转炉生产高品质齿轮钢冶炼工艺与产品质量研究
|
摘要
跟电炉流程相比,转炉流程具有高效、批量大、原材料杂质少、成本低等优点,但大多因为转炉流程节奏快,缺乏足够的精炼时间,钢水的纯净度满足不了要求,缺乏专业的轧钢及后部处理设施与之配套等诸多原因,钢材的实物质量还存在许多问题。因此,转炉流程生产特殊钢的冶炼工艺研究具有重要的意义。
本文针对首钢转炉流程的冶炼特点,采用了历史生产数据的统计与分析、钢材的检验与分析、钢材疲劳性能台架测试、理论分析、现场工业试验相结合的方法,对转炉流程生产特殊钢的低氧冶炼工艺、精炼渣系选择、夹杂物钙处理及淬透性带宽的控制进行了研究,取得了很好的效果。主要成果包括以下几部分：
(1)首钢产品质量现状的分析
本文在首钢现有生产线上,通过对历史生产数据的统计与分析,查阅大量的相关文献资料,解剖国内外先进流程生产的钢材实物质量找出首钢的差距；结果表明,转炉下渣量大,达12-15kg/t,炉渣改质效果差,精炼进站w(FeO)达3.0%,精炼出站w(FeO)达2.0%；w[T.O]达27×10-6,精炼时间比转炉和连铸多出近30min,成为流程的制约环节；钢水中A和B类夹杂物仍然很多,钙处理不充分,钙铝酸盐夹杂大部分落在高熔点的CA6-CA2区域内；窄成分控制精度低,成分波动比较大,淬透性窄带合格率低,J9值带宽36.0±3.00HRC的合格率为81.3%,36.0±2.00HRC的合格率为65.3%；需要对冶炼工艺和成分控制进行改进。
(2)转炉流程生产齿轮钢的冶炼工艺的研究
以热力学计算为依据,结合历史数据的分析,对铝脱氧钢的铝氧平衡关系和夹杂物的生成及去除条件进行了研究,结果表明,控制好转炉下渣量和炉渣改质是缩短精炼周期、提高齿轮钢纯净度、打通转炉流程的关键,控制w[Al]s>0.03%,控制精炼渣系的αAl2O3<0.02,w(CaO)/w(SiO2)=3.5-5, w(CaO)/w(Al2O3)=1.5-2.0,可以获得w[T.O]小于15×10-6低氧精炼钢水；要求20CrMnTiH1精炼钢水在1600℃以上的温度下进行钙处理,当酸溶铝质量分数为0.03%,控制钢水中的w[Ca]/w[Al]s比在0.017～0.19,使A1203变性为液态所需钙的质量分数范围为(5～57)×10-6,能生成低熔点的Ca·Al203和12CaO7Al203,只要w[S]高于0.01%,则钢水中A1203夹杂很难完全变性为C12A7,而相对容易变性为CA。
(3)铝脱氧钢精炼过程稳定氧化物夹杂的研究
采用小样电解方法分析铝脱氧钢精炼过程反应产物中稳定氧化物的种类和分布及变化规律,结果表明,首钢齿轮钢中氧化物夹杂总量质量分数平均为0.0027%,其中A1203质量分数高达0.00219%,占全部氧化物夹杂的69.30%。而MnO、FeO、CaO等钢中常见的氧化物夹杂质量分数极低,不超过0.0002%；稳定氧化物中的氧占到w[T.O]的90%,而A1203和Si02氧化物中氧占到w[T.O]的～80%,稳定氧化物夹杂去除是生产低氧齿轮钢的关键；首钢齿轮钢中存在大量的M3C和M(C,N)析出物,它们主要是在凝固过程中析出的,对热处理工序中细化晶粒有较大的好处。
(4)非金属夹杂物对齿轮钢疲劳寿命的影响研究
采用旋转弯曲疲劳实验方法,对具有不同冶金质量的齿轮钢进行了疲劳破坏行为研究,测定了齿轮钢的S-N曲线,对断口断裂机制进行了分析和讨论。结果表明,齿轮钢热轧态、锻态及正火态的旋弯疲劳实验数据的分散性均较小,体现出材料具有较好的冶金质量特别是微观组织的均匀性；不同氧质量分数20CrMnTi钢正火态之间的旋弯疲劳性能差异不明显。这主要与二者的疲劳裂纹均起源于表面基体的疲劳断裂机制有关；对于起裂于表层夹杂物的试样,夹杂物十分粗大,为Al2O3·(CaO)x类复合氧化物夹杂。因此,应特别注意控制钢中此类异常夹杂物。
(5)齿轮钢窄淬透性带宽控制的研究
采用数学统计和回归的方法,找出影响齿轮钢淬透性带宽的因素,基于多支持向量机的方法建立齿轮钢淬透性值预报模型并进行在线预报,指导成分调整。取得了很好的成果：窄成分控制精度为w[C]±0.01%、w[Si]±0.02%、w[Ti]±0.02%、w[Mn]±0.02%、w[Cr]±0.03%、w[Ti]±0.01%时,所占比例分别为96.1%、94.7%、90.5%、87.4%、81.9%、90.5%,较优化前有了很大的提高；淬透性窄带J9值±2.0合格率达到93.3%,J9值±3.0合格率达到99.3%；J15值±2.0合格率达到89.2%,J9值±3.0合格率达到98.4%,较优化前有了很大地提高。
通过工艺优化,各项主要技术指标较优化前有很大地提高,w[T.O]≤15×10-6,夹杂物评级A+B+C+D<5.0级,晶粒度≥8.5级；淬透性带宽≤6HRC合格率达到99.3%,≤4HRC合格率达到93.3%；一般疏松小于0.5级,中心疏松小于1.0级,未见裂纹、夹杂、气孔等低倍缺陷；较优化前,精炼时间缩短了15.6min。
综上所述,本文对转炉流程冶炼工艺进行了细致地研究,制定了转炉生产齿轮钢的工艺制度,实现了转炉流程生产高品质齿轮钢的目的,实物质量达到国际先进水平。
In comparision with EAF process, BOF special steel making has many characteristics such as high efficientcy, large batch production, low residual impurity of material and low cost, etc. However, some quality problems exist in the BOF special steel product due to the rapid process rhythm, lacking of enough refining time, less purity, shortage of professional steel rolling and subsequent equipments. Therefore, research on smelting process to improve BOF special steel quality has a great significance.
This paper focus on BOF steelmaking process to produce special steel in Shougang steelworks. It investigated the low oxygen smelting technology of gear steel with BOF process; studied the refining slag, calcium treatment of inclusion and control of wideness harden ability bands by analyzing the statistics of history production data, testing of steel product and theoretical analysis.
