轮轨作用力和接触点位置在线测量理论研究
|
摘要
轮轨作用力的测量对于车辆动力学、脱轨机理、轮轨接触理论研究和列车的安全检测都具有十分重要的意义。铁路运输不断向高速、重载、大运量和高密度方向发展,对轮轨作用力测量的精度和速度提出了更高的要求。测力轮对方法是目前最直接、最准确的轮轨力测量技术,它以车轮作为轮轨力的检测传感器,通过测量车轮辐板有限点处的应变实现轮轨接触力的连续检测。车辆运行过程中,车轮受到轨道不平顺作用,轮轨作用力是时变、非平稳的,轮轨力引起的辐板应变又被车轮转角所调制。测力轮对测量精度受到车轮均匀性、贴片工艺、车轮转速、轮轨力作用点位置变化和外界干扰的影响,测力轮对技术在测量理论和数据处理方法上都需要进一步研究。
论文在回顾了测力轮对的发展历程和国内外测力轮对的发展现状的基础上,分析了轮轨载荷作用下轮对辐板的应力分布,提出了消除车轮旋转影响的测力轮对测量方法,研究了轮轨力作用点位置对测量精度的影响和测力轮对数据处理方法,其内容与成果如下:
1、提出轮轨力测量模型
测力轮对技术经过几十年的发展,在车辆动力学试验中得到广泛使用,但在轮轨力连续测量中仍存在很多问题。目前的连续测力轮对技术要求轮轨横向力、垂向力产生的辐板应变组桥后沿圆周方向按正、余弦分布,这不仅对轮对的均匀性和贴片工艺提出了很高的要求,实际上在轮对辐板大部分区域是根本做不到,电桥输出中包含的高次谐波可能带来很大的测量误差。测力轮对存在的另一个问题是没有考虑轮轨作用点位置变化对测量精度的影响。车辆在运行过程中,轮对与轨道的接触点在滚动圆附近不停变化,尤其是在过曲线、道盆和蛇形运动时,接触点位置远离滚动圆。国内现有的测力轮对都没有考虑接触点位置变化,标定实验、数据处理都假设接触点在滚动圆上,测量结果误差较大。
论文在有限元分析和实验基础上,分析了轮对辐板应变在横向力、垂向力和作用点位置变化产生的弯矩作用下的分布规律,采用同一半径上多个应变片组合消除高次谐波,研究了需要消除的谐波分量阶次与电桥的数量、各电桥相位差之间的关系,为提高测力轮对测量精度提供了理论依据。在此基础上,建立轮对辐板应变与轮轨作用力、作用点位置的非线性方程组。
论文研究了测量模型非线性方程组的直接求解方法。非线性方程组一般没有解析解,通常采用迭代数值方法求解。迭代解法需要选择初值,初值不仅影响计算速度,同时影响计算精度,甚至导致计算不收敛,误差分析也比较困难。论文对非线性方程组进行合理简化,推导出方程组的解析解。
论文还研究了将非线性方程组简化为线性方程组的方法。使用主分量分析,得到各载荷单独作用时辐板应变信号的子空间,将测量信号投影到各个子空间从而建立线性方程组。可以利用线性方程组解的稳定性理论选择合理的贴片位置,减少应变测量误差对轮轨力计算值的影响。
针对实际轮对均匀性和贴片位置达不到测力轮对要求的情况,论文提出了新的解决方法。通过标定试验得到转角、作用点位置与对应的横向力、垂向力作用系数的数表,并用神经网络拟合辐板应变和车轮转角间的非线性关系,采用遗传算法获得最佳的作用点位置和电桥直流偏置,对每个电桥建立辐板应变与轮轨力间的线性方程组。这种方法可以有效地降低测力轮对的开发成本和研制时间。
2、提出轮轨作用点位置在线测量方法
轮轨作用点位置发生变化时,垂向力会产生一个附加弯矩,在轮对辐板产生附加应变,这一直被作为测力轮对的一个误差源对待。与横向力、垂向力在辐板上产生的应变相比,附加弯矩产生的应变比较微弱,论文提出的轮轨力测量方案,不仅提高了横向力和垂向力的测量精度,而且可以检测轮轨作用点位置。
3、分析车轮辐板应变的时频特性
轮轨力虽然是非平稳信号,但在很短的时间内可以看成平稳的。论文对滚动振动试验台实验数据进行了时频分析,结果表明用短时傅立叶分析是可行的。轮对辐板应变在纯滚动时为单频谐波信号,加轨道谱激振后变为调制信号,轮对蛇行时成为几个调制信号的叠加。分析车轮辐板在不同工况下的时频特性,有助于了解辐板应变信号的构成,根据信号的时频特性采用合适的信号处理方法滤除或减小噪声干扰。
The measure of wheel/rail loads has great significance on vehicle dynamics, derailment mechanism, the theory of wheel/rail contact and the vehicle security monitor. Railway transportation is now developing towards high speed, heavy-loading, massive carry and high density. It also demands high precision and speed on the measure of wheel/rail loads. The instrumented wheelset is now the most direct and most precise technology of measuring the wheel/rail loads, which takes the wheel as the detecting sensor of wheel/rail loads and realizes continuous measurement of wheel/rail loads by measuring the strain on several places on the wheel plate. During the running of the vehicle, the wheel/rail loads is non-stationary because of the track irregularity and the plate strain which is produced by the wheel/rail loads is modulated by the wheel rolling angle. The accurance of the instrumented wheelset is influenced by the inequality of the wheel, the technics of sticking the strain gagues, rotate speed of the wheel, and the change of position of the wheel/rail contact point , and the measuring theory and data processing method of the wheel plate strain instrumented wheelset need further study.
