基于环境监测数据的源项估算技术研究
|
摘要
随着我国经济的持续发展和公众环境意识的不断增强,环境安全问题已经成为不可回避的社会焦点。对于固定设施的环境安全监控、处置,我国已经拥有一整套比较的成熟技术、方法。但对于未知地点、未知强度的危险气体释放事件的评估与处置,由于其释放及迁移过程具有较大地不确定性,使得决策者无法在突发事件早期采取有效的应急措施,导致整体应急响应行动滞后、应急行动效率低下。
本项研究的目的就在于：建立一套技术方案,通过特定的优化算法,使其能够有效利用环境监测数据,实现对环境大气中未知释放过程进行再现模拟。模拟结果可以是用以描述释放的释放参数的估计,也可以是对危险气体整个释放迁移过程及特定时刻危险气体地面浓度分布的再现。
在开展了针对国际上多个代表性的核辐射应急响应系统相关技术发展跟踪的同时,对卡尔曼滤波、贝叶斯推导、非参数化回归以及启发式算法等多种优化算法进行了充分调研,并以此为基础,首次提出利用遗传优化技术的良好鲁棒性,结合蒙特卡洛粒子扩散模式、逆向粒子扩散／轨迹模拟技术,以及皮尔森相关系数等统计分析方法,实现对未知释放参数(释放水平位置、释放高度、释放强度、释放起止时间)的估算以及气载污染物迁移扩散过程的再现。
模式开发与验算的基础来自近年来开展的多例高质量SF6野外大气扩散试验数据。数据包括地形高程数据、气象观测数据、边界层流场数据等。应用Forrtran语言实现了基于环境监测数据与气象数据的源项估算模式-Inverse开发,并应用多线程计算技术,使模式计算更为充分地利用计算资源,有效提高了模式计算效率
进而,本文通过对不同假定与真实条件下的数值模拟试验,对源项估算方案可行性、模式结果有效性、模式运行时效性等进行了分析。同时,通过对不同采样点布设的差异对模拟结果的影响开展了数值模拟试验,对该模式的应用条件进行了尝试,初步提出了敏感地区采样点布设的基本原则。最后,就下阶段Inverse模式的进一步完善提出了较系统的数值验证方案,为该模式在不同突发事件场景条件下的应用奠定基础。
综合上述研究成果,分析认为：
1、Inverse模式选择遗传算法为主要优化手段,结合扩散／轨迹模拟技术,利用PCC(皮尔森相关系数)、NMSE(标准均方差)作为适宜的评判因子搜寻源项参量的技术方案是合理可行的；
2、首次提出在模拟早期,利用逆向轨迹／扩散模式合理限定解域区间的技术路线。这种做法不仅可以有效缩小优化算法的搜索区间,提高模式运算效率,而且其模拟结果可以直接为决策者事故提高重要决策信息；
3、程序实现中对已有成熟扩散模式的调用方式采用外部可执行程序的直接调用。这种灵活的调用方式,便于该模式功能拓展,极大拓展了程序未来应用领域；
4、基于程序可应用条件所开展的模拟与分析,首次提出敏感地区采样点布设的基本原则,为程序未来实际应用进行有重要的尝试。
To deal with the unknown releasing event of dangerous gases effectively, on the basis of the full research, with good robustness, the genetic optimization (GA) techniques combined with Monte-Calo particle dispersion model, particle reverse trajectory models and other technical means are used to built the source rebuld model-Inverse, which is developed to estimate the source parameters based on environmental monitoring data and meteorological information, depending on the meso/small-scale field experiment data.
According to numerical experiments under different assumptions and the real conditions, the feasibility of source rebuild scheme, the effectiveness of the model result and the running aging of model are analyzed. At the same time, numerical testes are carned out to analyse the effect of difference deployment of sampling points by the simulation results, by which the application conditions of the Inverse analyzed preliminarily. And the basic principle of the sampling points deployment in sensitive areas in this stage. Finally, the systematic numerical scheme is put forward to improve of Inverse model in next stage, which would establish the application foundation in different emergency situation.
Summed up the above research, analysis is:
1. Scheme of rebuilding source parameters in Inverse modelwhich are choosing genetic algorithms as its main optimization means, combined with diffusion/trajectory simulation techniques and appropriate evaluation factor, is reasonable and feasible.