The main research achievements are as follow:
(1) Shougang products and process analysis
According to the results of industry test and historical statistics data, combining with literature survey, the shortages of shougang product compared with domestic and overseas advanced manufacturer were found. The results show that the BOF slag flowover was up to12-15kg/t; w(FeO) that entered LF was up to3.0%and at the end of LF refining was2.0%; w[T.O]was up to27×10-6of mass concentration; LF refining time was close to30minutes which longer than BOF or CC.
There were excessive A and B type inclusions in steel. The inclusions modification was insufficient; most calcium aluminate inclusions located in CA6-CA2area which showed the calcium treatment was deficient. The accuracy on precise composition control was low, qualification rate of hardenability narrow band control was low, it was81.3%if J9defined as36.0±3.00HRC, however, decreased to65.3%if J9defined as36.0±2.00HRC. It means that improve the smelting process and composition control is necessary.
(2) Research on smelt Technology of gear steel with BOF process
According to thermodynamics calculating, the relationship between aluminium and oxygen content in the aluminum deoxidization steel, and the condition of inclusion generation and the condition of inclusion removing were studied by plant historical statistics data analysis. The results showed that the key points were to shorten refining time, to control slag flowover and to enhance slag modification. In order to produce steel with low oxygen w[T.O]<15x10-6, follow conditions must be satisfied: w[Al]s>0.03%, αAl2O3<0.02, w[CaO]/w[SiO2]=3.5-5and w[CaO]/w[Al2O3]=1.5-2.0. Calcium treatment should be made when the temperature of refining steel is above1600℃. The ratio of calcium and aluminum in steel should be controlled in0.017～0.19when acid-soluble aluminum is0.03%. The CaO·Al203and12CaO·7Al2O3with low melting point can be precipitated because Al2O3is liquid state when the calcium is (5-57)×10-6. When the sulphur content is higher than0.01%, the Al2O3inclusion will turn into CA easily, not C12A7.
(3) Research on oxide inclusion in aluminum deoxidization steel during refining process
Electrolysis method was used to analysis type, distribution and variability regulation of oxide inclusion in refining process of aluminum deoxidization steel, The result indicated, for Shougang steel, total quantity of oxide inclusion in the gear steel was0.0027%equally, A12O3reaches as high as0.00219%, counted69.30%of the oxide inclusion, the content of the common oxide inclusion in steel such as MnO、FeO、CaO are extremely low, less than0.0002%. Oxygen content from oxide inclusion almost took90%in ths steel,80%are from Al2O3and SiO2, so that it is the key to remove stable oxide. Shougang gear steel has massive M3C and M(C,N) precipitate, mainly precipitated during the solidification process, have great advantage to the grain refinement in heat treatment.
(4) Research on the influence of non-metallic inclusions to gear steel fatigue life
Through the rotating bending fatigue experimental, fatigue breakdown behaviours of gear steel in different metallurgy quality from industrial producing were researched, and the gear steel S-N curve was determined, also fracture break mechanism analyzed and discussed. The result indicated that, to the20CrMnTi steel in hot rolling, hammers and the normalizing condition, rotating bar bending fatigue test showed that the disparity of dates was low, and the metallurgy quality was high especially the uniformity of micro-structure, there was no obvious difference on the rotating bar bending fatigue quality between20CrMnTi steel in the normalizing condition of different oxygen content. This mainly was related to that their fatigue cracking stems from the fatigue cracking mechanism from the superficial substrate. To the sample cracked from the inclusion of the surface layer, the inclusion was Al2O3·(CaO)x and large. Therefore, special attention should be paid to the unusual inclusions.
(5) Research on the narrow hardenability band control technologies of gear steel
Impact of gear steel hardenability bandwidth factor was identified using mathematical statistics and regression analysis.
The value of gear steel hardenability prediction models was established to predict the hardenablility values on-line. The composition adjustment was guided based on Multi-SVM prediction model. Narrow component control precision is w[C]±0.01%、 w[Si]±0.02%、w[Ti]±0.02%、w[Mn]±0.02%、w[Cr]±0.03%、w[Ti]±0.01%, the respective proportions is96.1%、94.7%、90.5%、87.4%、81.9%、90.5%. The data were improved greatly than before; J9±2.0HRC qualification rate of narrow hardenability bandwidth value was93.3%, qualification rate of J9value±3.0was99.3%; Qualification rate of J15value±2.0was89.2%, qualification rate of J9value±3.0was98.4%.
Through optimization process, the main technical data were greatly improved. The results were as follows:w[T.O]≤15×106, inclusion rating A+B+C+D<5.0seris, grain size rating>8.5, percent of pass of hardenability bandwidth<6HRC was99.3%and <4HRC was93.3%, macroscopic test of general loosen was less than0.5grade, center loosen was less than1.0grade, no cracks, inclusions, porosity and other defects in the low-fold and the refining time was shorten15.6min than before.
In conclusion, this paper gave a detailed study on the converter smelting process to produce gear steel, and realized the object to produce high-quality gear steel by converter process. The statistics of the product qualities were up to international advanced level.
