基于人工免疫算法的电梯群控系统的研究与设计
|
摘要
在现代化的智能大厦中,由于交通流量特别大,不确定因素多,因而往往在大厦中配置多台电梯,电梯群控系统采用优化的控制策略来协调多台电梯的运行,以提高电梯的运行效率和服务质量。随着人工智能控制技术的发展,电梯群控系统的目标不仅限于缩短乘客乘梯时间,而是对乘客乘梯时间以及系统能耗等多个不同目标进行优化。本文针对电梯客流高峰期的特点及存在的问题,设计了基于人工免疫算法的电梯群控系统。
人工免疫系统是人工智能中的一种新兴的方法,它具有数据特征提取的作用。本文采用人工免疫算法对两组模拟电梯客流特点的数据进行了特征提取,得到很好的效果。另外,在对人工免疫算法的深入研究中,本文发现基于克隆选择学的人工免疫算法在解决优化问题时存在一定的不足,主要是在保持最大亲和力方面并不总能得到令人满意的曲线。本文针对此问题进行了仔细的分析,找出原因并对基于克隆选择学的人工免疫算法进行了改进,通过仿真验证得到了满意的效果。
在高峰客流期间,分区是解决电梯群调度问题的一种很好方法。但是,现存的分区控制模式是固定的或按时间预先确定的,它们不能适应实际的交通模式,这也使得一些人对电梯的分区研究持有反对的态度。动态分区具有很大的灵活性,它在增大载客量,降低高峰期乘客的候梯时间、乘梯时间,降低能耗等方面都有很大的优越性。然而,如何快速地确定各台电梯的最优的动态分区方式,是问题的难点。经过本文的研究,人工免疫算法恰好可以很好地解决此问题。本文将人工免疫算法应用于电梯动态分区过程,并进行了仿真实验,验证了其可行性与有效性,使得电梯动态分区的思想得以很好地实现。
In modern intelligent buildings, people always equip for a building several elevators because of big flux of passengers and varities of factors uncertian. Elevator Group Control System manages the operation of the elevators by optimizing control strategy in order to enhance running efficiency of elevator and improve service quality. With the development of artificial intelligent control technology, the objective of elevator group control system has not limited to reducing waiting time, some other objectives such as reducing riding time and reducing the power consumption are also taken into consideration. The thesis designs a Elevator Group Control System based on Artificial Immune Algorithm, in order to suit for the feature of peak passenger flow time and solve the problem remaining during this time.
Artificial Immune System is a newly emerging method'in Artificial Intelligence. It has data feature extraction function.The thesis does a simulation experiment of feature extraction to two groups the data which simulate the elevator passenger flow using the Artificial Immune Algorithm,whose effect is well. In the deeply research of Artificial Immune Algorithm, however, the thesis found that, there is some insufficiency when the thesis solutes optimization question with Artificial Immune Algorithm based on Clonal Selective Theory. Mainly, the insufficiency is that the thesis couldn't always get a satisfying curve about the aspect of keeping the maxmum affinity. The thesis analyses the problem carefully and find out the reason, then, improves the Artificial Immune Algorithm based on Clonal Selective Theory. Through the simulation, the thesis gets a satisfying effect.
On peak passenger flow time, zoning is a effictive method on soluting elevator group scheduling problem.However, the zoning method that we use now is fixed or determined according to the time in advance. They cannot adapt the actual transportation pattern. It also makes some people to have the opposition manner to zoning research. Dynamic zoning has the very big flexibility, it also has the very big superiority on the aspect of increasing the seating capacity, reducing the waiting time and riding time, reducing the energy consumption and so on. Whereas, how to fix the zoning mode quickly is the difficulty of the problem.The thesis found that the Artificial Immune Algorithm may solve it well exactly. The thesis uses the Artificial Immune Algorithm in the elevator dynamic zoning process, and carrys on a simulation experment.At last the thesis confirms its feasibility and validity, then makes the elevator dynamic idea ot be able to realize well.