The paper looks back on the develop course of instrumented wheelset as well as its developing status domestic and oversea, and analyzes the stress distribution of wheelset plate under the wheel/rail loads, and proposes the measuring method of instrumented wheelset which can eliminates the influence of the wheel rotation, and studies the influence on the measuring precision of the wheel/rail contact point and the data processing method of instrumented wheelset. The detail and achievement is as below:
1、proposing the measuring model of wheel/rail loads Through decades of development, the technology of instrumented wheelset has great use in vehicle dynamics experiment, but still has many problems in the continuous detecting of the wheel/rail loads. Nowadays, the continuous instrumented wheelset technology demands the plate strain produced by the wheel horizontal and vertical force follows the sine and cosine distribution after designing the measuring bridge. As it demands high wheel equality and technics of sticking the strain gages, however, the fact is most regions of the wheelset plates cannot meet the demands, and the high-order harmonic in the electrical bridge output could bring more measuring error. Another problem in instrumented wheelset is that it does not consider the influence on the measuring precision from the change of wheel/rail contact point. During the running of the vehicle, the contacting point between the wheelset and the track is changing near the rolling circle, especially in the pass of curve, switch and serpentine movement in which the contacting point is far away from the rolling circle. Nowadays, almost all the instrumented home dose not takes the change of contact point into consideration. The calibrating test and data processing all have the hypothesis that the contact point is on the rolling circle. So, as a matter of fact, it has great error.
This paper analyzes the distribution rule of the wheelset plate strain influenced by the horizontal and vertical force and bending moment produced by the change of contact point based on the finite element analysis and experiment. Then it adopts several strain gages combination on the same radius to eliminate the high-order harmonic and studies the relationship between the order if harmonic component which needs to be eliminated and the amount of the electrical bridge as well as the phase difference among each electrical bridge which bring theoretical foundation to the improving of the measuring precision of the instrumented wheelset. On such basis, it sets up anon-linear equation sets of wheelset plate strain, wheel/rail loads and contact point.
The paper studies the direct solution of the non-linear equation sets of the measuring model. As there is no analytical solution, so the only solution is to use Newton-Simpson iteration. This method needs to choose an initial value which influence the computing speed and precision and can lead to no convergence, and can has difficulty in error analyzing. This paper reduces the non-linear equation group reasonably, deduces the analytical solution and discusses the method of eliminating invalid solution.
This paper again studies the method of reducing the non-linear equation group to linear equation group. By using principal component analysis, it gets the subspace of plate strain signal when each loads separately influence on it and through projection of measuring signal to each subspace, it sets up the linear equation group. Then we can choose reasonable strain gages' location according to the theory of stability of linear equation group and thus reduces the influence of the error of the strain measuring on the calculation of wheel/rail loads.
As a matter of fact that equality of wheelset and plate gages' location can not meet the demand, this paper proposes new solution. Firstly, it confirm the horizontal and vertical force factor according to each steering angle and contact point location through the calibrating test, and then gets the best wheel rotated angle and contact point through Genetic Algorithm, finally get the horizontal and vertical force factor by checking the table and sets up the linear equation group of plate strain and wheel/rail loads on several radius. This method can effectively reduce the cost of the exploitation of instrumented wheelset as well as the researching time.
2、When wheel/rail contact point location changes, the vertical force can bring a accessory bending moment which bring accessory strain on the wheelset plate and this is always a source of error of instrumented wheelset. Comparing to the strain produced by the horizontal and vertical force, the strain is lower which is produced by accessory bending moment. This paper proposes the detecting method of contact point based on the studying on the distribution rule of plate strain brought by the bending moment.
