摘要
目的对核电厂运行瞬态识别进行有效识别。方法基于动态时间弯曲算法,计算运行瞬态中温度、压力和流速数据与设计瞬态中对应数据之间的相似程度,通过定义等效相似度反应温度、压力和流速的不同权重。在完成当前运行瞬态与各条设计瞬态之间的相似程度后,通过比较相似度值,将运行瞬态归类为设计瞬态。结果通过对"华龙一号"核电厂的设计瞬态和基于设计瞬态摄动获得的虚拟运行瞬态进行验证,文中提出的方法能够快速有效地对运行瞬态进行分类,结果显示,有95%以上的运行瞬态能够被正确识别。结论基于动态时间弯曲算法和等效相似度建立的瞬态识别方法具有高效、准确等优点,能够有效应用于核电厂的疲劳监测系统。
Objective To identify the transients in the nuclear power plant effectively.Methods Based on the dynamic time warping algorithm,the similarity between the temperature,pressure and flow rate in the operational transient and these in the design transient was calculated,respectively.Meanwhile,an effective similarity for the transient was defined to represent the different weight factor of temperature/pressure and flow rate.After the effective similarities for all the design transients were calculated,the current operational transient can be classified in to the best similar design transient.Results The design transient of Hualong 1 nuclear power plant and virtual operation transient obtained based on disturbance of design transient verified that the method proposed in this paper could effectively classify the operation transient.In the verification,more than 95% transients could be identified correctly.Conclusion Based on the DTW based algorithm and effective similarity,a transient identification method is proposed in present paper,this method is effective and accurate,and can be used in the fatigue monitor system of nuclear power plant.
引文
[1]MOL A C A,MARTINEZ A S,SCHIRRU R.A Neural Model for Transient Identifi-cation in Dynamic Processes With“Do Not Know”Response[J].Annals of Nuclear Energy,2003,30:1365-1381.
[2]Mol A C A,Almeida J C S,Pereira C M N A,et al.Neural and Genetic-based Approaches to Nuclear Transient Identification Including“Don′t Know”Response[J].Progress in Nuclear Energy,2006,48:268-282.
[3]MEDEIROS J A C C,SCHIRRU R..Identification of Nuclear Power Plant Transients Using the Particle Swarm Optimization algorithm[J].Annals of Nuclear Energy,2008,35:576-582.
[4]GOTTLIEB C,ARZHANOV V,GUDOWSKI W.Feasibility Study on Transient Identification in Nuclear Power Plants Using Support Vector machines[J].Nuclear Technology,2006,155:67-77.
[5]KWON K C.HMM-based Transient Identification in Dynamic Process[J].Trans-actions on Control,Automation,and Systems Engineering,2000(2):40-46.
[6]KWON K C,KIM J H.Accident Identification in Nuclear Power Plants Using Hidden Markov Models[J].Engineering Applications of Artificial Intelligence,1999,12:491-501.
[7]KWON K C,KIM J H,SEONG P H.Hidden Markov Model-based Real-time Transient Identifications in Nuclear Power Plants[J].International Journal of Intelligent Systems,2002,17:791-811.
[8]常远,郝轶,黄晓津,等.基于模糊熵的核电站瞬态识别方法[J].原子能科学技术,2014,48(9):1640-1645.
[9]常远,黄晓津,李春文,等.基于模糊熵的核电站瞬态分段识别方法[J].核动力工程,2014,35(1):106-109.