本文针对首钢转炉流程的冶炼特点,采用了历史生产数据的统计与分析、钢材的检验与分析、钢材疲劳性能台架测试、理论分析、现场工业试验相结合的方法,对转炉流程生产特殊钢的低氧冶炼工艺、精炼渣系选择、夹杂物钙处理及淬透性带宽的控制进行了研究,取得了很好的效果。主要成果包括以下几部分：
(1)首钢产品质量现状的分析
本文在首钢现有生产线上,通过对历史生产数据的统计与分析,查阅大量的相关文献资料,解剖国内外先进流程生产的钢材实物质量找出首钢的差距；结果表明,转炉下渣量大,达12-15kg/t,炉渣改质效果差,精炼进站w(FeO)达3.0%,精炼出站w(FeO)达2.0%；w[T.O]达27×10-6,精炼时间比转炉和连铸多出近30min,成为流程的制约环节；钢水中A和B类夹杂物仍然很多,钙处理不充分,钙铝酸盐夹杂大部分落在高熔点的CA6-CA2区域内；窄成分控制精度低,成分波动比较大,淬透性窄带合格率低,J9值带宽36.0±3.00HRC的合格率为81.3%,36.0±2.00HRC的合格率为65.3%；需要对冶炼工艺和成分控制进行改进。
(2)转炉流程生产齿轮钢的冶炼工艺的研究
以热力学计算为依据,结合历史数据的分析,对铝脱氧钢的铝氧平衡关系和夹杂物的生成及去除条件进行了研究,结果表明,控制好转炉下渣量和炉渣改质是缩短精炼周期、提高齿轮钢纯净度、打通转炉流程的关键,控制w[Al]s>0.03%,控制精炼渣系的αAl2O3<0.02,w(CaO)/w(SiO2)=3.5-5, w(CaO)/w(Al2O3)=1.5-2.0,可以获得w[T.O]小于15×10-6低氧精炼钢水；要求20CrMnTiH1精炼钢水在1600℃以上的温度下进行钙处理,当酸溶铝质量分数为0.03%,控制钢水中的w[Ca]/w[Al]s比在0.017～0.19,使A1203变性为液态所需钙的质量分数范围为(5～57)×10-6,能生成低熔点的Ca·Al203和12CaO7Al203,只要w[S]高于0.01%,则钢水中A1203夹杂很难完全变性为C12A7,而相对容易变性为CA。
(3)铝脱氧钢精炼过程稳定氧化物夹杂的研究
采用小样电解方法分析铝脱氧钢精炼过程反应产物中稳定氧化物的种类和分布及变化规律,结果表明,首钢齿轮钢中氧化物夹杂总量质量分数平均为0.0027%,其中A1203质量分数高达0.00219%,占全部氧化物夹杂的69.30%。而MnO、FeO、CaO等钢中常见的氧化物夹杂质量分数极低,不超过0.0002%；稳定氧化物中的氧占到w[T.O]的90%,而A1203和Si02氧化物中氧占到w[T.O]的～80%,稳定氧化物夹杂去除是生产低氧齿轮钢的关键；首钢齿轮钢中存在大量的M3C和M(C,N)析出物,它们主要是在凝固过程中析出的,对热处理工序中细化晶粒有较大的好处。
(4)非金属夹杂物对齿轮钢疲劳寿命的影响研究
采用旋转弯曲疲劳实验方法,对具有不同冶金质量的齿轮钢进行了疲劳破坏行为研究,测定了齿轮钢的S-N曲线,对断口断裂机制进行了分析和讨论。结果表明,齿轮钢热轧态、锻态及正火态的旋弯疲劳实验数据的分散性均较小,体现出材料具有较好的冶金质量特别是微观组织的均匀性；不同氧质量分数20CrMnTi钢正火态之间的旋弯疲劳性能差异不明显。这主要与二者的疲劳裂纹均起源于表面基体的疲劳断裂机制有关；对于起裂于表层夹杂物的试样,夹杂物十分粗大,为Al2O3·(CaO)x类复合氧化物夹杂。因此,应特别注意控制钢中此类异常夹杂物。
(5)齿轮钢窄淬透性带宽控制的研究
采用数学统计和回归的方法,找出影响齿轮钢淬透性带宽的因素,基于多支持向量机的方法建立齿轮钢淬透性值预报模型并进行在线预报,指导成分调整。取得了很好的成果：窄成分控制精度为w[C]±0.01%、w[Si]±0.02%、w[Ti]±0.02%、w[Mn]±0.02%、w[Cr]±0.03%、w[Ti]±0.01%时,所占比例分别为96.1%、94.7%、90.5%、87.4%、81.9%、90.5%,较优化前有了很大的提高；淬透性窄带J9值±2.0合格率达到93.3%,J9值±3.0合格率达到99.3%；J15值±2.0合格率达到89.2%,J9值±3.0合格率达到98.4%,较优化前有了很大地提高。
通过工艺优化,各项主要技术指标较优化前有很大地提高,w[T.O]≤15×10-6,夹杂物评级A+B+C+D<5.0级,晶粒度≥8.5级；淬透性带宽≤6HRC合格率达到99.3%,≤4HRC合格率达到93.3%；一般疏松小于0.5级,中心疏松小于1.0级,未见裂纹、夹杂、气孔等低倍缺陷；较优化前,精炼时间缩短了15.6min。
综上所述,本文对转炉流程冶炼工艺进行了细致地研究,制定了转炉生产齿轮钢的工艺制度,实现了转炉流程生产高品质齿轮钢的目的,实物质量达到国际先进水平。
In comparision with EAF process, BOF special steel making has many characteristics such as high efficientcy, large batch production, low residual impurity of material and low cost, etc. However, some quality problems exist in the BOF special steel product due to the rapid process rhythm, lacking of enough refining time, less purity, shortage of professional steel rolling and subsequent equipments. Therefore, research on smelting process to improve BOF special steel quality has a great significance.
This paper focus on BOF steelmaking process to produce special steel in Shougang steelworks. It investigated the low oxygen smelting technology of gear steel with BOF process; studied the refining slag, calcium treatment of inclusion and control of wideness harden ability bands by analyzing the statistics of history production data, testing of steel product and theoretical analysis.
The main research achievements are as follow:
(1) Shougang products and process analysis
According to the results of industry test and historical statistics data, combining with literature survey, the shortages of shougang product compared with domestic and overseas advanced manufacturer were found. The results show that the BOF slag flowover was up to12-15kg/t; w(FeO) that entered LF was up to3.0%and at the end of LF refining was2.0%; w[T.O]was up to27×10-6of mass concentration; LF refining time was close to30minutes which longer than BOF or CC.
There were excessive A and B type inclusions in steel. The inclusions modification was insufficient; most calcium aluminate inclusions located in CA6-CA2area which showed the calcium treatment was deficient. The accuracy on precise composition control was low, qualification rate of hardenability narrow band control was low, it was81.3%if J9defined as36.0±3.00HRC, however, decreased to65.3%if J9defined as36.0±2.00HRC. It means that improve the smelting process and composition control is necessary.
(2) Research on smelt Technology of gear steel with BOF process
According to thermodynamics calculating, the relationship between aluminium and oxygen content in the aluminum deoxidization steel, and the condition of inclusion generation and the condition of inclusion removing were studied by plant historical statistics data analysis. The results showed that the key points were to shorten refining time, to control slag flowover and to enhance slag modification. In order to produce steel with low oxygen w[T.O]<15x10-6, follow conditions must be satisfied: w[Al]s>0.03%, αAl2O3<0.02, w[CaO]/w[SiO2]=3.5-5and w[CaO]/w[Al2O3]=1.5-2.0. Calcium treatment should be made when the temperature of refining steel is above1600℃. The ratio of calcium and aluminum in steel should be controlled in0.017～0.19when acid-soluble aluminum is0.03%. The CaO·Al203and12CaO·7Al2O3with low melting point can be precipitated because Al2O3is liquid state when the calcium is (5-57)×10-6. When the sulphur content is higher than0.01%, the Al2O3inclusion will turn into CA easily, not C12A7.