2. Program calls way is flexible in Inverse model, which makes the function of model enlarged easily. The design of key links is helpful to the practical application of model.
3. The results of model are effective, and the built small and medium-scale regional source parameter rebuild model has the estimating ability based on the data of environmental monitoring and the unknown source basic parameters.
本项研究的目的就在于：建立一套技术方案,通过特定的优化算法,使其能够有效利用环境监测数据,实现对环境大气中未知释放过程进行再现模拟。模拟结果可以是用以描述释放的释放参数的估计,也可以是对危险气体整个释放迁移过程及特定时刻危险气体地面浓度分布的再现。
在开展了针对国际上多个代表性的核辐射应急响应系统相关技术发展跟踪的同时,对卡尔曼滤波、贝叶斯推导、非参数化回归以及启发式算法等多种优化算法进行了充分调研,并以此为基础,首次提出利用遗传优化技术的良好鲁棒性,结合蒙特卡洛粒子扩散模式、逆向粒子扩散／轨迹模拟技术,以及皮尔森相关系数等统计分析方法,实现对未知释放参数(释放水平位置、释放高度、释放强度、释放起止时间)的估算以及气载污染物迁移扩散过程的再现。
模式开发与验算的基础来自近年来开展的多例高质量SF6野外大气扩散试验数据。数据包括地形高程数据、气象观测数据、边界层流场数据等。应用Forrtran语言实现了基于环境监测数据与气象数据的源项估算模式-Inverse开发,并应用多线程计算技术,使模式计算更为充分地利用计算资源,有效提高了模式计算效率
进而,本文通过对不同假定与真实条件下的数值模拟试验,对源项估算方案可行性、模式结果有效性、模式运行时效性等进行了分析。同时,通过对不同采样点布设的差异对模拟结果的影响开展了数值模拟试验,对该模式的应用条件进行了尝试,初步提出了敏感地区采样点布设的基本原则。最后,就下阶段Inverse模式的进一步完善提出了较系统的数值验证方案,为该模式在不同突发事件场景条件下的应用奠定基础。
综合上述研究成果,分析认为：
1、Inverse模式选择遗传算法为主要优化手段,结合扩散／轨迹模拟技术,利用PCC(皮尔森相关系数)、NMSE(标准均方差)作为适宜的评判因子搜寻源项参量的技术方案是合理可行的；
2、首次提出在模拟早期,利用逆向轨迹／扩散模式合理限定解域区间的技术路线。这种做法不仅可以有效缩小优化算法的搜索区间,提高模式运算效率,而且其模拟结果可以直接为决策者事故提高重要决策信息；
3、程序实现中对已有成熟扩散模式的调用方式采用外部可执行程序的直接调用。这种灵活的调用方式,便于该模式功能拓展,极大拓展了程序未来应用领域；
4、基于程序可应用条件所开展的模拟与分析,首次提出敏感地区采样点布设的基本原则,为程序未来实际应用进行有重要的尝试。
To deal with the unknown releasing event of dangerous gases effectively, on the basis of the full research, with good robustness, the genetic optimization (GA) techniques combined with Monte-Calo particle dispersion model, particle reverse trajectory models and other technical means are used to built the source rebuld model-Inverse, which is developed to estimate the source parameters based on environmental monitoring data and meteorological information, depending on the meso/small-scale field experiment data.
According to numerical experiments under different assumptions and the real conditions, the feasibility of source rebuild scheme, the effectiveness of the model result and the running aging of model are analyzed. At the same time, numerical testes are carned out to analyse the effect of difference deployment of sampling points by the simulation results, by which the application conditions of the Inverse analyzed preliminarily. And the basic principle of the sampling points deployment in sensitive areas in this stage. Finally, the systematic numerical scheme is put forward to improve of Inverse model in next stage, which would establish the application foundation in different emergency situation.
Summed up the above research, analysis is:
1. Scheme of rebuilding source parameters in Inverse modelwhich are choosing genetic algorithms as its main optimization means, combined with diffusion/trajectory simulation techniques and appropriate evaluation factor, is reasonable and feasible.
2. Program calls way is flexible in Inverse model, which makes the function of model enlarged easily. The design of key links is helpful to the practical application of model.
