离心压缩机叶轮材料FV520B超高周疲劳行为与机理研究

发布时间：2018-04-23 02:11

本文选题：超高周疲劳 + FV520B　；参考：《山东大学》2015年博士论文

【摘要】：离心压缩机叶轮属于高速回转类零件,由于尾流激振、机组振动等原因,其往往承受高频低应力幅循环载荷作用,疲劳寿命要求远远超过107周次,进入超高周疲劳范围。离心压缩机叶轮重要材料FV520B的超高周疲劳行为与机理研究不仅为离心压缩机叶轮的长寿命疲劳设计与疲劳失效预防提供了指导,也为后续离心压缩机叶轮可再制造临界阈值的判定提供了有力支持。作者首先对某离心压缩机叶轮力学特性进行了分析,在此基础上研究了高低温时效处理、试样尺寸、表面粗糙度、焊接对FV520B超高周疲劳行为与机理的影响。利用有限单元法,对空分空压机组H418低压缸一级叶轮及磨损致叶片局部减薄后该叶轮的相关力学特性进行了分析。结果表明：(1)仅承受离心载荷时,叶根中部应力最大。(2)叶轮在额定转速附近运行时,应力集中主要发生在叶片前缘中部,与该叶轮服役时的疲劳断裂部位吻合。(3)磨损导致的叶片局部减薄为“小失谐”,叶轮失谐前后的结构拓扑形式基本不变,失谐叶轮模态频率和各阶振型与谐调叶轮相比变化不大。(4)尾流激振引起叶轮共振时,叶片前缘中部最危险,可作为叶轮疲劳试验的重点考察部位,此位置的最大交变载荷幅值为33.3MPa,叶片前缘的振动与一端为滑动质量块的悬臂梁的弯曲振动相似。研究了高低温时效处理对FV520B超高周疲劳行为与机理的影响。通过对称拉压超声疲劳试验方法对FV520B经过不同温度时效处理后得到的FV520B-I、 FV520B-S的超高周疲劳性能进行测试,获得超高周疲劳S-N曲线,并通过断口形貌观察和特征区域尺寸测量,对其超高周疲劳机理进行研究。FV520B-I与FV520B-S的超高周疲劳行为与机理存在较大差异。FV520B-I的S-N曲线在109周次试验范围内呈持续下降的趋势,起裂模式的不同导致曲线分为斜率明显不同的两部分。FV520B-S的S-N曲线大幅低于FV520B-I,甚至从106循环周次处便出现“传统疲劳极限”。在超高周疲劳阶段,FV520B-I裂纹主要萌生于内部夹杂物,少数裂纹萌生于表面基体或内部较软的铁素体。FV520B-I超高周疲劳断口鱼眼区的形成与裂纹源距试样表面距离有关,裂纹源距试样表面越近,内部裂纹转变为表面裂纹越快,鱼眼区内外裂纹扩展速率差别越大,鱼眼区边界越明显。FV520B-I超高周疲劳寿命随GBF (granular bright facet)区直径的增大而增大,而与其他特征区域尺寸的关系并不明显。夹杂物处起裂裂纹的萌生寿命不是超高周疲劳寿命的主要部分,超高周疲劳寿命主要消耗在GBF区的形成上。夹杂物直径对GBF区直径有一定影响,夹杂物直径显著减小时,GBF区直径也会显著减小。在Murakami模型的基础上,通过断裂力学分析考察了夹杂物形状对疲劳强度预测的影响。考虑夹杂物形状时,修正模型的超高周疲劳强度预测值比Murakami模型预测值略高2%,修正模型对FV520B-I超高周疲劳强度的预测更接近试验结果。基于Paris公式和腐蚀疲劳裂纹萌生寿命模型对FV520B-I超高周疲劳寿命进行拟合,基于腐蚀疲劳裂纹萌生寿命模型的拟合结果明显优于基于Paris公式的拟合结果。通过对称拉压超声疲劳试验方法对FV520B-I试样尺寸增大时的超高周疲劳性能进行测试,考察试样尺寸对FV520B-I超高周疲劳行为与机理的影响。试样尺寸增大后,试样中夹杂物尺寸增大,试样热效应更明显,S-N曲线下移,相同应力水平下,疲劳寿命缩短,109循环周次下的疲劳强度降低,但FV520B-I超高周疲劳机理未发生本质变化。与小尺寸试样试验结果相似,大尺寸试样的超高周疲劳寿命与裂纹源距试样表面距离、鱼眼区直径没有明显关系。载荷一定时,超高周疲劳寿命随夹杂物直径的减小总体上呈增大的趋势,随GBF区直径与夹杂物直径比值的增大而增大。对于大尺寸试样,夹杂物处起裂裂纹的萌生寿命同样不是超高周疲劳寿命的主要部分,超高周疲劳寿命主要消耗在GBF区的形成上。采用统计极值方法对不同尺寸试样中可能的最大夹杂物尺寸进行预测,并进一步通过Murakami模型以及考虑夹杂物形状影响后的修正模型对超高周疲劳强度进行估算,相较于对小尺寸试样试验结果的预测,上述模型特别是修正模型对大尺寸试样超高周疲劳强度的预测更接近试验结果。利用腐蚀疲劳裂纹萌生寿命模型对大尺寸试样的超高周疲劳寿命进行拟合,相较于对小尺寸试样超高周疲劳寿命的拟合,拟合效果有所下降。采用对称拉压超声疲劳试验方法测试了接近叶轮真实表面粗糙度情况下FV520B-I在109周次范围内的超高周疲劳性能,并与前两组光滑试样的试验结果进行对比,分析了表面粗糙度对FV520B-I超高周疲劳行为与机理的影响。随着试样表面粗糙度的增大,S-N曲线下移,表面裂纹向内部裂纹转变的应力幅值降低,直至出现“传统疲劳极限”。将试样危险截面处的最深沟槽作为一条单独裂纹处理,预测得到的表面疲劳极限相较于Murakami模型预测值与试验结果更为接近。对于Ra≈0.05,Ra≈0.2的试样,应力幅值较高时,很多试样中的裂纹未能充分扩展,测试得到的疲劳寿命偏小,试验结果小于平行层模型预测值。对于Ra≈0.6的试样,应力幅值较高时,裂纹能充分扩展,测试结果较为准确,平行层模型预测值与试验结果吻合较好。试样从表面或内部起裂是由表面裂纹或内部裂纹进入到扩展阶段的先后决定的,GBF区内裂纹生长极慢,该区域的形成与裂纹萌生相近,可采用参数D*=NG/NS表征裂纹起裂模式的竞争。采用漏斗形试样着重对表面粗糙度不同时FV520B-I焊缝的超高周疲劳行为与机理进行研究。相较于母材,FV520B-I焊缝试样的起裂模式更多,焊接过程中产生的气孔和药皮夹渣都能引起疲劳开裂。FV520B-I焊缝高周疲劳裂纹主要萌生于表面,超高周疲劳裂纹主要萌生于熔池冶金反应形成的复杂非金属氧化物。GBF区存在C富集,佐证了GBF区形成的碳化物弥散减聚机制。GBF区的形成是裂纹由短裂纹群体行为逐渐演化为单条长裂纹行为的过程。靠近夹杂物,应力集中越大,有效短裂纹密度越高,断面粗糙度越大；远离夹杂物,应力集中越小,有效短裂纹密度越低,断面粗糙度越小。FV520B-I光滑焊缝具有明显的超高周疲劳特征,疲劳寿命高于107周次时,裂纹主要在内部萌生。表面粗糙度增大后,S-N曲线有所下降,裂纹倾向于表面萌生。Murakami模型对表面疲劳极限的预测远大于试验值,平行层模型则能较好的预测表面起裂高周疲劳裂纹的疲劳寿命。夹杂物(气孔)处起裂裂纹的萌生寿命不是超高周疲劳寿命的主要部分,超高周疲劳寿命随夹杂物(气孔)直径的减小而增大,随GBF区直径与夹杂物(气孔)直径比值的增大而增大。夹杂物(气孔)直径减小导致应力集中降低,减缓了有效短裂纹密度的增加,GBF直径的增大增加了主导有效短裂纹的扩展距离。总之,通过离心压缩机叶轮的力学特性分析,确定了叶轮典型的高频低幅疲劳载荷及危险部位,为后续离心压缩机叶轮材料FV520B的超高周疲劳试验提供了参考。通过对FV520B的超高周疲劳行为与机理进行研究,揭示了热处理状态、试样尺寸、表面粗糙度、焊接对FV520B超高周疲劳行为与机理的影响规律,为离心压缩机叶轮的剩余寿命估算以及可再制造临界阈值的判定提供了支持。然而,由于时间和试验条件的限制,FV520B-I母材狗骨型试样与FV520B-I焊接接头狗骨型试样的超高周疲劳试验有待继续开展。