Although the wheel/rail loads is the non-stationary signal, it is quite stationary in very short timestamp. This paper did the time-frequency analysis on the experiment data of rolling and vibration test rig and finally finds that it is feasible to use the short-time Fourier. When in the pure rolling, the wheelset plate strain is single frequency harmonic signal, and becomes modulation signal when it is added by track spectrum excitation. When in the serpentine movement, the wheelset plate strain is the overlap of several modulation signals. Analyzing the time-frequency characteristic of wheel plate under different conditions help to filter or reduce noise interference through using reasonable method of signal processing according to the time-frequency characteristic.
论文在回顾了测力轮对的发展历程和国内外测力轮对的发展现状的基础上,分析了轮轨载荷作用下轮对辐板的应力分布,提出了消除车轮旋转影响的测力轮对测量方法,研究了轮轨力作用点位置对测量精度的影响和测力轮对数据处理方法,其内容与成果如下:
1、提出轮轨力测量模型
测力轮对技术经过几十年的发展,在车辆动力学试验中得到广泛使用,但在轮轨力连续测量中仍存在很多问题。目前的连续测力轮对技术要求轮轨横向力、垂向力产生的辐板应变组桥后沿圆周方向按正、余弦分布,这不仅对轮对的均匀性和贴片工艺提出了很高的要求,实际上在轮对辐板大部分区域是根本做不到,电桥输出中包含的高次谐波可能带来很大的测量误差。测力轮对存在的另一个问题是没有考虑轮轨作用点位置变化对测量精度的影响。车辆在运行过程中,轮对与轨道的接触点在滚动圆附近不停变化,尤其是在过曲线、道盆和蛇形运动时,接触点位置远离滚动圆。国内现有的测力轮对都没有考虑接触点位置变化,标定实验、数据处理都假设接触点在滚动圆上,测量结果误差较大。
论文在有限元分析和实验基础上,分析了轮对辐板应变在横向力、垂向力和作用点位置变化产生的弯矩作用下的分布规律,采用同一半径上多个应变片组合消除高次谐波,研究了需要消除的谐波分量阶次与电桥的数量、各电桥相位差之间的关系,为提高测力轮对测量精度提供了理论依据。在此基础上,建立轮对辐板应变与轮轨作用力、作用点位置的非线性方程组。
论文研究了测量模型非线性方程组的直接求解方法。非线性方程组一般没有解析解,通常采用迭代数值方法求解。迭代解法需要选择初值,初值不仅影响计算速度,同时影响计算精度,甚至导致计算不收敛,误差分析也比较困难。论文对非线性方程组进行合理简化,推导出方程组的解析解。
论文还研究了将非线性方程组简化为线性方程组的方法。使用主分量分析,得到各载荷单独作用时辐板应变信号的子空间,将测量信号投影到各个子空间从而建立线性方程组。可以利用线性方程组解的稳定性理论选择合理的贴片位置,减少应变测量误差对轮轨力计算值的影响。
针对实际轮对均匀性和贴片位置达不到测力轮对要求的情况,论文提出了新的解决方法。通过标定试验得到转角、作用点位置与对应的横向力、垂向力作用系数的数表,并用神经网络拟合辐板应变和车轮转角间的非线性关系,采用遗传算法获得最佳的作用点位置和电桥直流偏置,对每个电桥建立辐板应变与轮轨力间的线性方程组。这种方法可以有效地降低测力轮对的开发成本和研制时间。
2、提出轮轨作用点位置在线测量方法
轮轨作用点位置发生变化时,垂向力会产生一个附加弯矩,在轮对辐板产生附加应变,这一直被作为测力轮对的一个误差源对待。与横向力、垂向力在辐板上产生的应变相比,附加弯矩产生的应变比较微弱,论文提出的轮轨力测量方案,不仅提高了横向力和垂向力的测量精度,而且可以检测轮轨作用点位置。
3、分析车轮辐板应变的时频特性
轮轨力虽然是非平稳信号,但在很短的时间内可以看成平稳的。论文对滚动振动试验台实验数据进行了时频分析,结果表明用短时傅立叶分析是可行的。轮对辐板应变在纯滚动时为单频谐波信号,加轨道谱激振后变为调制信号,轮对蛇行时成为几个调制信号的叠加。分析车轮辐板在不同工况下的时频特性,有助于了解辐板应变信号的构成,根据信号的时频特性采用合适的信号处理方法滤除或减小噪声干扰。
The measure of wheel/rail loads has great significance on vehicle dynamics, derailment mechanism, the theory of wheel/rail contact and the vehicle security monitor. Railway transportation is now developing towards high speed, heavy-loading, massive carry and high density. It also demands high precision and speed on the measure of wheel/rail loads. The instrumented wheelset is now the most direct and most precise technology of measuring the wheel/rail loads, which takes the wheel as the detecting sensor of wheel/rail loads and realizes continuous measurement of wheel/rail loads by measuring the strain on several places on the wheel plate. During the running of the vehicle, the wheel/rail loads is non-stationary because of the track irregularity and the plate strain which is produced by the wheel/rail loads is modulated by the wheel rolling angle. The accurance of the instrumented wheelset is influenced by the inequality of the wheel, the technics of sticking the strain gagues, rotate speed of the wheel, and the change of position of the wheel/rail contact point , and the measuring theory and data processing method of the wheel plate strain instrumented wheelset need further study.