(3) Research on oxide inclusion in aluminum deoxidization steel during refining process
Electrolysis method was used to analysis type, distribution and variability regulation of oxide inclusion in refining process of aluminum deoxidization steel, The result indicated, for Shougang steel, total quantity of oxide inclusion in the gear steel was0.0027%equally, A12O3reaches as high as0.00219%, counted69.30%of the oxide inclusion, the content of the common oxide inclusion in steel such as MnO、FeO、CaO are extremely low, less than0.0002%. Oxygen content from oxide inclusion almost took90%in ths steel,80%are from Al2O3and SiO2, so that it is the key to remove stable oxide. Shougang gear steel has massive M3C and M(C,N) precipitate, mainly precipitated during the solidification process, have great advantage to the grain refinement in heat treatment.
(4) Research on the influence of non-metallic inclusions to gear steel fatigue life
Through the rotating bending fatigue experimental, fatigue breakdown behaviours of gear steel in different metallurgy quality from industrial producing were researched, and the gear steel S-N curve was determined, also fracture break mechanism analyzed and discussed. The result indicated that, to the20CrMnTi steel in hot rolling, hammers and the normalizing condition, rotating bar bending fatigue test showed that the disparity of dates was low, and the metallurgy quality was high especially the uniformity of micro-structure, there was no obvious difference on the rotating bar bending fatigue quality between20CrMnTi steel in the normalizing condition of different oxygen content. This mainly was related to that their fatigue cracking stems from the fatigue cracking mechanism from the superficial substrate. To the sample cracked from the inclusion of the surface layer, the inclusion was Al2O3·(CaO)x and large. Therefore, special attention should be paid to the unusual inclusions.
(5) Research on the narrow hardenability band control technologies of gear steel
Impact of gear steel hardenability bandwidth factor was identified using mathematical statistics and regression analysis.
The value of gear steel hardenability prediction models was established to predict the hardenablility values on-line. The composition adjustment was guided based on Multi-SVM prediction model. Narrow component control precision is w[C]±0.01%、 w[Si]±0.02%、w[Ti]±0.02%、w[Mn]±0.02%、w[Cr]±0.03%、w[Ti]±0.01%, the respective proportions is96.1%、94.7%、90.5%、87.4%、81.9%、90.5%. The data were improved greatly than before; J9±2.0HRC qualification rate of narrow hardenability bandwidth value was93.3%, qualification rate of J9value±3.0was99.3%; Qualification rate of J15value±2.0was89.2%, qualification rate of J9value±3.0was98.4%.
Through optimization process, the main technical data were greatly improved. The results were as follows:w[T.O]≤15×106, inclusion rating A+B+C+D<5.0seris, grain size rating>8.5, percent of pass of hardenability bandwidth<6HRC was99.3%and <4HRC was93.3%, macroscopic test of general loosen was less than0.5grade, center loosen was less than1.0grade, no cracks, inclusions, porosity and other defects in the low-fold and the refining time was shorten15.6min than before.
In conclusion, this paper gave a detailed study on the converter smelting process to produce gear steel, and realized the object to produce high-quality gear steel by converter process. The statistics of the product qualities were up to international advanced level.
引文
1.中国质量协会冶金分会.中国钢材质量状况分析报告[J],中国钢铁业,2004,18(5)：17-21.
2.吴树漂,刘占江,武云峰.我国齿轮钢的生产与应用[J],特殊钢,2003,24(5)：30-33.
3.汪凡,王声堂.2005年中国齿轮钢市场分析[J],莱钢科技,2006,21(6)：4-8.
4.刘云旭,刘澄,朱启惠.汽车齿轮用渗碳钢的发展现状及前景[J],汽车工艺与材料,1998,11(9)：1-4.
5.闫咏春,杜丽娜.齿轮钢生产现状及发展[J],山西机械,2003,32(1)：15-20.
6.杨勇波.齿轮行业的现状及发展趋势[J],通用机械,2004,25(4)：34-37.
7.杨莉.齿轮钢的生产现状及发展方向[J],太钢科技,2003,19(5)：45-47.
8.陈国民.齿轮钢材和热处理质量及其控制[J],汽车工艺与材料,2003,16(1)：11-14.
9.李艳清,孙进,闫志华20CrMnTi齿轮钢产品生命周期初探与分析[J],莱钢科技,2006,21(3)：44-46.
10.卢明霞,聂爱诚,周小明.齿轮钢20CrMnTiH连铸坯轧材的质量分析[J],特殊钢,2001,22(6)：50-51.
11.陈锋.短流程生产20CrMnTi齿轮钢[J],特殊钢,2000,21(2)：50-51,
12.贾连弟.转炉试制22MnTiBRE齿轮钢[J],钢铁钒钛,1994,15(4)：15-20.
13. GAO Y, Kenichi S. Effect of Mn and Ti precipitates on the hot ductility of low carbon and ultra low carbon steels [J], ISIJ,1995,35(7):914-919.
14. Abushosha R, Vipond R, Mintz B. Influence of titanium hot ductility of as cast steels [J], Master Sci & Tech,1991,7(7):613-621.
15. Mintz B, Yue S, Jonas J. Hot ductility of steels and its relationship to the problem of transverse cracking during continous casting [J], International Materials Reviews,1991,36(5):187-216.
16.张海20CrMnTi钢中钛质量分数控制的研究[J],物理测试,2001,34(3)：1-6.
17. Ladman T G. Grain size control in steel [J], Metal Science,1974,8(6):147-175.
18.任元和.冶金因素对齿轮钢淬透性的影响[J],炼钢,1995,11(6)：16-19.
19. Iguchi M, Chihara T. Water model study of the frequency of bubble formation under reduced and elevated pressures [J], Metallurgical and Materials Transactions,1998,29(8):755-761.
20.齐欣,徐太安,高勇.降低汽车齿轮钢氧质量分数的工艺实践[J],特殊钢,2005,26(2)：54-55.
21.高志新,李强,颜丙银.电炉20CrMnTiH齿轮钢质量现状与质量分析[J],莱钢科技,2006,21(5)：9-12.
22.陈良,吴锋,魏萧20CrMnTiH钢水口堵塞机理研究与解决方案[J],莱钢科技,2006,21(4)：38-39.
23. Han Z, Holappa L. Bubble bursting phenomenon in gas/metal/slag system [J], Metallurgical and Materials Transactions,2003,34(5):525-532.
24.康显澄.高强度齿轮钢的开发[J],特钢技术,1995,3(3)：1-13.