[Abstract]:Centrifugal compressor impeller is a high speed rotary part. Due to the wake excitation and vibration of the unit, it often bears the function of high frequency and low stress amplitude cycle load. The fatigue life requirement is far more than 107 weeks and enters the ultra high cycle fatigue range. The ultra high cycle fatigue behavior and mechanism of the important material FV520B of centrifugal compressor impeller is not only used for the study of the fatigue behavior and mechanism of the centrifugal compressor impeller. The long life fatigue design and fatigue failure prevention of centrifugal compressor impeller provide guidance, and provide strong support for the determination of the critical threshold for the remanufacture of the centrifugal compressor impeller. First, the author analyses the mechanical characteristics of a centrifugal compressor impeller. On this basis, the high and low temperature aging treatment, the sample size, and the table are studied. The effect of surface roughness and welding on the high cycle fatigue behavior and mechanism of FV520 B ultrasonic. The finite element method was used to analyze the mechanical properties of the impeller after the partial reduction of the first stage impeller and the worn blade of the H418 low pressure cylinder of air separation air compressor unit. The results showed that (1) the maximum stress in the middle part of the blade root was the maximum. (2) the impeller was in the amount. When the fixed speed is running, the stress concentration occurs mainly in the middle of the blade front, which is consistent with the fatigue fracture site when the impeller is in service. (3) the local thinning of the blade is reduced to "small detuning", and the structure topology of the impeller is basically unchanged before and after the detuning, and the frequency of the mode state of the detuning impeller and the modes of each order have little change compared with the harmonic impellers. (4) when the wake excited vibration causes the impeller resonance, the central part of the blade front is the most dangerous. It can be used as the key site of the impeller fatigue test. The maximum amplitude of the alternating load is 33.3MPa. The vibration of the front edge of the blade is similar to the bending vibration of a cantilever beam with a sliding mass. The high and low temperature aging treatment has been studied for the high cycle fatigue of FV520 B ultrasonic. The ultra high cycle fatigue performance of FV520B-I and FV520B-S obtained by FV520B after different temperature aging treatment was tested by symmetrical tension pressure ultrasonic fatigue test. The ultra high cycle fatigue S-N curve was obtained. The ultrahigh cycle fatigue mechanism was studied by the fracture morphology observation and the characteristic regional scale measurement. There is a great difference between the ultra high cycle fatigue behavior and mechanism of FV520B-I and FV520B-S. The S-N curve of.FV520B-I has a tendency to decline continuously within the range of 109 weeks. The S-N curve of the curve divided into a distinct slope with different slopes is significantly lower than that of FV520B-I, and the "tradition" appears even from the 106 cycle. In the stage of ultra high cycle fatigue, the FV520B-I cracks mainly erupt in the internal inclusions, a few cracks occur on the surface of the surface or the soft ferrite in the.FV520B-I ultra high cycle fatigue fracture area, which is related to the distance between the crack source and the sample surface. The crack source is closer to the specimen surface, and the internal crack is transformed into a surface crack. The faster the difference in the rate of crack propagation in the fish eye area, the more obvious the.FV520B-I ultra high cycle fatigue life of the fish eye area increases with the increase of the diameter of the GBF (granular bright facet) region, but the relationship with the other characteristic region size is not obvious. The ultrahigh cycle fatigue life is mainly consumed in the formation of the GBF region. The diameter of inclusions has a certain effect on the diameter of the GBF region, the diameter of the inclusion is significantly reduced, and the