin elevator group control system |
|
DING Bao, LI Qingchao, ZHANG Yongming, ZHANG Jin, QI Weigui
|
|
(School of Electrical Engineering and Automation,Harbin Institute of Technology, 150001 Harbin, China)
|
| Abstract: |
| As the traditional elevator traffic pattern recognition exists mode lag and it is difficult to adapt the performance requirements of group control system for the new type of twin elevator, this paper presents an elevator traffic mode prediction method which is based on traffic flow prediction. The support vector machine(SVM) is used for elevator traffic flow prediction, the BP neural network for elevator traffic flow pattern recognition and the genetic algorithms(GA) is adopted to optimize the BP neural network. Meanwhile, by fusing elevator traffic flow prediction and traffic pattern recognition together, the neural network is utilized to recognize the pattern of the predicted traffic flow again, thereby the pattern prediction is realized. Research results show that the traffic pattern of predicted traffic flow is consistent with the traffic pattern of actual traffic flow,which verifies the accuracy of traffic pattern prediction, avoids the defect of mode lag and provides a theoretical basis for engineering application of twin elevator group control system. |
| Key words: twin elevator elevator traffic pattern recognition traffic flow prediction pattern prediction |