25.姜正义,刘相华,王国栋.化学成分热轧及淬火工艺对20CrMnTiH淬透性J9的影响[J],热加工工艺,1997,26(6)：18-20.
26.赵亚平,于平波.影响20CrMnTi齿轮钢冲击性能的因素研究[J],上海钢研,2005,32(1)：34-38.
27.陈锋.汽车齿轮钢带状组织与热处理[J],上海钢研,2006,33(2)：44-49.
28.王明林,赵沛,罗海文20CrMnTi齿轮钢热塑性的研究[J],钢铁,2001,36(4)：50-53.
29. Malkiewicz T, Rudnik S. Deformation of non-metallic inclusion during rolling of steel [J], Journal of the Iron & Steel Institute,1963,8(1):33-38.
30. Tomka G, Murray N, Gore J. Effects of biaxial stress on martensitic steels [J], Journal of Magnetism and Magnetic Materials,2001,25(5):226-230.
31. Baker T. Inclusion deformation and toughness anisotropy in hot rolled steel [J], Metal Technology,1976,16(4):183-193.
32. Rudnik S. Discontinuities in hot-rolled steel caused by non-metallic inclusions [J], JISI,1966, 204(4):374-376.
33. Kiessling R. The influence of non-metallic inclusion on the properties of steel [J], Jounal of Metals,1969,17(10):48-54.
34. Bernard G, Riboud P, Urbain G. Oxide inclusions plasticity [J], Metal Technology,1981,32(5): 121-133.
35. Maeda S, Soejima T. Shape control of inclusions in wire rods for high tensile tire cord by refining with synthetic slag [A], Steelmaking Conference Proceedings[C], New York:1989, 379-385.
36. Ahmad S, Saeid H. Simulation-based prediction of hot-rolled coil forced cooling [J], Applied Thermal Engineering,2008,28(13):1630-1637.
37. Mcpherson N, Mclean A. Non-metallic inclusion in continuously cast steel [J], Continuous Casting,1995,11(7):1-16.
38.李为缪.钢中非金属夹杂物[M],北京：冶金工业出版社,1988,45-53.
39.蒋国昌.纯净钢及二次精炼[M],上海：上海科学技术出版社,1996,7-15.
40.章守华.合金钢[M],北京：冶金工业出版社,1981,307-316.
41. Brooksbank D, Andrews K. Stress field around inclusions and their relation to mechanical properties [J], The Iron & Steel Institute,1970,23(6):186-198.
42.王常珍主编.钢铁冶金物理化学研究方法[M],北京：冶金工业出版社,1982,433-439.
43. Brooksbank D, Andrews K. Thermal expension of some inclusions found in steels and relation to tessellated stresses [J], ISIJ,1968,9(6):595-599.
44. Brooksbank D. Thermal expension of calcium-aluminate inclusions and relation to tessellated stresses [J], ISIJ,1970,11(5):495-499.
45. Kishimoto Y, Yamaguchi K, Sakuraya T. Decarburization reaction in ultra-low carbon iron melt under reduced pressure [J], ISIJ,1993,33(3):391-399.
46.陈家祥.炼钢常用图表数据手册[M],北京：冶金工业出版社,1984,51-61.
47.魏季和,曲英译.高温工艺物理化学[M],北京：冶金工业出版社,1988,38-43.
48. Hirter K. The Japan society for the promotion steelmaking data sourcebook [M], Tokyo:Gordon and Breach Science Publishers,1988,125-135.
49. Sigworth K, Elliott J. The thermodynamics of liquid dilute iron alloys [J], Metal Science,1974, 18(3):298-316.
50. Song B, Han Q. Equilibrium of calcium vapours with liquid iron and the interaction of third elements [J], Metallurgical and Materials Transactions,1998,29(8):415-420.
51. Cho S, Suito H. Assessment of liquid calcium-oxygen equilibrium in liquid iron [J], ISIJ,1994, 34(3):265-269.
52.魏寿昆.冶金过程热力学[M],上海：上海科学技术出版社,1980,37-40.
53. Hiroki O, Hideaki S. Activities in CaO-SiO2-Al2O3 slag and deoxidation equilibria of Si and Al [J], Metallurgical and Materials transactions,1996,27(2):943-953.
54. Larsen K, Fruehan R J. Calcium modification of oxide inclusions [A], Steelmaking Conference Proceedings[C], New York:1990,497-506.
55. Hiroki O, Hideaki S. Activities of SiO2 and Al2O3 and activity coefficients of FeO and MnO in CaO-SiO2-Al2O3-MgO slags [J], Metallurgical and Materials transactions,1998,29(4):119-129.
56. Yoon B, Heo K. Improvement of steel cleanliness by controlling slag composition [J], Ironmaking and Steelmaking,2002,29(3):215-218.
57. Holappa L. Ladle injection metallurgy [J], International Metals Reviews,1982,27(2):53-56.
58.胡真明.正确使用钙合金脱氧[J],钢铁,1988,23(4)：13-16.
59.音谷登平.钙洁净钢[M],北京：冶金工业出版社,1994,25-31.
60. Dunn E. Calcium treatment of steel from tap stream to wire to log [A], Electric Furnace Conference Proceedings[C], New York:1988,46(2):125-130.
61. Villette F, Chapeland B. Proceedings of shenyang symposium oil line-based slagformers, refining and alloying powders [J], Casting Mold Fluxes in Iron and Steel Industry,1988,18(3): 209-216.
62.张建,姜钧普,周国平.灰色系统理论在钢包喂丝工艺中的应用[J],特殊钢,1998,19(2)：43-46.
63.宋波.钙处理钢水中非金属夹杂物的形态[J],北京科技大学学报,1995,17(2)：125-128.
64.龚坚,王庆祥.钢水钙处理的热力学分析[J],炼钢,2003,19(3)：56-59.
65. Ye G, Jonsson P. Thermodynamics and kinetics of the modification of Al2O3 inclusions [J], ISIJ, 1996,36(2):105-108.
66. Holappa L, Hamalainen M, Liukkonen M. Thermodynamic examination of inclusion modification and precipitation from calcium treatment to solidified steel [J], Inormaking and Steelmaking,2003,30(2):111-115.
67. Fujisawa T, Yamaguchi C, Sakao H. Inclusion modification in Al-killed steel [A], Proceeding of the Sixth Internatinal Iron and Steel Congress[C], Toyoko:1990,201-205.
68. Presern V, Korousic B, Hastie J. hermodynamic conditions for inclusions modification in calcium treated steel [J], Steel Research,1991,62(7):289-295.