3. The results of model are effective, and the built small and medium-scale regional source parameter rebuild model has the estimating ability based on the data of environmental monitoring and the unknown source basic parameters.
引文
[1]P.J. Vogt, B.M. Pobanz, F.J. Aluzzi, etc. ARAC Simulation of the Algeciras, Spain Steel Mill Cs-137 Release [C]. Proceedings of the American Nuclear Society 7th Topical Meeting on Emergency Preparedness and Response, Santa Fe, NM, American Nuclear Society, Inc., La Grange Park, IL, September 1999. Also LLNL UCRL-JC-131330.
[2]B. Kosovic, G. Sugiyama, K. Dyer, etc. Dynamic Data-Driven Event Reconstruction for Atmospheric Releases [C].8th Annual George Mason University Transport and Dispersion Modeling Conference, Fairfax, VA, July 2004. Also LLNL UCRL-PRES-205169.
[3]B. Kosovic, G. Sugiyama, S. Chan, etc. Stochastic Source Inversion Methodology and Optimal Sensor Network Design [C].9th Annual George Mason University Conference on Atmospheric Transport and Dispersion Modeling. July 2005. Also LLNL UCRL-PRES-213633.
[4]T. Chow. B. Kosovic. and S. Chan 2005a. Source inversion for contaminant plume dispersion in urban environments using buildingresolving simulations [C].6th Symposium on Urban Environment, Annual American Meteorological Society Meeting, Atlanta, GA, January 2006. Also LLNL UCRL-CONF-216903.
[5]T. Chow, B. Kosovic, and S. Chan,2005b. Source inversion for contaminant plume dispersion in urban environments using buildingresolving simulations [C].9th Annual George Mason University Conference on Atmospheric Transport and Dispersion Modeling, July 2005. Also LLNL UCRL-PRES-213634.
[6]Michael M. Bradley, Branko Kosovic, etc. Models and Measurements: Complementary Tools for Predicting Atmospheric Dispersion and Assessing the Consequences of Nuclear and Radiological Emergencies [C]. November 2005, LLNL UCRL-PROC-216924.
[7]Florian Gering, Klaus Richter, Heinz Miiller. Model Description of the Deposition Monitoring Module DeMM in RODOS PV5.0 [C]. RODOS (RA5)-TN(02) 03, November 2002.
[8]Klaus Richter, Heinz Muller, Florian Gering. Model Description of the Food Monitoring Module FoMM in RODOS PV 6.0 [C]. RODOS (RA5)-TN(03)01, Octomber 2003
[9]Chino M, Nagai H, Furuno A, et al. New technical functions for WSPEEDI: Worldwide version of System for Prediction of Environmental Emergency Dose Information [R]. Proceedings of IRPA-10(M/CD), T-16-2, P-1 1-277,2000.
[10]王跃山.数据同化-它的缘起、含义和主要方法,海洋预报[J],1999,16(1),11-20.
[11]Rojas-Palma, C., etal. Data assimilation in the decision support system RODOS [R]. Radiation Protestion Dosimetry,2003,104,31-40.
[12]D.Q. Zheng, J.K.C. Leung, B.Y. Lee. Online update of model state and parameters of a Monte Carlo atmospheric dispersion model by using ensemble Kalman filter [J]. Atmospheric Environment,2009,43,2005-2011.
[13]Seibert, P.. Inverse modelling of atmospheric trace substances on the regional scale with Lagrangian models[R]. In:Proceedings of the EUROTRAC-2 Symposium, 11-15 March 2002.
[14]Flescha, T., Wilsona, J.D. etc.. Estimating gas emissions from a farm with an inverse dispersion technique [J]. Atmospheric Environment 39 (27),4863-4874,2005.
[15]Thomas K. Flesch, John D. Wilson, Lowry A. Harper, etc.. Seasonal NH3 emissions for the continental united states:Inverse model estimation and evaluation [J]. Atmospheric Environment 40,4986-4998,2006.
[16]邢文训,谢金星.《现代优化计算方法》[M].北京：清华大学出版社,2005.
[17]陈军明,徐大海等.遗传算法在点源扩散浓度反演排放源强中的应用[R].《国家重点基础研究规划项目》：首都北京及周边地区大气、水、土环境污染机理 与调控原林,2006.