diameter of the GBF region will be reduced significantly. On the basis of the Murakami model, the influence of the inclusion shape on the fatigue strength prediction is investigated by the fracture mechanics analysis. The prediction value of ultra high cycle fatigue strength of the modified model is slightly higher than that of the Murakami model. The modified model is closer to the test results for the prediction of FV520B-I ultra high cycle fatigue strength. Based on the Paris formula and the corrosion fatigue crack initiation life model, the ultra high cycle fatigue life of the FV520B-I is fitted and the fatigue crack initiation life model is based on the corrosion fatigue crack initiation life model. The fitting results are obviously superior to the fitting results based on the Paris formula. The ultra high cycle fatigue performance of the FV520B-I specimen is tested by the symmetrical tension pressure ultrasonic fatigue test, and the effect of the sample size on the ultra high cycle fatigue behavior and mechanism of FV520B-I is investigated. The size of the inclusions in the sample increases after the sample size increases, and the sample size increases. The thermal effect is more obvious, the S-N curve moves down, the fatigue life shortens under the same stress level, and the fatigue strength decreases under the 109 cycle cycle, but the ultra high cycle fatigue mechanism of FV520B-I does not change essentially. When the load is fixed, the ultra high cycle fatigue life increases with the decrease of the inclusion diameter, and increases with the increase of the ratio of the diameter of the GBF area to the inclusion diameter. For large size specimens, the initiation life of the crack crack at the inclusion is not the main part of the ultra high cycle fatigue life, and the ultra high cycle fatigue life is the same. It is mainly consumed in the formation of the GBF region. The maximum inclusion size in different size samples is predicted by the statistical extremum method, and the super high cycle fatigue strength is estimated by the Murakami model and the modified model considering the influence of the inclusion shape, compared with the prediction of the test results for small size samples. The model, especially the modified model, is more close to the test results for the prediction of the ultra high cycle fatigue strength of large size specimens. Using the corrosion fatigue crack initiation life model to fit the super high cycle fatigue life of large size specimens, compared with the fitting of the ultra-high cycle fatigue life of small size specimens, the fitting effect is reduced. The ultrasonic fatigue test method was used to test the ultra high cycle fatigue performance of FV520B-I in the 109 week range near the actual surface roughness of the impeller, and compared with the test results of the previous two groups of smooth samples. The effect of surface roughness on the ultra high cycle fatigue behavior and mechanism of FV520B-I was analyzed. With the increase of the surface roughness of the sample, S-N When the curve moves down, the stress amplitude of the change of the surface crack to the internal crack is reduced until the "traditional fatigue limit" appears. The most deep groove at the dangerous section of the specimen is treated as a single crack. The predicted surface fatigue limit is closer to the predicted value of the Murakami model than the test result. For the test of Ra 0.05, the test of Ra 0.2 When the stress amplitude is high, the cracks in many samples are not fully expanded, the fatigue life of the test is smaller and the test results are less than the predicted value of the parallel layer