69. Sun Z Q, Che Y C, Jiang M F. Equilibrium Diagram for Fe-Al-Ca-O-S Systems [J], Journal of Iron and Steel Research,2005,17(5):39-42.
70. Wei J, Yan G, Tian Z, Huang D H. Improvement of CSP castability in low carbon Al-killed steel [J], Journal of University of Science and Technology Beijing,2005,27(6):666-670.
71.韩志军20CrMnTiH1齿轮钢水钙处理热力学[J],钢铁,2007,36(4)：50-53.
72. Kitada T, Yamaguchi T M. Matsumura New technologies of steel bridges in Japan [J], Journal of Constructional Steel Research,2002,58(1):21-27.
73.黄希祜.钢铁冶金原理[M],北京：冶金工业出版社,2004,111-120.
74.魏季和,曲英译.高温工艺物理化学[M],北京：冶金工业出版社,1988,236-246.
75.项成云.合金结构钢[M],北京：冶金工业出版社,2004,171-190.
76.张建.化学成分对20CrMnTi钢淬透性影响的研究[J],特钢技术,1998,34(1)：36-37.
77. Brown M, Harris C. Neural fuzzy adaptive modeling and control [M], New York: Englewood Cliffs Prentice-Hall Press,1994:16-52.
78.杨钟瑾,史忠科.神经网络结构优化方法[J],计算机工程与应用,2004,25(3)：52-54.
79.咸葵.神经网络设计[M],北京：机械工业出版社,2002,227-255.
80. Vapnik C. The nature statistical learning theory [M], New York: Spring Verlag Press,1995,4-6.
81. Field P, Heymsfield A, Bansemer A. A test of ice self-selection kernels using aircraft [J], Journal of the Atmospheric Science,2006,63(4):651-666.
82.边肇棋,张学工.模式识别[M],北京：清华大学山版社,2000,179-184.
83.崔玉珍.国外汽车齿轮钢晶粒度的研究[J],机械工程材料,1990,23(3)：10-12.
84.杜丽娜.窄淬透性带齿轮钢生产试验研究[J],太钢科技,2002,14(2)：54-58.
85.张海,于辉20CrMnTiH钢的成分控制规范[J],钢铁研究学报,2001,13(4)：42-45.
86. Barron A R. Approximation and estimation bounds for artificial neural networks [J], Machine Learning,1994,23(14):115-133.
87. Guyon I, Matic N, Vapnik V. Discovering formative patterns and data cleaning [J], Advances in Knowledge Discovery & Data Mining,1996,34(2):181-203.
88. Burges C, Scholkopf B. Improving the accuracy and speed of support vector machines [J], Neural Information Processing Systems,1997,34(8):29-35.
89. Mukherjee S, Osuna E, Girosi F. Nonlinear prediction of chaotic time series using a support vector machine [A], NNSP 97[C],Toyoko:1997,159-163.
90. Kwok J. Support vector mixture for classification and regression problems [J], ICPR,1998, 12(2):35-45.
91. Suykens J, Vandewalle J. Least squares support vector machine classifiers [J], Neural Processing Letters,1999,9(3):293-300.
92. Vapnik V N, Levin E, Cun Y L. Measuring the VC dimension of a learning machine [J], Neural Computation,1994,33(6):851-876.
93.邓乃扬,田英杰.数据挖掘中的新方法-支持向量机[M],北京：科学出版社出版,2004,134-135.
94. Holland J H. Adaptation in natural and artificial system [M], Detroit: University of Michigan Press,1975,654-687.
95.王小平,曹立明.遗传算法理论应用与软件实现[M],西安：西安交通大学出版社,2002,324-365.
96. Suykens J, Vandewalle J, Moor B. Optimal control by least squres support vector machine [J], Neural Network,2001,14(2):23-35.
97. Barron A R. Approximation and estimation bounds for artificial neural networks [J], Machine Learning,1994,14(2):115-133.
98. Guyon I, Matic N, Vapnik V. Discovering formative patterns and data cleaning[J], Advances in Knowledge Discovery & Data Mining,1996,24(5):181-203.
99. Stensem S. Interactions and constitutive models for calculating quench stresses in steel [J], Materials Science and Technology,1985,32(1):823-830.
100. Allan R S, Charles G E, Mason S G. The approach of gas bubbles to a gas/liquid interface [J], Colloid Sci,1961,16(3):150-165.
101. Hartland S, Robinson J D, Can J. The profile of the draining film beneath a liquid drop approaching a plate interface [J], Chemical Engineering Science,1973,51(12):647-655.
102. Chevrier V, Cramb A. Observation and measurement of bubble separation at liquid-liquid interfaces [J], Steel Research International,2007,75(10):645-658.
103. Szekely J, Fang S D. Studies in vacuum degassing mass and momentum transfer to gas bubbles rising in melts [J], Metallurgical Transactions,1974,16(5):1429-1436.
104.宋广胜,刘相华,王国栋.齿轮钢淬火过程多场耦合分析[J],特殊钢,2007,28(2)：7-9.
2.吴树漂,刘占江,武云峰.我国齿轮钢的生产与应用[J],特殊钢,2003,24(5)：30-33.
3.汪凡,王声堂.2005年中国齿轮钢市场分析[J],莱钢科技,2006,21(6)：4-8.
4.刘云旭,刘澄,朱启惠.汽车齿轮用渗碳钢的发展现状及前景[J],汽车工艺与材料,1998,11(9)：1-4.
5.闫咏春,杜丽娜.齿轮钢生产现状及发展[J],山西机械,2003,32(1)：15-20.
6.杨勇波.齿轮行业的现状及发展趋势[J],通用机械,2004,25(4)：34-37.
7.杨莉.齿轮钢的生产现状及发展方向[J],太钢科技,2003,19(5)：45-47.
8.陈国民.齿轮钢材和热处理质量及其控制[J],汽车工艺与材料,2003,16(1)：11-14.
9.李艳清,孙进,闫志华20CrMnTi齿轮钢产品生命周期初探与分析[J],莱钢科技,2006,21(3)：44-46.
10.卢明霞,聂爱诚,周小明.齿轮钢20CrMnTiH连铸坯轧材的质量分析[J],特殊钢,2001,22(6)：50-51.
11.陈锋.短流程生产20CrMnTi齿轮钢[J],特殊钢,2000,21(2)：50-51,
12.贾连弟.转炉试制22MnTiBRE齿轮钢[J],钢铁钒钛,1994,15(4)：15-20.
13. GAO Y, Kenichi S. Effect of Mn and Ti precipitates on the hot ductility of low carbon and ultra low carbon steels [J], ISIJ,1995,35(7):914-919.