人工免疫系统是人工智能中的一种新兴的方法,它具有数据特征提取的作用。本文采用人工免疫算法对两组模拟电梯客流特点的数据进行了特征提取,得到很好的效果。另外,在对人工免疫算法的深入研究中,本文发现基于克隆选择学的人工免疫算法在解决优化问题时存在一定的不足,主要是在保持最大亲和力方面并不总能得到令人满意的曲线。本文针对此问题进行了仔细的分析,找出原因并对基于克隆选择学的人工免疫算法进行了改进,通过仿真验证得到了满意的效果。
在高峰客流期间,分区是解决电梯群调度问题的一种很好方法。但是,现存的分区控制模式是固定的或按时间预先确定的,它们不能适应实际的交通模式,这也使得一些人对电梯的分区研究持有反对的态度。动态分区具有很大的灵活性,它在增大载客量,降低高峰期乘客的候梯时间、乘梯时间,降低能耗等方面都有很大的优越性。然而,如何快速地确定各台电梯的最优的动态分区方式,是问题的难点。经过本文的研究,人工免疫算法恰好可以很好地解决此问题。本文将人工免疫算法应用于电梯动态分区过程,并进行了仿真实验,验证了其可行性与有效性,使得电梯动态分区的思想得以很好地实现。
In modern intelligent buildings, people always equip for a building several elevators because of big flux of passengers and varities of factors uncertian. Elevator Group Control System manages the operation of the elevators by optimizing control strategy in order to enhance running efficiency of elevator and improve service quality. With the development of artificial intelligent control technology, the objective of elevator group control system has not limited to reducing waiting time, some other objectives such as reducing riding time and reducing the power consumption are also taken into consideration. The thesis designs a Elevator Group Control System based on Artificial Immune Algorithm, in order to suit for the feature of peak passenger flow time and solve the problem remaining during this time.
Artificial Immune System is a newly emerging method'in Artificial Intelligence. It has data feature extraction function.The thesis does a simulation experiment of feature extraction to two groups the data which simulate the elevator passenger flow using the Artificial Immune Algorithm,whose effect is well. In the deeply research of Artificial Immune Algorithm, however, the thesis found that, there is some insufficiency when the thesis solutes optimization question with Artificial Immune Algorithm based on Clonal Selective Theory. Mainly, the insufficiency is that the thesis couldn't always get a satisfying curve about the aspect of keeping the maxmum affinity. The thesis analyses the problem carefully and find out the reason, then, improves the Artificial Immune Algorithm based on Clonal Selective Theory. Through the simulation, the thesis gets a satisfying effect.
On peak passenger flow time, zoning is a effictive method on soluting elevator group scheduling problem.However, the zoning method that we use now is fixed or determined according to the time in advance. They cannot adapt the actual transportation pattern. It also makes some people to have the opposition manner to zoning research. Dynamic zoning has the very big flexibility, it also has the very big superiority on the aspect of increasing the seating capacity, reducing the waiting time and riding time, reducing the energy consumption and so on. Whereas, how to fix the zoning mode quickly is the difficulty of the problem.The thesis found that the Artificial Immune Algorithm may solve it well exactly. The thesis uses the Artificial Immune Algorithm in the elevator dynamic zoning process, and carrys on a simulation experment.At last the thesis confirms its feasibility and validity, then makes the elevator dynamic idea ot be able to realize well.
引文
1.朱昌明,毕晓亮.电梯智能群控系统研究概况[J],现代城市研究电梯专刊,2003,1-2.
2. Kuo-ming Chang, Tzong-lin Wu. More analyses for the service performance of elevator group systems [J], SAMS,2002,42(4):539-557.
3. Seng Hong, Gary Thiele and Vipperla Venkayya. Stochastic Mutation Algorithm in Multi-objective Design Optimization[J], IEEE Circuits and Systems,2006(2),41-45.