[18]Haupt S.E., Young G.S., Allen C.T.. Validation of a receptor/dispersion model coupled with a genetic algorithm using synthetic data [J]. Journal of Applied Meteorology, submitted,2005.
[19]Haupt S.E., Young G.S., Allen C.T.. Validation of a receptor/dispersion model with a genetic algorithm using synthetic data [J]. Journal of Applied Meteorology 45, 476-490,2006.
[20]Allen C.T., Haupt S.E., Young G.S.. Source characterization with a genetic algorithm-coupled dispersion-backward model incorporating SCIPUFF [J]. Journal of Applied Meteorology 41,465-479,2007.
[21]Kerrie J. Long, Sue Ellen Haupt, George S. Young. Assessing sensitivity of source term estimation [J]. Atmospheric Environment 44.1558-1567,2010.
[22]MacDonald I.R., Reilly Jr., J.R. etc. A remote sensing inventory of active oil seeps and chemosynthetic communities in the northern Gulf of Mexico [R]. In: Schumacher, D., Abrams, M.A. (Eds.). Hydrocarbon Migration and its near-surface expression, AAPG Mem., vol.66. pp.27-37,1996.
[23]Laura C. Thomsona, Bill Hirstb, Graham Gibsona etc. An improved algorithm for locating a gas source using inverse methods [J]. Atmospheric Environment,41, 1128-1134,2007.
[24]James W. Psychology(Briefer Course) [M]. New York:Holt,1890.
[25]McCulloch W, Pitts W. A logical calculus of the ideas imminent in nervous activity [J]. Bullentin of Mathematical Biophysics,5:115-133,1943.
[26]Hopfield J, Tank D.'Neural' computation of decision in optimization problem [J]. Biological Cybernetics,52:141-152.1985.
[27]Hopfield J, Tank D. Computing with neural circuits:a model [J]. Science, 233:625-633,1986.
[28]H. Pfeiffer, G. Baumbach, I,. Sarachaga-Ruiz etc. Neural modelling of the spatial distribution of air pollutants [J], Atmospheric Environment 43 3289-3297,2009.
[29]Inane Senocak, Nicolas W. Hengartner, Margaret B. Short etc. Stochastic event reconstruction of atmospheric contaminant dispersion using Bayesian inference [J]. Atmospheric Environment 42,7718-7727,2008.
[30]Andrew Keats, Eugene Yee, Fue-Sang. Lien Bayesian inference for source determination with applications to a complex urban environment [J]. Atmospheric Environment 41 465-479,2007.
[31]Ronald C. Henrya, Yu-Shuo Changa, Clifford H. Spiegelman. Locating nearby sources of air pollution by nonparametric regression of atmospheric concentrations on wind direction [J]. Atmospheric Environment 36,2237-2244,2002.
[32]J. A. Nelder and R. Mead, A Simplex Method for Function Minimization [J]. Computer Journal 7 308-313,1965.
[33]Allen, C.T., Young, G.S., Haupt, S.E. Improving pollutant source characterization by better estimating wind direction with a genetic algorithm [J]. Atmospheric Environment 41,2283-2289,2007.
[34]S. Mosca, G. Graziani, W. Klug etc. A statistical methodology for the evaluation of long-range dispersion models:an application to the ETEX exercise[J]. Atmospheric Evironment 32,4307-4325,1998.
[35]徐向军,姚仁太.大尺度迁移模式及其验证技术研究[C]//俞学曾.第十届全国大气环境学术论文集.济南：中国环境科学学会大气环境.
[36]D.L. Carroll. D.L. Carroll's FORTRAN Genetic Algorithm Driver [EB/OL]. http://cuaerospace.com/carroll/ga.html,2009.
[37]David Goldberg. Genetic Algorithms in Search, Optimization and Machine Learning [J]. Addison-Wesley,1989.
[38]Goldberg, D. E., and Richardson, J., Genetic Algorithms with Sharing for Multimodal Function Optimization [C]. Genetic Algorithms and their Applications: Proceedings of the Second International Conference on Genetic Algorithms,41-49.1987.
[39]Goldberg, D. E., "A Note on Boltzmann Tournament Selection for Genetic Algorithms and Population-Oriented Simulated Annealing," in:Complex Systems, Vol.4, Complex Systems Publications, Inc.,1990, pp.445-460.
