基于量子蛙跳算法和过程神经网络的抽油机故障诊断

中国机械工程 ›› 2014, Vol. 25 ›› Issue (12): 1609-1615.

基于量子蛙跳算法和过程神经网络的抽油机故障诊断

张强;许少华;李盼池

东北石油大学,大庆,163318

出版日期:2014-06-26 发布日期:2014-06-27
基金资助:
国家自然科学基金资助项目(61170132) ；中国博士后科学基金资助项目(20090460864)；黑龙江省教育厅科学技术研究资助项目(11551015)

Pumping Unit Fault Diagnosis Based on Quantum Shuffled Frog Leaping Algorithm and Process Neural Networks

Zhang Qiang;Xu Shaohua;Li Panchi

Northeast Petroleum University,Daqing,Heilongjiang,163318

Online:2014-06-26 Published:2014-06-27
Supported by:
National Natural Science Foundation of China（No. 61170132）;Supported by China Postdoctoral Science Foundation（No. 20090460864）;Heilongjiang Provincial Science and Technology Research Program of Ministry of Education of China（No. 11551015）

摘要/Abstract

摘要：

提出了一种量子混合蛙跳算法，该算法采用量子位的Bloch球面坐标编码个体，利用量子位在Bloch球面上绕轴旋转的方法实施优化搜索，采用Hadamard门实现个体变异以避免早熟，增强解空间的遍历性，可以快速逼近全局最优解。对过程神经网络的网络结构、网络参数和展开项数统一编码，并利用该算法进行优化，把优化后的神经网络应用到抽油机故障诊断中，结果表明，用量子混合蛙跳算法优化的神经网络对抽油机进行故障诊断较传统BP算法更具准确性与快速性。

关键词: 过程神经网络, 混合蛙跳算法, 示功图, 故障诊断

Abstract:

A quantum shuffled frog leaping algorithm(QSFLA) was presented herein. In this algorithm, the individuals were expressed with Bloch spherical coordinates of qubits, the evolution search was realized with the rotation of qubits in Bloch sphere. The mutation of individuals was achieved with Hadamard gates to avoid premature convergence. Above operations enhanced the ergodicity of the solution space and approximate global optimal solution fast. The network structure, network parameters and expand the number of items of PNN were encoded uniformly, and were optimized by the QSFLA. The optimizated neural network was used in pumping unit fault diagnosis. The diagnostic results between the new QSFLA and BP algorithm were compared. The conclusion is that the PNN based on QSFLA has better training performance, faster convergence rate and higher accuracy.

Key words: process neural network(PNN), shuffled frog leaping algorithm, indicator diagram, fault diagnosis

中图分类号:

TP183

张强, 许少华, 李盼池. 基于量子蛙跳算法和过程神经网络的抽油机故障诊断[J]. 中国机械工程, 2014, 25(12): 1609-1615.

Zhang Qiang, Xu Shaohua, Li Panchi. Pumping Unit Fault Diagnosis Based on Quantum Shuffled Frog Leaping Algorithm and Process Neural Networks[J]. China Mechanical Engineering, 2014, 25(12): 1609-1615.

参考文献

[1]任伟建,陶琳.基于粒子群算法的抽油机故障诊断研究[J].系统仿真学报,2012,24(2):482-487.
[2]王洪岩,乔磊,樊瑞筱.基于不变矩的示功图故障诊断方法研究[J].科学技术与工程,2012,20(17):4308-4311.
[3]李正勤,黎洪生.基于示功图面积变化的抽油机故障诊断模型[J].油气田地面工程,2008,27(9):3-4.
[4]潘志坚,葛家理.自适应神经网络示功图识别[J].石油学报,1996,17(3):104-109.
[5]杨理践,徐世文,高松巍.基于BP神经网络的有杆抽油系统故障智能判断[J].沈阳工业大学学报,2004,26(6):667-669.
[6]段玉波,谭金库,张力鹏,等.基于神经网络的抽油机故障诊断[J].自动化技术与应用,2012(4):1-3.
[7]张强,许少华.智能动态诊断模型及在示功图识别中的应用[J].计算机工程与应用,2009,45(4):215-217.
[8]何新贵,许少华著.过程神经元网络[M],2007:189.
[9]许少华,何新贵,刘坤,等.关于连续过程神经元网络的一些理论问题[J].电子学报,2006,34(10):1838-1841.
[10]许少华 ,何新贵.基于函数正交基展开的过程神经网络学习算法[J].计算机学报,2004,27(5):645-650.
[11]许少华,蔡月芹,谢树民.基于傅立叶函数基变换的过程神经元网络学习算法[J].计算机工程与设计,2006,27(23):4413-4415.
[12]李盼池,许少华.基于连续沃尔什变换的过程神经元网络训练[J].计算机工程与设计,2005,26(3):702-703.
[13]钟诗胜,朴树学,丁刚.改进BP算法在过程神经网络中的应用[J].哈尔滨工业大学学报,2006,38(6):840-842.
[14]张强,许少华,刘丽杰,等.基于遗传—模拟退火算法的过程神经网络的训练及应用[J].长春理工大学学报,2009(1):119-122.
[15]葛利.一种基于混合遗传算法学习的过程神经网络[J].哈尔滨工业大学学报,2005,37(7):986-988.
[16]Eusuff M M, Lansey K E. Optimization of Water Distributionnetwork Design Using the Shuffled Frog Leaping Algorithm[J]. Journal of Water Sources Planning and Management, 2003, 129 (3) :210-225.
[17]刘坤,谭营,何新贵.基于粒子群优化的过程神经网络学习算法[J].北京大学学报：自然科学版,2011,47(2):238-244.
[18]Elbeltagi E, Hegazy T. A Modified Shuffled Frog-leaping Optimization Algorithm Applications to Project Management[J]. Structure and Infrastructure Engineering,2007,3(1) :53-60.
[19]赵鹏军,刘三阳.求解复杂函数优化问题的混合蛙跳算法[J].计算机应用研究,2009(7):2435-2437.
[20]Ajit N, Mark M. Quantum-inspired Genetic Algorithms[C]//Proe, of IEEE Int. Conference on Ew olutionary Computation. Nagoya, 1996 : 61-66.
[21]郑仕链,楼才义,杨小牛.基于改进混合蛙跳算法的认知无线电协作频谱感知[J].物理学报,2010(5):3611-3617.
[22]Han K H, Park K H, Lee C H. Parallel Quantum-inspired Genetic Algorithmfor Combinatorial Optimization Problem[C]//Proe. of IEEE Int. Conference on Evolutionary Computation. Seoul, 2001 : 1422-1429.
[23]Han K H, Kim J H. Genetic Quantum Algorithm And Its Applicationto Combinational Optimization Problem[C]// Proc. of IEEE Int. Conference on Evolutionary Computation. La Jolla, 2000: 1354- 1360.
[24]汪鹏君,李辉,吴文晋,等.量子遗传算法在多输出Reed—Muller逻辑电路最佳极性搜索中的应用[J].电子学报, 2010, 38(5): 1058-1063. 4
[25]Han K H, Kim J H. Quantum-inspired Evolutionary Algorithm for a Class of Combinatorial Optimization[J]. IEEE Transactions on Evolutionary Computation, 2002,6(6): 580-593.