4. De Castro L N, Von Zuben F J. Artificial Immune Systems:Part Ⅰ Basic Theory and Application[J]. Technical Report-RT DC A,1999(1),89.
5.吕岗.免疫算法及其应用研究[D],北京：中国矿业大学,2003.
6.孙勇智,韦巍.人工免疫系统及其应用[J],计算机工程,2003,29(15)：1-3.
7. Sakai Y, Kurosawa K. Development of elevator supervisory group control system with artificial intelligence [J], Hitachi Review,1984,33(1):25-30.
8. K.Igarashi, S. Take and T. Ishikawa. Supervisory control for elevator group with fuzzy expert system[J]. Proceedings of the 1994 IEEE International Conference on Industrial Technology,1994,133-137.
9.朱德文,付国江.电梯群控技术[M],北京：中国电力出版社,2006.
10.Kenji Yoneda. Multi-objective elevator supervisory control system with individual floor-situation control [J], Elevator World,1999,47(5):90-95.
11. P. J. Costa Branco, J.A. Dente, R. Vilela Mendes. Using immunology principles for fault detection [J], IEEE Transactions on industrial electronics,2003,50(2):362-373.
12.杨祯山,邵诚.电梯群控技术的现状与发展方向[J],控制与决策,2005,20(12)：1321-1331.
13. Takashi Nagatani. Complex behavior of elevators in peak traffic [J], PHYSICA A,2003, 326:556-566.
14. Crites R H, Barto A G. Elevator group control using multiple reinforcement learning Agents [J], Machine Learning,1998,33(2-3):235-262.
15. A. T. P. So, J. K. L. Yu and W. L. Chan. Dynamic zoning based supervisory control for elevators[J], Proceedings of the 1999 IEEE International Conference on Control Applications,1999,1591-1596.
16.许玉格,罗飞,曹建忠.目的层预约的模糊神经网络电梯群控策略[J],华南理工大 学学报,2007,35(1)：13-18.
17. D. Castro and F. J. V. Zuben, Learning and Optimization Using the Clonal Selection Principle[J], IEEE Transactions on Evolutionary Computauton,2002,6(3),239-251.
18.郑延军,张惠侨.电梯动态分区算法及其遗传进化研究[J],计算机工程与应用,2001,(22)： 142-148.
19. C.B.Kim, K. A. Seong, H. Lee-Kwang and J.O. Kim. Design and Implementation of a fuzzy elevator group control system[J], IEEE Transactions on Systems,Man,and Cybernetics, Part A,1998,28(3):277-287.
20. Zhonghua Li, Zongyuan Mao and Jianping Wu. Research on Dynamic Zoning of Elevator Traffic Based on Artificial Immune Algorithm[J], Procedings of the Eighth International Conference on Control, Automation, Robotics and Vision Kunming, China,2004, 2170-2175.
21.宋涵.电梯群控系统智能控制的研究[D],浙江：浙江工业大学,2005,1.
22.林琳.群控电梯多目标智能最优调度算法的研究[D],沈阳：东北大学,2005.
23. G. C. Barney, S. M. dos Santos. Elevator traffic analysis, design and control[M], London: Peter Peregrinus,1985.
24. Y. Sogawa, T. Ishikawa, K. Igarashi. Supervisory control for elevator group by using fuzzy expert system which addresses the riding time[J], Proceedings of the 1996 IEEE International Conference on Industrial Electronics, Control, and Instrumentation. Taipei, 1996,419-424.
25.宗群,岳有军.电梯群控系统交通流的预测方法[J],系统工程与电子技术,2001,23(7)： 103-105.
26. C. B. Kim, K. A. Seong, H. Lee-Kwang and Jeong O. Kim. Design and Implementation of FEGCS:fuzzy elevator group control systems[J], Proceedings of the 1996 Conference of the North American on Fuzzy Information Processing. Berkeley,1996,109-113.