14. Abushosha R, Vipond R, Mintz B. Influence of titanium hot ductility of as cast steels [J], Master Sci & Tech,1991,7(7):613-621.
15. Mintz B, Yue S, Jonas J. Hot ductility of steels and its relationship to the problem of transverse cracking during continous casting [J], International Materials Reviews,1991,36(5):187-216.
16.张海20CrMnTi钢中钛质量分数控制的研究[J],物理测试,2001,34(3)：1-6.
17. Ladman T G. Grain size control in steel [J], Metal Science,1974,8(6):147-175.
18.任元和.冶金因素对齿轮钢淬透性的影响[J],炼钢,1995,11(6)：16-19.
19. Iguchi M, Chihara T. Water model study of the frequency of bubble formation under reduced and elevated pressures [J], Metallurgical and Materials Transactions,1998,29(8):755-761.
20.齐欣,徐太安,高勇.降低汽车齿轮钢氧质量分数的工艺实践[J],特殊钢,2005,26(2)：54-55.
21.高志新,李强,颜丙银.电炉20CrMnTiH齿轮钢质量现状与质量分析[J],莱钢科技,2006,21(5)：9-12.
22.陈良,吴锋,魏萧20CrMnTiH钢水口堵塞机理研究与解决方案[J],莱钢科技,2006,21(4)：38-39.
23. Han Z, Holappa L. Bubble bursting phenomenon in gas/metal/slag system [J], Metallurgical and Materials Transactions,2003,34(5):525-532.
24.康显澄.高强度齿轮钢的开发[J],特钢技术,1995,3(3)：1-13.
25.姜正义,刘相华,王国栋.化学成分热轧及淬火工艺对20CrMnTiH淬透性J9的影响[J],热加工工艺,1997,26(6)：18-20.
26.赵亚平,于平波.影响20CrMnTi齿轮钢冲击性能的因素研究[J],上海钢研,2005,32(1)：34-38.
27.陈锋.汽车齿轮钢带状组织与热处理[J],上海钢研,2006,33(2)：44-49.
28.王明林,赵沛,罗海文20CrMnTi齿轮钢热塑性的研究[J],钢铁,2001,36(4)：50-53.
29. Malkiewicz T, Rudnik S. Deformation of non-metallic inclusion during rolling of steel [J], Journal of the Iron & Steel Institute,1963,8(1):33-38.
30. Tomka G, Murray N, Gore J. Effects of biaxial stress on martensitic steels [J], Journal of Magnetism and Magnetic Materials,2001,25(5):226-230.
31. Baker T. Inclusion deformation and toughness anisotropy in hot rolled steel [J], Metal Technology,1976,16(4):183-193.
32. Rudnik S. Discontinuities in hot-rolled steel caused by non-metallic inclusions [J], JISI,1966, 204(4):374-376.
33. Kiessling R. The influence of non-metallic inclusion on the properties of steel [J], Jounal of Metals,1969,17(10):48-54.
34. Bernard G, Riboud P, Urbain G. Oxide inclusions plasticity [J], Metal Technology,1981,32(5): 121-133.
35. Maeda S, Soejima T. Shape control of inclusions in wire rods for high tensile tire cord by refining with synthetic slag [A], Steelmaking Conference Proceedings[C], New York:1989, 379-385.
36. Ahmad S, Saeid H. Simulation-based prediction of hot-rolled coil forced cooling [J], Applied Thermal Engineering,2008,28(13):1630-1637.
37. Mcpherson N, Mclean A. Non-metallic inclusion in continuously cast steel [J], Continuous Casting,1995,11(7):1-16.
38.李为缪.钢中非金属夹杂物[M],北京：冶金工业出版社,1988,45-53.
39.蒋国昌.纯净钢及二次精炼[M],上海：上海科学技术出版社,1996,7-15.
40.章守华.合金钢[M],北京：冶金工业出版社,1981,307-316.
41. Brooksbank D, Andrews K. Stress field around inclusions and their relation to mechanical properties [J], The Iron & Steel Institute,1970,23(6):186-198.
42.王常珍主编.钢铁冶金物理化学研究方法[M],北京：冶金工业出版社,1982,433-439.
43. Brooksbank D, Andrews K. Thermal expension of some inclusions found in steels and relation to tessellated stresses [J], ISIJ,1968,9(6):595-599.
44. Brooksbank D. Thermal expension of calcium-aluminate inclusions and relation to tessellated stresses [J], ISIJ,1970,11(5):495-499.
45. Kishimoto Y, Yamaguchi K, Sakuraya T. Decarburization reaction in ultra-low carbon iron melt under reduced pressure [J], ISIJ,1993,33(3):391-399.
46.陈家祥.炼钢常用图表数据手册[M],北京：冶金工业出版社,1984,51-61.
47.魏季和,曲英译.高温工艺物理化学[M],北京：冶金工业出版社,1988,38-43.
48. Hirter K. The Japan society for the promotion steelmaking data sourcebook [M], Tokyo:Gordon and Breach Science Publishers,1988,125-135.
49. Sigworth K, Elliott J. The thermodynamics of liquid dilute iron alloys [J], Metal Science,1974, 18(3):298-316.
50. Song B, Han Q. Equilibrium of calcium vapours with liquid iron and the interaction of third elements [J], Metallurgical and Materials Transactions,1998,29(8):415-420.
51. Cho S, Suito H. Assessment of liquid calcium-oxygen equilibrium in liquid iron [J], ISIJ,1994, 34(3):265-269.
52.魏寿昆.冶金过程热力学[M],上海：上海科学技术出版社,1980,37-40.
53. Hiroki O, Hideaki S. Activities in CaO-SiO2-Al2O3 slag and deoxidation equilibria of Si and Al [J], Metallurgical and Materials transactions,1996,27(2):943-953.
54. Larsen K, Fruehan R J. Calcium modification of oxide inclusions [A], Steelmaking Conference Proceedings[C], New York:1990,497-506.
55. Hiroki O, Hideaki S. Activities of SiO2 and Al2O3 and activity coefficients of FeO and MnO in CaO-SiO2-Al2O3-MgO slags [J], Metallurgical and Materials transactions,1998,29(4):119-129.
56. Yoon B, Heo K. Improvement of steel cleanliness by controlling slag composition [J], Ironmaking and Steelmaking,2002,29(3):215-218.
57. Holappa L. Ladle injection metallurgy [J], International Metals Reviews,1982,27(2):53-56.
58.胡真明.正确使用钙合金脱氧[J],钢铁,1988,23(4)：13-16.
