复杂诊断系统的MAS协作求解方法

doi:10.3969/j.issn.1000-0801.2010.10.034

摘要/Abstract

摘要：

为适应故障诊断中动态变化的环境和满足求解复杂系统全部任务的要求，研究一种基于MAS的分布式智能监控系统。文中以诊断决策问题的结构及建模过程的层次性为基本思想，基于诊断任务的多层次混合分解策略，提出了融合不同表达方式和推理机制的多Agent并行诊断结构，分析了管理Agent、诊断Agent和决策Agent的基本职能，探讨了实时Agent的组织和演化机制，给出了并行推理过程中的冲突消解算法，建立了具有公共属性的智能体模块化层次结构，实现了基于MAS多层分布式黑板模型的智能监控系统。工程应用效果验证了多智能体结构框架下复杂系统分布式求解的可行性和有效性。

关键词: 复杂故障诊断, 多Agent系统, 分布式求解, 协同协作, 融合算法

Abstract:

To adapt the model in reconfiguring fault diagnosing to dynamic environment and the needs of solving the tasks of complex system fully，the paper introduced multi-agent and related technology to the complicated fault diagnosis，an integrated intelligent control system is studied in this paper. Based on the thought of the structure of diagnostic decision and hierarchy in modeling，based on multi-layer decomposition strategy of diagnosis task，a multi-agent synchronous diagnosis federation integrated different knowledge expression modes and inference mechanisms are presented，the functions of management agent， diagnosis agent and decision agent are analyzed，the organization and evolution of agents in the system are proposed，and the corresponding conflict resolution algorithm in given，layered structure of abstract agent with public attributes is build. System architecture is realized based on MAS distributed layered blackboard. The real world application shows that the proposed control structure successfully solves the fault diagnose problem of the complex plant，and the special advantage in the distributed domain.

Key words: fault diagnosis, multi-agent system, distributed solution, cooperation, amalgamation algorithm

蒋伟进,曹东波,许宇晖,史德嘉. 复杂诊断系统的MAS协作求解方法[J]. 电信科学, 2010, 26(10): 131-140.

Weijin Jiang,Dongbo Cao,Yuhui Xu,Dejia Shi. Research on Distributed Cooperation Solution Method of Complex Diagnosis System Based on MAS[J]. Telecommunications Science, 2010, 26(10): 131-140.

图/表 4

图1

图2

图3

表1

参考文献 18

1	Shahbazain E ， et al. Multi-agent data fusion workstation architecture. SPIE-the International Society for Optical Engineering， 1998（3376）:60～68
2	Jennigngs N R ， Vargs L Z ， Aarnts R P . Transforming standalone expert systems into a community of cooperating agents. Engineering Application Artificial Intelligence， 1993，6（4）:317～331
3	Jennings N R . Controlling cooperative problem solving in industrial multiage systems using joint intentions. Artificial Intelligence， 1995，75（2）:195～240
4	Draa B C ， Millot P . A framework for cooperative work：an approach based on the intentionality. Artificial Intelligence in Engineering， 1990，5（4）:199～205
5	Hartvigsen G ， MilJohansen D . Cooperation in a distributed artificial intelligence environment-the storm cast application. Engineering Application of Artificial Intelligence， 1990，3（3）:229～237
6	Russell S J . Provably bounded optimal agents. Journal of Artificial Intelligence Research， 1995，23（2）:575～609
7	Zhang J ， Collani Y V . An interactive assembly by a two-arm agent. Robotic and Autonomous Systems， 1999，29（12）:91～100
8	Hu Wenbing ， Wang Shaomei . Study on logistics decision supported system based on multi-agent. In：The 8th International Conference on Computer Supported Cooperative Work in Design， Xiamen，China, 2004
9	Daniel D C . Collaborating software blackboard and multi-agent systems ＆ the future. In：Proceedings of the International Lisp Conference， New York,October 2003
10	Grosz B J ， Kraus S . Collaborative plans for complex group action. Artificial Intelligence， 1996，86（2）:269～357
11	De Visser J ， Wieringa P A ， Moss J ， et al. Supporting distributed planning in a dynamic environment：an observational study in operating room management. In：Proc of the 21st European Conference on Human Decision Making and Control， Glasgow，Scotland, 2002
12	Subbu R ， Sanderson A C . Network based distributed planning for design and manufacturing. In：Proc of the 4th IEEE International Symposium on Assembly and Task Planning Soft Research Park， Fukuoka，Japan, 2001
13	Jiang Weijin . Research on distributed diagnoses method based on MAS. ACTA Electronica Sinica， 2004，32（12A）:235～237
14	Jiang Weijin . Research and implementation of distributed MSP algorithm based on GA ＆ MAS. Computer Science， 2002，29（9）:443～447
15	Jiang Weijin ， Xu Yusheng . Research on diagnosis model distributed intelligence and key technique based on MAS. Journal of Control Theory ＆ Applications， 2004，20（6）:231～231
16	朱孟潇，宋志伟，蔡庆生 . 一个基于模拟退火的多主体模型及其应用. 软件学报， 2004，15（4）： 537～544
17	刘贵全，陈小平，范焱等. 多主体协作系统的一种形式模型. 计算机学报， 2001，24（5）： 529～535
18	刘波，李伟，罗军舟等. 网络管理中多Agent 的半在线调度算法. 计算机研究与发展， 2006，43（4）： 571～578

故障模式	传感器			信度函数
故障模式	传感器	不平衡	不对中	零件松动	无故障	不确定度	诊断结果
不平衡	速度	0.2060	0.1519	0.1462	0.1326	0.3633	不确定
	加速度	0.3709	0.0907	0.2378	0.1024	0.1983	不确定
	模糊融合	0.2885	0.1213	0.1920	0.1175	0.2808	不确定
	本文方法	0.4098	0.1225	0.2394	0.1138	0.1144	不平衡
不对中	速度	0.2928	0.4680	0.2269	0.0092	0.0031	不对中
	加速度	0.1630	0.1679	0.1645	0.1243	0.3803	不确定
	模糊融合	0.2279	0.3180	0.1957	0.0067	0.1917	不对中
	本文方法	0.2917	0.4670	0.2269	0.0091	0.0022	不对中
零件松动	速度	0.1352	0.1411	0.3115	0.1103	0.3020	不确定
	加速度	0.1827	0.1769	0.2240	0.1035	0.3111	不确定
	模糊融合	0.1589	0.1590	0.2678	0.1069	0.3066	不确定
	本文方法	0.1873	0.1878	0.3599	0.1182	0.1443	零件松动

[1]	周杰, Esono Mikue Bernardo Esono, 王学英, 周惠婷, 罗宏. 基于SLM-PTS算法融合的NC-OFDM峰均比优化[J]. 电信科学, 2022, 38(7): 63-74.
[2]	张科,孙越佳,韩海. 基于协同过滤与社交网络混合算法的客户信用建模及授信方法[J]. 电信科学, 2020, 36(2): 52-60.