27. T. Ishikawa, A. Miyauchi and M. Kaneko. Supervisory control of elevator group by using fuzzy expert system which also addressing traveling time [J], Proceedings of the 2000 IEEE International Conference on Industrial Technology. Bangalore,2000,87-94.
28.张泽明.人工免疫算法及其应用研究[D],合肥：中国科学技术大学,2007,5.
29. D. Castro and J. Timmis, Artificial Immune Systems:A New Computational Intelligence Approach[J], London:Springer-Verlag,2002.
30. A.Tarakanov and D.Dasgupta. A Formal Model of an Artificial Immune System. BioSystem[J],2002(55),151-158.
31.宗群,童玲,薛丽华.电梯群控系统智能优化调度方法的研究[J],控制与决策,2004,19(8)：939-942.
32.刘园园.基于多Agent的电梯群控系统的研究与设计[D],沈阳：东北大学,2008.
33. L. N. de Castro and F. J. Von Zuben. An evolutionary immune system network for data clustering[J], Procedings of the Sixth Brazilian Symposium on Neural Networks, Rio de Janeiro,2000,84-89.
34.宗群,尚晓光.电梯群控系统的交通模式识别[J],控制与决策,2001,16(2)：163-166.
35. L. N. de Castro and J. Timmis. An artificial immune network for multimodal function optimization[J], Proceedings of the 2002 International Conference on Evolutionary computation, Honolulu,2002,699-674.
36.肖人彬,王磊.人工免疫系统：原理、模型、分析及展望[J],计算机学报,2002,25(12)：1282-1291.
37.莫洪伟.人工免疫系统原理及应用[M],哈尔滨：哈尔滨工业大学出版社,2001.
38.Chin-teng Lin, George Lee. Neural-network-based fuzzy logic control and decision system[J].IEEE Trans on Computers,1991,40(12):1320-1336.
39. Markon S. H, Kita and Y. Nishkawa. A adeptive optimum elevator group control by use of neural network[J], Transaction of the Institute of System,Control and Information Engineers,1994,7(12):487-497.
40.王国萍.多目标优化的电梯群控算法[D],武汉：华中科技大学,2005,11.
41.黄静,郭勇.对目前国内电梯控制技术的研究[J],电脑与信息技术,2000,5(17)：64-66.
42.郑延军,张惠侨.电梯群控系统客流分析与仿真[J],计算机工程与应用,2001,22(2)：139-141.
43. C. H. Wu, J. M. Ho and D. T. Lee. Travel-time prediction with support vector regression[J], IEEE Transactions on Intelligent Transportation Systems,2004,5(4):276-281.
44. A.L. Ding, X.M. Zhao and L.C. Jiao. Traffic flow time series prediction based on statistics learning theory[J], Proceedings of the IEEE 5th International Conference on Intelligent Transportation Systems,2002,727-730.
45.李中华.垂直交通客流分析与电梯群控制优化研究[D],广州：华南理工大学,2005.
46.孙勇智.人工免疫系统能够模型、算法及其应用研究[D],浙江：浙江大学,2004.
47. Yasuhiro Ogoshi, Haruhiko Kimura et al. Elevator group control system using multiagent system[J], Systems and Computers in Japan,2003,34(1):191-202.
48.李涛.计算机免疫学[M],北京,电子工业出版社,2004.
49.莫洪伟.人工免疫系统原理及应用[M],哈尔滨：哈尔滨工业大学出版社,2001.
50.李中华,朱燕飞,李春华.面向对象的电梯群控系统仿真平台开发与实现[J],计算机仿真,2005,22(2)：156-159.
51. Xiaoke Yan, Yu Liu and Zongyuan Mao. SVM-based Elevator Traffic Flow Prediction[J], Proceedings of the 6th Word Congress on Intelligent Control and Automation, Dalian, China,2006,8814-8818.
2. Kuo-ming Chang, Tzong-lin Wu. More analyses for the service performance of elevator group systems [J], SAMS,2002,42(4):539-557.