model. For the sample of Ra 0.6, the crack can be fully expanded, the test result is more accurate, the prediction value of the parallel layer model is in good agreement with the test result. The crack from surface or inside is determined by the entry of the surface crack or the internal crack to the expansion stage. The crack growth in the GBF region is very slow. The formation of the region is similar to the crack initiation. The competition of the crack initiation mode can be characterized by the parameter D*=NG/NS. The use of the funnel-shaped specimen is focused on the super high surface roughness of the FV520B-I weld. The behavior and mechanism of the cycle fatigue are studied. Compared with the parent material, the crack initiation mode of the FV520B-I welds is more. The pores produced in the welding process and the dregs can cause the fatigue cracking.FV520B-I weld high cycle fatigue crack initiation on the surface, and the ultra high cycle fatigue crack mainly originated from the complex non-metal formed by the metallurgical reaction of the weld pool. The presence of C enrichment in the oxide.GBF region shows that the formation of the carbide dispersion reduction mechanism of the GBF zone is the process of the crack evolution from the short crack group behavior to the single long crack. The greater the stress concentration is near the inclusion, the higher the density of the effective short crack, the greater the roughness of the section, and the smaller the stress concentration away from the inclusions. The lower the effective short crack density is, the smaller the cross section roughness is, the.FV520B-I smooth weld has obvious ultra high cycle fatigue characteristics. The fatigue life is higher than 107 weeks, and the crack is mainly in the interior. After the surface roughness increases, the S-N curve decreases, and the crack tends to the surface initiation.Murakami model to predict the surface fatigue limit far greater than the test. The parallel layer model can better predict the fatigue life of the high cycle fatigue crack on the surface. The initiation life of the crack in the inclusion (air hole) is not the main part of the ultra high cycle fatigue life. The ultra high cycle fatigue life increases with the decrease of the diameter of the inclusions (the pores), and increases with the ratio of the diameter of the GBF area to the inclusion (air hole) diameter. The decrease of the diameter of the inclusions (holes) leads to the decrease of the stress concentration and the increase of the effective short crack density. The increase of the GBF diameter increases the extended distance of the dominant and effective short crack. In a word, the typical high frequency and low amplitude fatigue load and dangerous part of the impeller are determined by the analysis of the mechanical characteristics of the impeller of the centrifugal compressor. The ultra high cycle fatigue test of the centrifugal compressor impeller material FV520B provides a reference. Through the study of the ultra high cycle fatigue behavior and mechanism of FV520B, the effect of heat treatment state, sample size, surface roughness, welding on the behavior and mechanism of FV520 B ultrasonic high cycle fatigue is revealed, and the residual life of the centrifugal compressor impeller is estimated. Support is provided for the determination of the remanufacturing critical threshold. However, due to time and test conditions, the ultra high cycle fatigue test of the FV520B-I parent dog bone specimen and the FV520B-I welded joint dog bone specimen needs to be carried out.

【学位授予单位】：山东大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：TH452

【引证文献】