基于量子神经网络的超深层储层评价
|
摘要
为解决深层、超深层储层勘探与准确识别问题,论文提出一种基于量子BP神经网络的新方法。首先通过模拟量子受控非门的受控关系构造了量子神经元,然后将该神经元同BP算法相结合,构建出一种量子BP网络模型。针对油气田超深层储层评价问题,研究了基于量子BP网络的解决方案。结果表明,论文所设计的量子BP网络模型,对于复杂地质环境下的深层、超深层储层识别问题具有较强的适应性,在收敛速度和识别率两方面均明显优于普通BP网络。
In order to solve the problem of deep and ultra-deep reservoir exploration and accurate identification,a new method based on quantum BP neural network is proposed in this paper.Firstly,a quantum neuron is constructed by simulating the controlled relation of quantum controlled non-gate,and then a quantum BP network model is constructed by combining the neuron with BP algorithm.Results show that the quantum of the BP network model,the complex geological conditions of deep and ultra deep reservoir identification has strong adaptability,both in convergence speed and recognition rate are significantly superior to general BP network.
In order to solve the problem of deep and ultra-deep reservoir exploration and accurate identification,a new method based on quantum BP neural network is proposed in this paper.Firstly,a quantum neuron is constructed by simulating the controlled relation of quantum controlled non-gate,and then a quantum BP network model is constructed by combining the neuron with BP algorithm.Results show that the quantum of the BP network model,the complex geological conditions of deep and ultra deep reservoir identification has strong adaptability,both in convergence speed and recognition rate are significantly superior to general BP network.
引文
[1]黄娟,叶德燎,韩彧.超深层油气藏石油地质特征及其成藏主控因素分析[J].石油试验地质,2016,38(5):635-639.HUANG Juan,YE Deliao,HAN Xun. Petoleum geologyfeatures and accumulation controls for ultra-deep oil andgas reservoirs[J]. Petroleum test geology,2016,38(5):635-639.
[2]S. Kak. On quantum neural computing[J]. Information Sci-ences,1995,83(3-4):143-160.
[3]N. Ajit and M. Tammy. Quantum artificial neural networkarchitectures and components[J]. Information Sciences,2000,128(3-4):231-255.
[4]N. Kouda,N. Matsui,H. Nishimura. Qubit neural net-work and its learning efficiency[J]. Neural Computing&Applications,2005,14(2):114-121.
[5]Adenliton Silva,Wilson de Oliveira,Teresa Ludermir. AWeightless Neural Node based on a Probailistic QuantumMemory:Eleventh Brazilian Symposium on Neural Net-works. Sao Paulo,2000[C]//IEEE,2010:259-264.
[6]Adenliton J. da Silva,Wilson R. de Oliverira,Teresa B.Ludermir. Classical and superposed learning for quantumweightless neural network[J]. Neurocomputing,2012,75(1):52-60.
[7] Takahashi,Kazukiho;Kurokawa,Motoki;Hashimoto,Masafumi. Multi-layer quantum neural network controllertrained by real-coded genetic alogorithm[J]. neurocom-puting,2014,134:159-164.
[8]解光军,庄镇泉.量子神经网络[J].计算机科学,2001,28(7):1-6.XIE Guangjun,ZHUANG Zhenquan. Quantum neural net-work[J]. Computer science,2001,28(7):1-6.
[9]周日贵.量子神经网络模型研究[D].南京:南京航空航天大学,2008:12-24.ZHOU Rigui. Research on quantum neural network model[D]. Nanjing:Nanjing University of Aeronautics and As-tronautics,2008:12-24.
[10]陈珊琳,黄春晖.连续变量相干态量子神经网络模型的构建[J].量子电子学报,2017,34(4):467-472.CHEN Shanlin,HUANG Chunhui. Construction of con-tinuous variable coherent state quantum neural networkmodel[J]. Journal of quantum electronics,2017,34(4):467-472.
[11]李滨旭,姚姜虹.一种量子衍生神经网络模型[J].计算机系统应用,2016,25(8):206-210.LI Binxu,YAO Jianghong. A quantum derived neural network model[J]. Computer system applications,2016,25(8):206-210.
[12]Panchi Li,Hong X.Model and algorithmof quantum in-spired neural network with sequence input based on con-trolled rotation gates[J]. Application Intelligence,2014,40(1):107-126.
[13]LI P C and XIAO H. A hybrid quantum-inspired neuralnetworks sequence inputs[J]. Neurocomputing,2013,117:81-90.
[14]李盼池,周红岩.基于受控Hadamard门的量子神经网络模型及算法[J].计算机研究与发展,2015,52(1):211-220.LI Panchi,ZHOU Hongyan. Model and Algorithm ofQuantum Neural Network Based on the Controlled Had-amard Gates[J]. Journal of Computer Resesearch andDevelopment,2015,52(1):211-220.
[15]赵阳,孙学斌,周正.一种改进的量子神经网络频谱感知算法[J].通信系统与网络技术,2015,41(2):7-11.ZHAO Yang,SUN Xuebin,ZHOU Zheng. A SpectrumSensing Algorithm in Cognitive Radio Based on Im-proved Quantum Neural Network[J]. Communication Systems and Network Technology,2015,41(2):7-11
[16]张翼鹏,陈亮,郝欢.采用量子神经网络的音频水印新算法[J].信号处理,2013,29(6):684-690.ZHANG Yipeng,CHEN Liang,HAO Huan. A new au-dio watermarking algorithm based on quantum neural net-work is proposed[J]. Signal Processing,2013,29(6):684-690.