3. Seng Hong, Gary Thiele and Vipperla Venkayya. Stochastic Mutation Algorithm in Multi-objective Design Optimization[J], IEEE Circuits and Systems,2006(2),41-45.
4. De Castro L N, Von Zuben F J. Artificial Immune Systems:Part Ⅰ Basic Theory and Application[J]. Technical Report-RT DC A,1999(1),89.
5.吕岗.免疫算法及其应用研究[D],北京：中国矿业大学,2003.
6.孙勇智,韦巍.人工免疫系统及其应用[J],计算机工程,2003,29(15)：1-3.
7. Sakai Y, Kurosawa K. Development of elevator supervisory group control system with artificial intelligence [J], Hitachi Review,1984,33(1):25-30.
8. K.Igarashi, S. Take and T. Ishikawa. Supervisory control for elevator group with fuzzy expert system[J]. Proceedings of the 1994 IEEE International Conference on Industrial Technology,1994,133-137.
9.朱德文,付国江.电梯群控技术[M],北京：中国电力出版社,2006.
10.Kenji Yoneda. Multi-objective elevator supervisory control system with individual floor-situation control [J], Elevator World,1999,47(5):90-95.
11. P. J. Costa Branco, J.A. Dente, R. Vilela Mendes. Using immunology principles for fault detection [J], IEEE Transactions on industrial electronics,2003,50(2):362-373.
12.杨祯山,邵诚.电梯群控技术的现状与发展方向[J],控制与决策,2005,20(12)：1321-1331.
13. Takashi Nagatani. Complex behavior of elevators in peak traffic [J], PHYSICA A,2003, 326:556-566.
14. Crites R H, Barto A G. Elevator group control using multiple reinforcement learning Agents [J], Machine Learning,1998,33(2-3):235-262.
15. A. T. P. So, J. K. L. Yu and W. L. Chan. Dynamic zoning based supervisory control for elevators[J], Proceedings of the 1999 IEEE International Conference on Control Applications,1999,1591-1596.
16.许玉格,罗飞,曹建忠.目的层预约的模糊神经网络电梯群控策略[J],华南理工大 学学报,2007,35(1)：13-18.
17. D. Castro and F. J. V. Zuben, Learning and Optimization Using the Clonal Selection Principle[J], IEEE Transactions on Evolutionary Computauton,2002,6(3),239-251.
18.郑延军,张惠侨.电梯动态分区算法及其遗传进化研究[J],计算机工程与应用,2001,(22)： 142-148.
19. C.B.Kim, K. A. Seong, H. Lee-Kwang and J.O. Kim. Design and Implementation of a fuzzy elevator group control system[J], IEEE Transactions on Systems,Man,and Cybernetics, Part A,1998,28(3):277-287.
20. Zhonghua Li, Zongyuan Mao and Jianping Wu. Research on Dynamic Zoning of Elevator Traffic Based on Artificial Immune Algorithm[J], Procedings of the Eighth International Conference on Control, Automation, Robotics and Vision Kunming, China,2004, 2170-2175.
21.宋涵.电梯群控系统智能控制的研究[D],浙江：浙江工业大学,2005,1.
22.林琳.群控电梯多目标智能最优调度算法的研究[D],沈阳：东北大学,2005.
23. G. C. Barney, S. M. dos Santos. Elevator traffic analysis, design and control[M], London: Peter Peregrinus,1985.
24. Y. Sogawa, T. Ishikawa, K. Igarashi. Supervisory control for elevator group by using fuzzy expert system which addresses the riding time[J], Proceedings of the 1996 IEEE International Conference on Industrial Electronics, Control, and Instrumentation. Taipei, 1996,419-424.
25.宗群,岳有军.电梯群控系统交通流的预测方法[J],系统工程与电子技术,2001,23(7)： 103-105.
26. C. B. Kim, K. A. Seong, H. Lee-Kwang and Jeong O. Kim. Design and Implementation of FEGCS:fuzzy elevator group control systems[J], Proceedings of the 1996 Conference of the North American on Fuzzy Information Processing. Berkeley,1996,109-113.