59.音谷登平.钙洁净钢[M],北京：冶金工业出版社,1994,25-31.
60. Dunn E. Calcium treatment of steel from tap stream to wire to log [A], Electric Furnace Conference Proceedings[C], New York:1988,46(2):125-130.
61. Villette F, Chapeland B. Proceedings of shenyang symposium oil line-based slagformers, refining and alloying powders [J], Casting Mold Fluxes in Iron and Steel Industry,1988,18(3): 209-216.
62.张建,姜钧普,周国平.灰色系统理论在钢包喂丝工艺中的应用[J],特殊钢,1998,19(2)：43-46.
63.宋波.钙处理钢水中非金属夹杂物的形态[J],北京科技大学学报,1995,17(2)：125-128.
64.龚坚,王庆祥.钢水钙处理的热力学分析[J],炼钢,2003,19(3)：56-59.
65. Ye G, Jonsson P. Thermodynamics and kinetics of the modification of Al2O3 inclusions [J], ISIJ, 1996,36(2):105-108.
66. Holappa L, Hamalainen M, Liukkonen M. Thermodynamic examination of inclusion modification and precipitation from calcium treatment to solidified steel [J], Inormaking and Steelmaking,2003,30(2):111-115.
67. Fujisawa T, Yamaguchi C, Sakao H. Inclusion modification in Al-killed steel [A], Proceeding of the Sixth Internatinal Iron and Steel Congress[C], Toyoko:1990,201-205.
68. Presern V, Korousic B, Hastie J. hermodynamic conditions for inclusions modification in calcium treated steel [J], Steel Research,1991,62(7):289-295.
69. Sun Z Q, Che Y C, Jiang M F. Equilibrium Diagram for Fe-Al-Ca-O-S Systems [J], Journal of Iron and Steel Research,2005,17(5):39-42.
70. Wei J, Yan G, Tian Z, Huang D H. Improvement of CSP castability in low carbon Al-killed steel [J], Journal of University of Science and Technology Beijing,2005,27(6):666-670.
71.韩志军20CrMnTiH1齿轮钢水钙处理热力学[J],钢铁,2007,36(4)：50-53.
72. Kitada T, Yamaguchi T M. Matsumura New technologies of steel bridges in Japan [J], Journal of Constructional Steel Research,2002,58(1):21-27.
73.黄希祜.钢铁冶金原理[M],北京：冶金工业出版社,2004,111-120.
74.魏季和,曲英译.高温工艺物理化学[M],北京：冶金工业出版社,1988,236-246.
75.项成云.合金结构钢[M],北京：冶金工业出版社,2004,171-190.
76.张建.化学成分对20CrMnTi钢淬透性影响的研究[J],特钢技术,1998,34(1)：36-37.
77. Brown M, Harris C. Neural fuzzy adaptive modeling and control [M], New York: Englewood Cliffs Prentice-Hall Press,1994:16-52.
78.杨钟瑾,史忠科.神经网络结构优化方法[J],计算机工程与应用,2004,25(3)：52-54.
79.咸葵.神经网络设计[M],北京：机械工业出版社,2002,227-255.
80. Vapnik C. The nature statistical learning theory [M], New York: Spring Verlag Press,1995,4-6.
81. Field P, Heymsfield A, Bansemer A. A test of ice self-selection kernels using aircraft [J], Journal of the Atmospheric Science,2006,63(4):651-666.
82.边肇棋,张学工.模式识别[M],北京：清华大学山版社,2000,179-184.
83.崔玉珍.国外汽车齿轮钢晶粒度的研究[J],机械工程材料,1990,23(3)：10-12.
84.杜丽娜.窄淬透性带齿轮钢生产试验研究[J],太钢科技,2002,14(2)：54-58.
85.张海,于辉20CrMnTiH钢的成分控制规范[J],钢铁研究学报,2001,13(4)：42-45.
86. Barron A R. Approximation and estimation bounds for artificial neural networks [J], Machine Learning,1994,23(14):115-133.
87. Guyon I, Matic N, Vapnik V. Discovering formative patterns and data cleaning [J], Advances in Knowledge Discovery & Data Mining,1996,34(2):181-203.
88. Burges C, Scholkopf B. Improving the accuracy and speed of support vector machines [J], Neural Information Processing Systems,1997,34(8):29-35.
89. Mukherjee S, Osuna E, Girosi F. Nonlinear prediction of chaotic time series using a support vector machine [A], NNSP 97[C],Toyoko:1997,159-163.
90. Kwok J. Support vector mixture for classification and regression problems [J], ICPR,1998, 12(2):35-45.
91. Suykens J, Vandewalle J. Least squares support vector machine classifiers [J], Neural Processing Letters,1999,9(3):293-300.
92. Vapnik V N, Levin E, Cun Y L. Measuring the VC dimension of a learning machine [J], Neural Computation,1994,33(6):851-876.
93.邓乃扬,田英杰.数据挖掘中的新方法-支持向量机[M],北京：科学出版社出版,2004,134-135.
94. Holland J H. Adaptation in natural and artificial system [M], Detroit: University of Michigan Press,1975,654-687.
95.王小平,曹立明.遗传算法理论应用与软件实现[M],西安：西安交通大学出版社,2002,324-365.
96. Suykens J, Vandewalle J, Moor B. Optimal control by least squres support vector machine [J], Neural Network,2001,14(2):23-35.
97. Barron A R. Approximation and estimation bounds for artificial neural networks [J], Machine Learning,1994,14(2):115-133.
98. Guyon I, Matic N, Vapnik V. Discovering formative patterns and data cleaning[J], Advances in Knowledge Discovery & Data Mining,1996,24(5):181-203.
99. Stensem S. Interactions and constitutive models for calculating quench stresses in steel [J], Materials Science and Technology,1985,32(1):823-830.
100. Allan R S, Charles G E, Mason S G. The approach of gas bubbles to a gas/liquid interface [J], Colloid Sci,1961,16(3):150-165.
101. Hartland S, Robinson J D, Can J. The profile of the draining film beneath a liquid drop approaching a plate interface [J], Chemical Engineering Science,1973,51(12):647-655.
102. Chevrier V, Cramb A. Observation and measurement of bubble separation at liquid-liquid interfaces [J], Steel Research International,2007,75(10):645-658.
103. Szekely J, Fang S D. Studies in vacuum degassing mass and momentum transfer to gas bubbles rising in melts [J], Metallurgical Transactions,1974,16(5):1429-1436.
104.宋广胜,刘相华,王国栋.齿轮钢淬火过程多场耦合分析[J],特殊钢,2007,28(2)：7-9.