[40]Syswerda, G., "Uniform Crossover in Genetic Algorithms," in: Proceedings of the Third International Conference on Genetic Algorithms, Schaffer, J. (Ed.), Morgan Kaufmann Publishers, Los Altos, CA, pp.2-9,1989.
[2]B. Kosovic, G. Sugiyama, K. Dyer, etc. Dynamic Data-Driven Event Reconstruction for Atmospheric Releases [C].8th Annual George Mason University Transport and Dispersion Modeling Conference, Fairfax, VA, July 2004. Also LLNL UCRL-PRES-205169.
[3]B. Kosovic, G. Sugiyama, S. Chan, etc. Stochastic Source Inversion Methodology and Optimal Sensor Network Design [C].9th Annual George Mason University Conference on Atmospheric Transport and Dispersion Modeling. July 2005. Also LLNL UCRL-PRES-213633.
[4]T. Chow. B. Kosovic. and S. Chan 2005a. Source inversion for contaminant plume dispersion in urban environments using buildingresolving simulations [C].6th Symposium on Urban Environment, Annual American Meteorological Society Meeting, Atlanta, GA, January 2006. Also LLNL UCRL-CONF-216903.
[5]T. Chow, B. Kosovic, and S. Chan,2005b. Source inversion for contaminant plume dispersion in urban environments using buildingresolving simulations [C].9th Annual George Mason University Conference on Atmospheric Transport and Dispersion Modeling, July 2005. Also LLNL UCRL-PRES-213634.
[6]Michael M. Bradley, Branko Kosovic, etc. Models and Measurements: Complementary Tools for Predicting Atmospheric Dispersion and Assessing the Consequences of Nuclear and Radiological Emergencies [C]. November 2005, LLNL UCRL-PROC-216924.
[7]Florian Gering, Klaus Richter, Heinz Miiller. Model Description of the Deposition Monitoring Module DeMM in RODOS PV5.0 [C]. RODOS (RA5)-TN(02) 03, November 2002.
[8]Klaus Richter, Heinz Muller, Florian Gering. Model Description of the Food Monitoring Module FoMM in RODOS PV 6.0 [C]. RODOS (RA5)-TN(03)01, Octomber 2003
[9]Chino M, Nagai H, Furuno A, et al. New technical functions for WSPEEDI: Worldwide version of System for Prediction of Environmental Emergency Dose Information [R]. Proceedings of IRPA-10(M/CD), T-16-2, P-1 1-277,2000.
[10]王跃山.数据同化-它的缘起、含义和主要方法,海洋预报[J],1999,16(1),11-20.
[11]Rojas-Palma, C., etal. Data assimilation in the decision support system RODOS [R]. Radiation Protestion Dosimetry,2003,104,31-40.
[12]D.Q. Zheng, J.K.C. Leung, B.Y. Lee. Online update of model state and parameters of a Monte Carlo atmospheric dispersion model by using ensemble Kalman filter [J]. Atmospheric Environment,2009,43,2005-2011.
[13]Seibert, P.. Inverse modelling of atmospheric trace substances on the regional scale with Lagrangian models[R]. In:Proceedings of the EUROTRAC-2 Symposium, 11-15 March 2002.
[14]Flescha, T., Wilsona, J.D. etc.. Estimating gas emissions from a farm with an inverse dispersion technique [J]. Atmospheric Environment 39 (27),4863-4874,2005.
[15]Thomas K. Flesch, John D. Wilson, Lowry A. Harper, etc.. Seasonal NH3 emissions for the continental united states:Inverse model estimation and evaluation [J]. Atmospheric Environment 40,4986-4998,2006.
[16]邢文训,谢金星.《现代优化计算方法》[M].北京：清华大学出版社,2005.
[17]陈军明,徐大海等.遗传算法在点源扩散浓度反演排放源强中的应用[R].《国家重点基础研究规划项目》：首都北京及周边地区大气、水、土环境污染机理 与调控原林,2006.
[18]Haupt S.E., Young G.S., Allen C.T.. Validation of a receptor/dispersion model coupled with a genetic algorithm using synthetic data [J]. Journal of Applied Meteorology, submitted,2005.
[19]Haupt S.E., Young G.S., Allen C.T.. Validation of a receptor/dispersion model with a genetic algorithm using synthetic data [J]. Journal of Applied Meteorology 45, 476-490,2006.