27. T. Ishikawa, A. Miyauchi and M. Kaneko. Supervisory control of elevator group by using fuzzy expert system which also addressing traveling time [J], Proceedings of the 2000 IEEE International Conference on Industrial Technology. Bangalore,2000,87-94.
28.张泽明.人工免疫算法及其应用研究[D],合肥：中国科学技术大学,2007,5.
29. D. Castro and J. Timmis, Artificial Immune Systems:A New Computational Intelligence Approach[J], London:Springer-Verlag,2002.
30. A.Tarakanov and D.Dasgupta. A Formal Model of an Artificial Immune System. BioSystem[J],2002(55),151-158.
31.宗群,童玲,薛丽华.电梯群控系统智能优化调度方法的研究[J],控制与决策,2004,19(8)：939-942.
32.刘园园.基于多Agent的电梯群控系统的研究与设计[D],沈阳：东北大学,2008.
33. L. N. de Castro and F. J. Von Zuben. An evolutionary immune system network for data clustering[J], Procedings of the Sixth Brazilian Symposium on Neural Networks, Rio de Janeiro,2000,84-89.
34.宗群,尚晓光.电梯群控系统的交通模式识别[J],控制与决策,2001,16(2)：163-166.
35. L. N. de Castro and J. Timmis. An artificial immune network for multimodal function optimization[J], Proceedings of the 2002 International Conference on Evolutionary computation, Honolulu,2002,699-674.
36.肖人彬,王磊.人工免疫系统：原理、模型、分析及展望[J],计算机学报,2002,25(12)：1282-1291.
37.莫洪伟.人工免疫系统原理及应用[M],哈尔滨：哈尔滨工业大学出版社,2001.
38.Chin-teng Lin, George Lee. Neural-network-based fuzzy logic control and decision system[J].IEEE Trans on Computers,1991,40(12):1320-1336.
39. Markon S. H, Kita and Y. Nishkawa. A adeptive optimum elevator group control by use of neural network[J], Transaction of the Institute of System,Control and Information Engineers,1994,7(12):487-497.
40.王国萍.多目标优化的电梯群控算法[D],武汉：华中科技大学,2005,11.
41.黄静,郭勇.对目前国内电梯控制技术的研究[J],电脑与信息技术,2000,5(17)：64-66.
42.郑延军,张惠侨.电梯群控系统客流分析与仿真[J],计算机工程与应用,2001,22(2)：139-141.
43. C. H. Wu, J. M. Ho and D. T. Lee. Travel-time prediction with support vector regression[J], IEEE Transactions on Intelligent Transportation Systems,2004,5(4):276-281.
44. A.L. Ding, X.M. Zhao and L.C. Jiao. Traffic flow time series prediction based on statistics learning theory[J], Proceedings of the IEEE 5th International Conference on Intelligent Transportation Systems,2002,727-730.
45.李中华.垂直交通客流分析与电梯群控制优化研究[D],广州：华南理工大学,2005.
46.孙勇智.人工免疫系统能够模型、算法及其应用研究[D],浙江：浙江大学,2004.
47. Yasuhiro Ogoshi, Haruhiko Kimura et al. Elevator group control system using multiagent system[J], Systems and Computers in Japan,2003,34(1):191-202.
48.李涛.计算机免疫学[M],北京,电子工业出版社,2004.
49.莫洪伟.人工免疫系统原理及应用[M],哈尔滨：哈尔滨工业大学出版社,2001.
50.李中华,朱燕飞,李春华.面向对象的电梯群控系统仿真平台开发与实现[J],计算机仿真,2005,22(2)：156-159.
51. Xiaoke Yan, Yu Liu and Zongyuan Mao. SVM-based Elevator Traffic Flow Prediction[J], Proceedings of the 6th Word Congress on Intelligent Control and Automation, Dalian, China,2006,8814-8818.