[17]唐彰国,李焕洲,张健.基于量子神经网络的网络攻击同源性判定方法[J].成都理工大学学报(自然科学版),2017,44(4):506-512.TANG Zhangguo,LI Huanzhou,ZHANG Jian. Networkattack homology determination method based on quantumneural network[J]. Journal of chengdu university of tech-nology(natural science edition),2017,44(4):506-512.
[18]徐春春,邹伟宏,杨跃明,等.中国陆上深层油气资源勘探开发现状及展望[J].天然气地球科学,2017,28(8):1139-1153.XU Chunchun,ZOU Weihong,YANG Yueming,et al.Current situation and prospect of exploration and exploi-tation of deep oil and gas resources in China[J]. NaturalGas Geoscience,2017,28(8):1139-1153.
[2]S. Kak. On quantum neural computing[J]. Information Sci-ences,1995,83(3-4):143-160.
[3]N. Ajit and M. Tammy. Quantum artificial neural networkarchitectures and components[J]. Information Sciences,2000,128(3-4):231-255.
[4]N. Kouda,N. Matsui,H. Nishimura. Qubit neural net-work and its learning efficiency[J]. Neural Computing&Applications,2005,14(2):114-121.
[5]Adenliton Silva,Wilson de Oliveira,Teresa Ludermir. AWeightless Neural Node based on a Probailistic QuantumMemory:Eleventh Brazilian Symposium on Neural Net-works. Sao Paulo,2000[C]//IEEE,2010:259-264.
[6]Adenliton J. da Silva,Wilson R. de Oliverira,Teresa B.Ludermir. Classical and superposed learning for quantumweightless neural network[J]. Neurocomputing,2012,75(1):52-60.
[7] Takahashi,Kazukiho;Kurokawa,Motoki;Hashimoto,Masafumi. Multi-layer quantum neural network controllertrained by real-coded genetic alogorithm[J]. neurocom-puting,2014,134:159-164.
[8]解光军,庄镇泉.量子神经网络[J].计算机科学,2001,28(7):1-6.XIE Guangjun,ZHUANG Zhenquan. Quantum neural net-work[J]. Computer science,2001,28(7):1-6.
[9]周日贵.量子神经网络模型研究[D].南京:南京航空航天大学,2008:12-24.ZHOU Rigui. Research on quantum neural network model[D]. Nanjing:Nanjing University of Aeronautics and As-tronautics,2008:12-24.
[10]陈珊琳,黄春晖.连续变量相干态量子神经网络模型的构建[J].量子电子学报,2017,34(4):467-472.CHEN Shanlin,HUANG Chunhui. Construction of con-tinuous variable coherent state quantum neural networkmodel[J]. Journal of quantum electronics,2017,34(4):467-472.
[11]李滨旭,姚姜虹.一种量子衍生神经网络模型[J].计算机系统应用,2016,25(8):206-210.LI Binxu,YAO Jianghong. A quantum derived neural network model[J]. Computer system applications,2016,25(8):206-210.
[12]Panchi Li,Hong X.Model and algorithmof quantum in-spired neural network with sequence input based on con-trolled rotation gates[J]. Application Intelligence,2014,40(1):107-126.
[13]LI P C and XIAO H. A hybrid quantum-inspired neuralnetworks sequence inputs[J]. Neurocomputing,2013,117:81-90.
[14]李盼池,周红岩.基于受控Hadamard门的量子神经网络模型及算法[J].计算机研究与发展,2015,52(1):211-220.LI Panchi,ZHOU Hongyan. Model and Algorithm ofQuantum Neural Network Based on the Controlled Had-amard Gates[J]. Journal of Computer Resesearch andDevelopment,2015,52(1):211-220.
[15]赵阳,孙学斌,周正.一种改进的量子神经网络频谱感知算法[J].通信系统与网络技术,2015,41(2):7-11.ZHAO Yang,SUN Xuebin,ZHOU Zheng. A SpectrumSensing Algorithm in Cognitive Radio Based on Im-proved Quantum Neural Network[J]. Communication Systems and Network Technology,2015,41(2):7-11
[16]张翼鹏,陈亮,郝欢.采用量子神经网络的音频水印新算法[J].信号处理,2013,29(6):684-690.ZHANG Yipeng,CHEN Liang,HAO Huan. A new au-dio watermarking algorithm based on quantum neural net-work is proposed[J]. Signal Processing,2013,29(6):684-690.
[17]唐彰国,李焕洲,张健.基于量子神经网络的网络攻击同源性判定方法[J].成都理工大学学报(自然科学版),2017,44(4):506-512.TANG Zhangguo,LI Huanzhou,ZHANG Jian. Networkattack homology determination method based on quantumneural network[J]. Journal of chengdu university of tech-nology(natural science edition),2017,44(4):506-512.
[18]徐春春,邹伟宏,杨跃明,等.中国陆上深层油气资源勘探开发现状及展望[J].天然气地球科学,2017,28(8):1139-1153.XU Chunchun,ZOU Weihong,YANG Yueming,et al.Current situation and prospect of exploration and exploi-tation of deep oil and gas resources in China[J]. NaturalGas Geoscience,2017,28(8):1139-1153.