[20]Allen C.T., Haupt S.E., Young G.S.. Source characterization with a genetic algorithm-coupled dispersion-backward model incorporating SCIPUFF [J]. Journal of Applied Meteorology 41,465-479,2007.
[21]Kerrie J. Long, Sue Ellen Haupt, George S. Young. Assessing sensitivity of source term estimation [J]. Atmospheric Environment 44.1558-1567,2010.
[22]MacDonald I.R., Reilly Jr., J.R. etc. A remote sensing inventory of active oil seeps and chemosynthetic communities in the northern Gulf of Mexico [R]. In: Schumacher, D., Abrams, M.A. (Eds.). Hydrocarbon Migration and its near-surface expression, AAPG Mem., vol.66. pp.27-37,1996.
[23]Laura C. Thomsona, Bill Hirstb, Graham Gibsona etc. An improved algorithm for locating a gas source using inverse methods [J]. Atmospheric Environment,41, 1128-1134,2007.
[24]James W. Psychology(Briefer Course) [M]. New York:Holt,1890.
[25]McCulloch W, Pitts W. A logical calculus of the ideas imminent in nervous activity [J]. Bullentin of Mathematical Biophysics,5:115-133,1943.
[26]Hopfield J, Tank D.'Neural' computation of decision in optimization problem [J]. Biological Cybernetics,52:141-152.1985.
[27]Hopfield J, Tank D. Computing with neural circuits:a model [J]. Science, 233:625-633,1986.
[28]H. Pfeiffer, G. Baumbach, I,. Sarachaga-Ruiz etc. Neural modelling of the spatial distribution of air pollutants [J], Atmospheric Environment 43 3289-3297,2009.
[29]Inane Senocak, Nicolas W. Hengartner, Margaret B. Short etc. Stochastic event reconstruction of atmospheric contaminant dispersion using Bayesian inference [J]. Atmospheric Environment 42,7718-7727,2008.
[30]Andrew Keats, Eugene Yee, Fue-Sang. Lien Bayesian inference for source determination with applications to a complex urban environment [J]. Atmospheric Environment 41 465-479,2007.
[31]Ronald C. Henrya, Yu-Shuo Changa, Clifford H. Spiegelman. Locating nearby sources of air pollution by nonparametric regression of atmospheric concentrations on wind direction [J]. Atmospheric Environment 36,2237-2244,2002.
[32]J. A. Nelder and R. Mead, A Simplex Method for Function Minimization [J]. Computer Journal 7 308-313,1965.
[33]Allen, C.T., Young, G.S., Haupt, S.E. Improving pollutant source characterization by better estimating wind direction with a genetic algorithm [J]. Atmospheric Environment 41,2283-2289,2007.
[34]S. Mosca, G. Graziani, W. Klug etc. A statistical methodology for the evaluation of long-range dispersion models:an application to the ETEX exercise[J]. Atmospheric Evironment 32,4307-4325,1998.
[35]徐向军,姚仁太.大尺度迁移模式及其验证技术研究[C]//俞学曾.第十届全国大气环境学术论文集.济南：中国环境科学学会大气环境.
[36]D.L. Carroll. D.L. Carroll's FORTRAN Genetic Algorithm Driver [EB/OL]. http://cuaerospace.com/carroll/ga.html,2009.
[37]David Goldberg. Genetic Algorithms in Search, Optimization and Machine Learning [J]. Addison-Wesley,1989.
[38]Goldberg, D. E., and Richardson, J., Genetic Algorithms with Sharing for Multimodal Function Optimization [C]. Genetic Algorithms and their Applications: Proceedings of the Second International Conference on Genetic Algorithms,41-49.1987.
[39]Goldberg, D. E., "A Note on Boltzmann Tournament Selection for Genetic Algorithms and Population-Oriented Simulated Annealing," in:Complex Systems, Vol.4, Complex Systems Publications, Inc.,1990, pp.445-460.
[40]Syswerda, G., "Uniform Crossover in Genetic Algorithms," in: Proceedings of the Third International Conference on Genetic Algorithms, Schaffer, J. (Ed.), Morgan Kaufmann Publishers, Los Altos, CA, pp.2-9,1989.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |