摘要
我国煤炭资源区域分布与主要消费地不均衡的矛盾十分突出。湖南电力企业每年消耗大量煤炭,但是本省缺少煤炭资源,尤其是优质煤炭资源。动力配煤具有提高煤炭燃烧效率、减少污染物排放、充分利用劣质煤种等优点,是解决这一矛盾的最有效的方法之一。
近年来,如何利用智能技术解决动力配煤存在的问题,是国内研究的热点之一。本文将Elman神经网络应用于动力配煤的煤质预测研究,利用模拟退火法优化配煤方案,具体研究内容如下。
根据单煤和混煤的工业分析、元素分析结果,利用t分布方法,验证了单煤煤质与混煤煤质之间的非线性关系。
建立了Elman煤质特性预测模型,利用该网络预测了混煤水分、灰分、挥发分、发热量和着火温度;针对不同元素建立了不同Elman神经网络预测模型,并预测了混煤的各元素含量,预测结果误差较小。
利用相关性分析法检验了预测结果与实测结果间的相关性,相关系数在0.95以上,表明预测值与实测值之间具有良好相关性;利用误差置信区间分析法检验了预测结果,结果表明Elman神经网络预测结果具有良好可靠性和可信度。
以锅炉对燃料的要求为约束条件,以混煤价格为目标函数,建立了动力配煤优化模型;引入模拟退火法对配煤优化模型进行求解,其中模型中煤质特性采用Elman神经网络预测值;将该算法所求结果与利用穷举法所求的最优结果进行比较,结果表明模拟退火算法求解结果与最优结果相差较小,算法效率比穷举法高8倍左右。
The contradiction of imbalance between regional distribution of coal resources and location of major consumer was very serious in China. The power plants of Hunan consume a great amount of coals, but Hunan is lack of coal, especially high-quality coal. Coal blending was the best way to overcome the contradiction, because they can raise combustion efficiency of coal, reduce pollutant emission and make full use of poor quality coal.
In resent years, it has become one of hot points to solve the problem of blended coals through artificial intelligence technology. In this paper, Elman neural network was used to predict characteristic parameter of blended coals. The Simulated Annealing method was adopted to optimize the components of blended coals. This main works of this dissertation are as following.
Based on element analysis and proximate analysis of blended coals and single coal, the non-linear relationship between qualities of the blended coals and its components was proved by t distribution method.
The Elman neural network prediction model was established. The moisture content, ash content, volatile content, ignition temperature and calorific value of the blended coals were predicted by the characteristic prediction model. At the same time, different elements of blended coals were also predicted by different Elman neural network element prediction models.
Relationship analysis method was used to testify the relationship between predictive values and actual values. It showed they had good dependencies, the correlation coefficients of all the parameters were more than 0.95. High reliability and confidence of the predicted result were verified using confidence interval method.
Power coal blending model was established, whose constraints were performance of the blended coals allowed by boiler, and the objective function was the price of the blended coals. Simulated annealing (SA) method was used to solve the power coal blending model to search the lowest cost of coals. Elman neural network was used to predict qualities of blended coals. The result solved by SA was compared with the optimal result solved by exhaustive method. It showed that they had a small error, but algorithm efficiency of SA is eight times higher than exhaustive method.
引文
[1]刘马群,黄跃年.浅谈煤炭资源保护与综合开发[J].资源节约与综合利用,1996,(1),38-40
[2]崔民选.中国能源发展报告[M].北京:社会科学文献出版社,2008
[3]周乃君.能源与环境[M].长沙:中南大学出版社,2008:1-74
[4]陈美英.湖南省煤炭资源构造特点及找煤潜力分析[J].资源与产业,2009,11(6):64-68
[5]湖南省国土资源厅调研组.湖南省煤炭资源对火电发展的保障程度-问题与对策[J].国土资源导刊,2004,4(4):9-12
[6]韩宏彬.火力发电厂劣质煤使用现状及配煤安全对策[J].实用节能技术,2009,(5):57-59
[7]陈文敏等.洁净煤技术基础[M].北京:煤炭工业出版社,1997:10-35
[8]戴财胜.动力配煤的理论与应用研究[D].北京:中国矿业大学(北京校区),2000:1-25
[9]张军,袁建伟,马毓义.煤的显微组分及其对粉煤焦形态的影响[J].燃烧科学与技术,1995,3(1):196~202
[10]姚强,岑可法,施正伦等.多煤种配煤特性的试验研究[J].动力工程,1997,17(2):16~20
[11]肖键.混煤燃烧过程中颗粒物排放特性的试验研究[D].浙江:浙江大学,2006
[12]阮伟,张伟宁,周俊虎等.电厂优化配煤多目标机会约束数学模型的建立[J].动力工程,2001,21(1):1090~1092
[13]曹欣玉,阮伟,周俊虎等.电厂优化配煤专家系统的研究与开发[J].电站系统工程,2000,16(1):4~6
[14]刘志.配煤煤灰内矿物质转变过程与熔融特性规律研究[D].浙江:浙江大学,2006
[15]樊增权.300MW锅炉掺烧长焰煤的运行特性及配煤研究[D]河北:华北电力大学,2004
[16]崔震华.基于神经网络的煤灰结渣特性的研究[D].保定:华北电力大学,2007
[17]邢诚.动力配煤原理在火力发电厂中的应用[D].北京:华北电力大学,2006
[18]方立军.混煤燃烧及污染物排放特性的实验研究和数值模拟[D].北京:华北电力大学,2001
[19]李永华.混煤高效低污染燃烧特性研究[D].北京:华北电力大学,2000
[20]杨劲.基于智能技术的混煤特性预测及配煤优化模型研究[D].长沙:中南大学,2009
[21]Zaixing Zhu, Jiemin Zhou, Ying Wang. Numerical simulation on fluid flow and combustion in a subcritical pressure boiler with various hybrid coal [J].2008 ASME International Mechanical Engineering Congress and Exposition.
[22]A.megaritis,R.C.Messenbock,et al.High-pressure pyrolysis and CO2-gasification of coal maceral concentrates:conversions and char combustion reactivities.Fuel,1999,78(8),871~882
[23]Barbara pilawa,Andrzej B.Wieckowski and Marek Lewandowski.E.P.r.studies of thermal decomposition of exinite and inertinite[J].Fuel,1996,75(10):1181-1185
[24]阴崇儒.欧洲煤炭能源情势[J].能源基地建设,1991(2):61-64
[25]日本燃煤技术的试验研究报告[J].电站系统工程,1988(4):35-64
[26]杨天正.国际能源机构诸国清洁煤技术开发状况[J].煤化工,:1.990(4):52-55
[27]邱建荣.混煤特性的综合性研究[D].武汉:华中理工大学,1993
[28]符曦.系统最优化及控制[M].北京:机械工业出版社,2004:1-35
[29]AsicMD,Kovacevic-VujcicVV.TABOO Search Methodology in Global Optimization[J]. Computers and Mathematics with application,1999,3(3):125-133
[30]Sung2 Hwan Min. Hybrid genetic algorithms and support vector machines for bankruptcy prediction[J]. Expert Systems with Applications,2006,3(31):652-660
[31]焦李成,杜海峰,刘芳有等.免疫优化计算、学习与识别[M]北京:科学出版社,2006:27-231
[32]蔡自兴,龚涛.免疫算法研究的进展[J].控制与决策,2004,19(8):841-845
[33]Marco Dorigo, Vittorio Maniezzo, Alberto Colomi. Ant System Optimization by a Colony of Cooperating Agents[J].IEEE Transactions on systems, man, and cybernetics,1996,1(26):29-41
[34]韩昭沧,燃料与燃烧[M]北京:冶金工业出版,2005:4-10
[35]陈晓玲.配煤线性规划模型的研究[J].煤炭加工与综合利用,2003,(2):34-38
[36]陈文敏,姜宁.动力配煤主要煤质的可加性[J],煤质技术与科学管理,1997,(4):30-32
[37]赵建国,赵虹,岑可法.基于遗传算法优化的煤粉着火温度BP神经网络预测模型[J].煤炭学报,2006,31(2):211-214
[38]李颖.基于非线性理论的动力配煤模型的研究[D].浙江:浙江大学,2002
[39]沈彬彬.基于遗传算法的动力配煤优化模型的研究及其软件实现[D].浙江:浙江大学,2004
[40]陈彦杰.基于神经网络与遗传算法的配煤优化模型的研究[D].黑龙江:黑龙江科技学院,2007
[41]陈怀珍,陈亚飞等.中国动力配煤工业分析和发热量的可加性研究[J].洁净煤技术,1999,5(2):51-54
[42]桂祥友,马云东.非线性优化理论在动力配煤中的应用[J].中国矿业,2005,14(6):49-51
[43]胡荣祖,史启祯.热分析动力学[M].北京:科学出版社,.2001
[44]梁向晖,钟伟强.热重差热联用热分析仪SDTQ600的特点及维护[J].现代仪 器,2010.22(6):75-77
[45]喻秋梅,庞亚军,陈宏国.煤燃烧试验中着火点确定方法的探讨[J].华北电力技术,2001,2(7):9-11
[46]Campos M, Lopez D. Neural network approach to locate motifs in biosequences[J]. Progress in Pattern Recognition, Image Analysis and Applications,2005, Proceedings 3773:214-221
[47]Coban R, Can B. Identification and control of ITU Triga Mark-Ⅱ Nuclear Research Reactor using neural networks and fuzzy logic[J]. Lecture Notes in Computer Science, 2005,1(3809):1057-1062
[48]Kim K.J. Artificial neural networks with evolutionary instance selection for financial forecasting[J]. Expert Systems with Applications,2006,30(3):519-526
[49]Enke D, Thawornwong S. The use of data mining and neural networks for forecasting stock market returns[J]. Expert Systems with Applications,2005,29(4):927-940
[50]Elman J.L. Finding Structure in Time [J]. Cognitive Science,1990,2(14):179-211
[51]Jordan M.I. Attractor dynamics and parallelism in a connectionist sequential machine[J]. In Proceedings of the Eighth Annual Conference of the Cognitive Science Society, 1986,1(144):531-546
[52]胡军,郑宝山,吴代赦.经验公式对全国煤实测发热量的检验[J].煤炭转化,2006,29(1):1-5
[53]刘青.浅谈煤炭发热量计算经验公式的实用性[J].山东煤炭科技,2005,1(4):71—73
[54]李徐萍,李国庆,李芳莲.煤的发热量经验公式及其在煤质分析中的应用[J].轻金属,2000(3):28-29
[55]刘光中,李晓峰.人工神经网络BP算法的改进和结构的自调整[J].运筹学学报,2001,5(1):81-88
[56]陈小前,罗世彬,王振国,陶玉静.BP神经网络应用中的前后处理过程研究[J].系统工程理论与实践,2002,(1):65-70
[57]冯玉蓉.模拟退火算法的研究及其应用[D].昆明:昆明理工大学,2005
[58]庞峰.模拟退火算法的原理及算法在优化问题上的应用[D].长春:吉林大学,2006
[59]陈华根,吴健生,王家林,陈冰.模拟退火算法机理研究[J].同济大学学报(自然科学版),2004,6(32):802-805
[60]明延峰,张永祥,朴甲哲.故障识别中并行组合模拟退火算法的应用[J].中国机械工程,2003,14(17):1496-1498
[61]康立山等.非数值并行算法(第一册)模拟退火算法[M].北京:科学出版社,2006
[62]王得燕.穷举法与粒子群算法的比较[J].无锡职业技术学院学报,2008,7(1):4445