The paper looks back on the develop course of instrumented wheelset as well as its developing status domestic and oversea, and analyzes the stress distribution of wheelset plate under the wheel/rail loads, and proposes the measuring method of instrumented wheelset which can eliminates the influence of the wheel rotation, and studies the influence on the measuring precision of the wheel/rail contact point and the data processing method of instrumented wheelset. The detail and achievement is as below:
1、proposing the measuring model of wheel/rail loads Through decades of development, the technology of instrumented wheelset has great use in vehicle dynamics experiment, but still has many problems in the continuous detecting of the wheel/rail loads. Nowadays, the continuous instrumented wheelset technology demands the plate strain produced by the wheel horizontal and vertical force follows the sine and cosine distribution after designing the measuring bridge. As it demands high wheel equality and technics of sticking the strain gages, however, the fact is most regions of the wheelset plates cannot meet the demands, and the high-order harmonic in the electrical bridge output could bring more measuring error. Another problem in instrumented wheelset is that it does not consider the influence on the measuring precision from the change of wheel/rail contact point. During the running of the vehicle, the contacting point between the wheelset and the track is changing near the rolling circle, especially in the pass of curve, switch and serpentine movement in which the contacting point is far away from the rolling circle. Nowadays, almost all the instrumented home dose not takes the change of contact point into consideration. The calibrating test and data processing all have the hypothesis that the contact point is on the rolling circle. So, as a matter of fact, it has great error.
This paper analyzes the distribution rule of the wheelset plate strain influenced by the horizontal and vertical force and bending moment produced by the change of contact point based on the finite element analysis and experiment. Then it adopts several strain gages combination on the same radius to eliminate the high-order harmonic and studies the relationship between the order if harmonic component which needs to be eliminated and the amount of the electrical bridge as well as the phase difference among each electrical bridge which bring theoretical foundation to the improving of the measuring precision of the instrumented wheelset. On such basis, it sets up anon-linear equation sets of wheelset plate strain, wheel/rail loads and contact point.
The paper studies the direct solution of the non-linear equation sets of the measuring model. As there is no analytical solution, so the only solution is to use Newton-Simpson iteration. This method needs to choose an initial value which influence the computing speed and precision and can lead to no convergence, and can has difficulty in error analyzing. This paper reduces the non-linear equation group reasonably, deduces the analytical solution and discusses the method of eliminating invalid solution.
This paper again studies the method of reducing the non-linear equation group to linear equation group. By using principal component analysis, it gets the subspace of plate strain signal when each loads separately influence on it and through projection of measuring signal to each subspace, it sets up the linear equation group. Then we can choose reasonable strain gages' location according to the theory of stability of linear equation group and thus reduces the influence of the error of the strain measuring on the calculation of wheel/rail loads.
As a matter of fact that equality of wheelset and plate gages' location can not meet the demand, this paper proposes new solution. Firstly, it confirm the horizontal and vertical force factor according to each steering angle and contact point location through the calibrating test, and then gets the best wheel rotated angle and contact point through Genetic Algorithm, finally get the horizontal and vertical force factor by checking the table and sets up the linear equation group of plate strain and wheel/rail loads on several radius. This method can effectively reduce the cost of the exploitation of instrumented wheelset as well as the researching time.
2、When wheel/rail contact point location changes, the vertical force can bring a accessory bending moment which bring accessory strain on the wheelset plate and this is always a source of error of instrumented wheelset. Comparing to the strain produced by the horizontal and vertical force, the strain is lower which is produced by accessory bending moment. This paper proposes the detecting method of contact point based on the studying on the distribution rule of plate strain brought by the bending moment.
Although the wheel/rail loads is the non-stationary signal, it is quite stationary in very short timestamp. This paper did the time-frequency analysis on the experiment data of rolling and vibration test rig and finally finds that it is feasible to use the short-time Fourier. When in the pure rolling, the wheelset plate strain is single frequency harmonic signal, and becomes modulation signal when it is added by track spectrum excitation. When in the serpentine movement, the wheelset plate strain is the overlap of several modulation signals. Analyzing the time-frequency characteristic of wheel plate under different conditions help to filter or reduce noise interference through using reasonable method of signal processing according to the time-frequency characteristic.
引文
[1]张卫华.空间状态轮轮(轨)接触点计算方法[J].中国铁道科学,2006,27(4):76-79.
[2]曾庆元等.列车脱轨分析理论与应用[M].长沙:中南大学出版社,2006,12:38-73.
[3]俞展猷等.车轮脱轨及其评价[J].铁道学报,1999,21(3):33-38.
[4]张卫华等.车辆运动稳定性试验台试验及与线路试验间误差分析[J].铁道机车车辆,1996,19(4):69-72.
[5]张卫华等.单轮对脱轨实验及其理论分析[J].铁道学报,1998,20(1):39-44.
[6]杨国桢.铁道车辆脱轨安全性的设计(上).铁道车辆,1994,(11):8-27.
[7]杨国桢.铁道车辆脱轨安全性的设计(下).铁道车辆,1994,(12):12-32.
[8]向俊,周智辉.列车脱轨的研究的最新进展[J].铁道科学与工程学报,2005,5(2):1-8.
[9]曾庆元等,列车脱轨的力学机理与防治脱轨的理论[J].铁道科学与工程学报,2004(11):19-31.
[10]俞展猷等.防脱轨性能的评定[J].机车电传动,1999,(5):1-4.
[11]曾宇清.车辆脱轨安全评判的动态限度[J].中国铁道科学,1999,20(4):71-76.
[12]宫本吕幸.铁道车辆的脱轨机制[J].国外铁道车辆,1997,6:36-43.
[13]石田弘明.铁路车辆安全性评定标准[J].国外铁道车辆,1996,5:27-34.
[14]翟婉明,陈果.根据车轮抬升量评判车辆脱轨的方法与准则[J].铁道学报,2001,23(2):17-26.
[15]Weber H.H.Zur Ermmilung der kraftor Zwischen Rad und Schiene[J].Eigenerlag des Institues fur Flugzeugustatick und Leichtbau der ETH Zurich,1968.
[16]A.RonardPocklington.The B.R.Load Measuring Wheel[J].proceedingds of international conference on Wheel/rail load and Displacement Measurement Techniques,1981.
[17]H.Kanehara.Measuring technique of dynamic characteristics of railway vehicles,Japan National Railways,1977(121):113-114.
[18]H.Kanehara.Study on online measurement of longitudinal creep force of railway vehicle[J].Proceedinds of the conference J-Rail' 98.
[19]Modransky.Instrumented Locomotive Wheels for Continuous Measurements of Vertical and Lateral Loads[J].ASME Publication,Paper 79-RT-8,Feb.1980.
[20]Kesler J K.Measurement techniques for onboard wheel/rail loads[J]. Proceeeds of International Conference on Wheel/Rail Load and Displacement Techniques,1981:19-20.
[21]Johnson M R,Development and Use of an instrumented wheelset for the measurement of wheel/rail interaction forces[J].Proceedings of international conference on wheel/rail load and displacement measurement techniques,1981,1:19-20.
[22]Olson P E,Johnson S.Lateral forces between wheels and rail,an experimental investigation[J].ASME Publication.Paper No.79-RT-8,1980.
[23]Anderson,T.J.Calibration guidelines and equipments,important characteristics and error types for instrumented wheelsets[J]Proceedings of the International Conference on Wheel/Rail and Displacement Measurement Techniques,1981(6):1-26.
[24]Bakken,G.B.Development of a Wheel/Rail Measurement System-From Concept to Implemententation[J].Proceedings of the International Conference on Wheel/Rail and Displacement Measurement Techniques,1981(2):1-30.
[25]Boyd,P.Measurement Techniques for Onboard Wheel/Rail Loads[J].Proceedings of the International Conference on Wheel/Rail and Displacement Measurement Techniques,1981(1):1-36.
[26]Pocklinton,A.R.Improved Data from Load-Measuring Wheels[J].Railway Engineer,1977,2(4):37-43.
[27]Swenson,C.A.Development and Use of Instrumented Locomotive Wheelsets[J].Proceedings of the International Conference on Wheel/Rail and Displacement Measurement Techniques,1981(5):1-10.
[28]Zuck,H.H.Determination of Wheel/Rail Forces by Forces by Means of Measuring Wheelsets on Deutsche Bundesbahn[J].Proceedings of the International Conference on Wheel/Rail and Displacement Measurement Techniques,1981(1):1-22.
[29]李富达等.机车与线路间垂向力和横向力的测试[J].中国铁道科学.1986,7(1):92-102.
[30]刘尚举等.提高测力轮对测量精度的研究[J].铁道车辆,1998,36(2):32-34.
[31]吴铎等.测力轮对轮/轨力数据处理迭代方法[J],中国铁道科学,1995,16(3):49-55.
[32]曾宇清等.测力轮对连续测量的理论与实践[J].铁道学报,1998,20(6):28-34.
[33]张立民,金学松.基于测力轮对的轮轨瞬态作用力仿真计算[J].铁道学报,1998,20(6):28-134.
[34]Hiroaki Ishida.轮轨接触力新的连续测量法[J].国外铁道车辆,1998,2:19-24
[35]Otter D E.A Design for Next Generation Load Measuring Wheelsets[J].IEEE/ASME,May 1991.
[36]Higgns R L.High Accuracy Load Measuring Wheelset.1992.
[37]Hiromichi Kanehara.Measuring rail/wheel contact points of running railway vehicles[J].Wear,253(2002):275-283.
[38]陈成元等.连续测量轮轨力的测力轮对技术[J].西南交通大学学报,1994,29(2):263-267.
[39]张绍华等.辐板式测力轮对的研制[J],铁道学报,1992,14(2):103-109.
[40]徐春明,姜海波,于建江.一种新的加权主分量分析方法[J].佳木斯大学学报,2007,25(6):821-822.
[41]胡煜.主分量分析方法和二次判别分析方法应用于基因芯片数据分析[J].广东技术师范学院学报,2007,10:25-27.
[42]王展青,刘小双.基于PCA与ICA的人脸识别算法研究[J].华中师范大学学报,2007,41(3):373-376.
[43]杨红,周伟.主成分分析法在土壤腐蚀性评价中的应用[J].油气储运 2006,26(10):43:46.
[44]刘直芳,游志胜,王运琼.基于PCA和ICA的人脸识别[J].激光技术,2004,28(1):78-81.
[45]杨竹青,李勇,胡德文.独立成分分析综述[J].自动化学报,2002,28(5):762-772.
[46]江泓.一种基于K-L变换和支持向量机的图像分割算法[J].电子测量技术,2006.29(5):81-83.
[47]Erwin E,Obermeyer K,Schulten K.Self-organizing maps:ordering,convergence properties and energy functions[J].Biol Cybern,1992,67(1):47-55.
[48]Cichocki A.Neural Networks for Optimization and Signal Processing[J].New York:John Wiley & Sons Inc,1993:38-84.
[49]Huang N E,et al.The empirical mode decomposition and t he Hilbert spect rum for nonlinear and nonstationary time series nalysis[J].Proceedings of the Royal Society,1998,903-995.
[50]Kwok H K,Jones D L.Improved Instantaneous Frequency Estimation Using an Adaptive Short-time Fourier Transform[J].IEEE Trans.on SP,2000,48(10):2964.
[51]Mallat S,Zhang Z.Matching Pursuits with Time-frequency Dictionaries[J].IEEE Trans.Signal Processing,1993,41(12):3397.
[52]Mann S,Haykin S.The Chirplet Transform-A Generalization of Gabors Logon Transform[J].Vision Interface'91,1991,3:6.
[53]Mann S,Haykin S.Chirplets and Warblers:Novel Time-frequency Representations.Electron.Lett.1992,28(2):114.
[54]Martin J G,Rasheed K.Using Singular Value Decomposition to Improve a Genetic Algorithm's Performance[J].The Congresson Evolutionary Computation,2003,3(8-12):612-617.
[55]Kirby M,Sirovich Application of Karhuner-Loeve Procedure For the Characterization of Human Faces[J]. IEEE Trans. 1990, 12(1):103-108.
[56] Algorithm for Time-varying Fourier[J]. Transform IEEE. Trans. SP. , 1993, 41(7).
[57] Yeung, K. Y. and Ruzzo, W. L. Principal compo-nent analysis for clustering gene expression data. 2001, 17(9):763-774.
[58] Erikssons J, Koivunen V. Complex random vectors and ICA models: Identifiability, uniqueness, and separability [J]. IEEE Trans on Information Theory, 2006, 52 (3):1017-1029.
[59] Bell A,Sejnowski T. An information maximaization approach to blind separation and blind deconvolution[J]. Neural Computation, 1995, 7:1129-1159.
[60] Weldon T P, Higgins W E. An algorithm for designing multiple gabor filters for segmenting multi-textured images [C]. In: Proc. Of IEEE Int'l. Conf. of ImageProc. , Chicago, IL, 1998,10.
[61] GirgisAA, PetersonWL. AdaPtive estintion of Powersystem frequency deviation and its rate change for calculating sudden powersystem overload[J]. IEEE Transactions on Power Delivery,1990,5(4):21-25.
[62] David W. P. Thomas, Malecolm. Evaluation of Frequency tracting methods [J]. IEEE Transctions on Power Delivery, 2001, 16(3):367-371.
[63] Taolin, Minel Tsuji,Eiji Yamada. A wavelet approach to real time estimation of power system frequeney[C], IEEE Digital object Identifier. 2001:58-65.
[64] T. S. Sidhu. Accurate measurement of Power system frequency using a digital signal Processing teehnique[J]. IEEE Transactions on Instrumenttation and Measumentl, 1999,48(1):75-80.
[65] Gusev V G, Yanover B I. The complex representation of a discrete signal on the basis of its Hilbert transform. Radio. Eng. and Electron Phys. , 1983, 28(1):83-89.
[66] Leung Y, Guo Y. Degree of Population Diversity—A Perspective on Premature Convergence in Genetic Algorithms and its Markov Chain Analysis[J]. IEEE Trans. On Neural Networks. 1997, 8 (5):1165-1176.
[67] Srinivas M, Patnaik L M. Adaptive Probabilities of Crossover and Mutation in Genetic Algorithms[J]. IEEE Trans on System. 1994, 24 (4):656-665.
[68] R. B RANDLL, LUO DEYANG , Hilber Transform Techniques for Torsional Vibration Anallysis[C]. Vibration and Noiseconference, 1990:15-25
[69] Mallat S,Hwang W L. Singularity detection and processing with wavelets[J]. IEEE Trans. on Information Theory, 1992 (2):617-643.
[70] Hammond J, White P. The analysis of non-stationary signals using time-frequency methods[J]. Journal of Sound and Vibration, 1996, 190:419-447.
[71]Huang N E,Shen Z,Long S R,et al.The empirical mode decomposition and the Hilbert spectrum for nonlinear and nonstationary time series analysis[J].Proceedings of the Royal Society of London,Series A,1998,454:903-995.
[72]Olhede S,Walden A T.The Hilbert spectrum via wavelet projections[J].Proceedings of the Royal Society of London,Series A,2004,460:955-975.
[73]Olhede S,Walden A T.A generalized demodulation approach to time-frequency projections for multicomponent signals[J].Proceedings of the Royal Society of London,Series A,2005,461:2159-2179.
[74]Nielsen M.On the construction and frequency localization of finite orthogonal quadrature filters[J].Journal of Approximation Theory,2001,108:36-52.
[75]Fort J C.Solving a combinatorial problem via self-organizing process:an application of the Kohonen algorithm to the travelling salesman problem[J].Biol Cybern,1988,59(1):33-40.
[76]杨培杰,印兴耀,张广智.希尔伯特-黄变换地震信号时频分析与属性提取[J].地球物理学进展,2007,22(5):1585-1590.
[77]邹红星等.时频分析:回溯与前瞻[J].电子学报,2000,28(9):78-84.
[78]刘庆云,李志舜,刘朝晖.时频分析技术及研究现状[J].计算机工程,2004,30(1):171-173.
[79]金雷,李月,杨宝俊.用时频峰值滤波方法消减地震勘探资料中随机噪声的初步研究[J].地球物理学进展,2005,20(3):724-728.
[80]彭勇,周轼尘,希尔伯特变换及解调方法的应用[J].武汉交通科技大学学报,1996,20(5):544-548.
[81]黄迪山,陆乃炎.应用希尔伯特变换提取相位信号[J].振动与动态测试,1988:68-71.
[82]党志敏,贺振华,黄德济.广义希尔伯特变换在含噪信号边缘检测中的应用效果分析[J].油气地球物 2007,5(2):25-27.
[83]苏峰,曲毅,陈军.线性时频分析在弱信号检测中的应用[J].广西民族学院学报2004,10(3):63-65.
[84]周丹 蔡坤宝.基于短时傅立叶变换的脉象信号的模式识别方法[J].重庆科技学院学报,2007,9(3):49-52.
[85]王胜文,陈伏兵.一种基于奇异值分解的特征抽取方法[J].电子与信息学报,2005,27(2):294-297.
[86]何正嘉,警艳阳.基于Hermtiina小波的信号奇异性识别[J].工程数学学报,2001,18(5):37-43.
[87]盖强,张海勇.Hilbert-Huang变换的自适应频率多分辨率分析研究[J].电子学报,2005,33(3):563-566.
[88]陈章位,潘宏.渐进信号瞬态频率的提取[J].Journal of vibration Engmeering,1997,10(4):451-457.
[89]庞浩,李东霞等.基于Hiibert移相滤波的全数字锁相环[J].电网技术,2003,27(11):55-59.
[90]何仁贵,黄登山.一种改进的正弦波频率谱估计方法[J].西安工业大学学报,2004.24(4):328-331.
[91]李钢虎等.Winger-HOugh变换在水下目标信号检测中的应用[J].兵工学报,2006,27(1):121-125.
[92]严光洪,束滋德.包络分析法在某航行体振动信号处理中的应用[J].振动、测试与诊断,2000,20(3):202-205.
[93]丁康等.复解析带通滤波器及其在解调分析中的应用[J].振动工程学报,2000,13(3):385-391.
[94]何岭松,杨叔子.包络检波的数字滤波算法[J].振动工程学报,1997,10(3):362-367.
[95]张建勋 俞智昆.希尔伯特变换原理在扭转振动分析中应用的理论研究[J].昆明理工大学学报,1998,23(1):90-94.
[96]黄迪山,陆乃炎.应用希尔伯特变换提取相位信号[J].振动与动态测试,1988:68-71.
[97]李生红,王树勋.利用希尔伯特变换进行谐波分析[J].电子学报,1997,25(1):98-101.
[98]王兆旗,丘斌煌.希尔伯特变换子波反褶积及其改进方案[J].山西建筑,2007,33(7):117-118.
[99]徐伯勋,林贵标.希尔伯特变换子波反褶积[J].石油物探,1987,26(1):84-85
[110]饶建中.子波反褶积程序[J].物探化探计算技术,1993,15(2):77-78.
[101]刘雯林.子波处理方法的研究[J].石油地球物理勘探,1985,20(2):64-65.
[102]周洁玲,成世琦.反褶积与信噪比的关系研究[J].西安石油学院学报,2002,17(4):21-22.
[103]张南平,黄延禾.低信噪比地震资料的子波处理方法[J].石油物探,1989,28(1):90-91.
[104]赵学智,符红羽.小波分析住信号包络提取中的应用[J].广东工业大学学报,2000.17(3):36-39.
[105]冯志鹏,李学军.基于平稳小波包分解的水轮机非平稳振动信号希尔伯特谱分析[J].中国电机工程学报,2006,26(12):79-84.
[106]彭志科,何永勇等.用小波时频分析方法研究发电机碰摩故障特征[J].中国电机工程学报,2003,23(5):75-79.
[107]杨世锡,胡劲松等.旋转机械振动信号基于EMD的希尔伯特变换和小波变换时频分析比较[J].中国电机工程学报,2003,23(6):102-107.
[108]张绪省等.信号包络提取-从希尔伯特变换到小波变换[J].电子科学学刊,1997,19(1):120-123.
[109]皮骏等.基于小波原理的振动信号包络提取法[J].中国民航学院学报,2005,23(2):12-15.
[110]王汉平等.基于遗传算法进化神经网络的潜射导弹筒盖压力预测[J].北京理 工大学学报,2006,26(1):23-26.
[111]郑志军,郑守淇.用基于实数编码的自适应遗传算法进化神经网络[J].计算机工程与应用,2000,15(9):36-37.
[112]袁慧梅.具有自适应交换率和变异率的遗传算法[J].首都师范大学学报,2000,21(3):14-20.
[113]陈明.基于进化遗传算法的优化计算[J].软件学报,1998,9(11):876-879.
[114]辛长欣等.BP神经网络在回归分析中的应用[J].西安工业学院学报,2002,22(2):129-135.
[115]雷英杰,张善文,李续武等.MATLAB遗传算法工具箱及应用[M].西安:西安电子科技大学出版社.2004.
[116]L.科恩.时-频分析:理论与应用[M].西安交通大学出版社,1998.
[117]张贤达,保铮.非平稳信号分析与处理[M].北京:国防工业出版社,2001.
[118]胡广月.数字信号处理-理论、算法与实现[M].清华大学出版社,1996.
[119]阎平凡,张长水.人工神经网络与模拟进化计算[M].北京:清华大学出版社,2001.
[120]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
[121]黄绪德.反褶积与地震道反演[M].北京:石油工业出版社,1992.
[122]秦前清,杨宗凯.实用小波分析[M].西安:西安电子科技大学出版社.1995.
[123]Thiran P,Hasler M.Self-organisation of a one dimensional Kohonen network with quantized weights and inputs[J].Neural Networks,1994,7(9):1427-1439.
[124]邵玉萍等.BP神经网络在测力轮对横垂向桥解耦中的应用研究[J],中国测试技术.2003,5(10):17-19.
[125]赵华敏,陈开周.用神经网络解非线性方程组[J].西安电子科技大学学报,2001,28(1):35-38.
[126]焦李成.神经网络系统理论[J].西安:西安电子科技大学出版社,1996:1-25.
[127]李春宇等.遗传算法在微弱GPS信号捕获方法中的应用[J].航空学报,2007,28(6):1433-1437.
[128]杨慧娟,曲喜强.基于小波变换的声波信号包络提取[J].华北工学院学报,2004,25(4):300-302.
[129]梅璐璐,林京.基于相移小波的信号包络提取方法研究[J].仪器仪表学报,2007,28(6):1122-1128.
[130]李智,陈祥初.基于小波分析的解调分析方法[J].华北工学院测试技术学报,1999,13(3):165-169.
[131]丁康,胡晓斌,谢明.解调分析的一种优化方法[J].振动工程学报,1998,11(2):235-240.
[132]Walden A T,Contreras Cristan A.The phase-corrected undecimated discrete wavelet packet transform and its application to interpreting the timing of events[J].Proceedings of the Royal Society of London,Series A,1998,454:2243-2266.
[133]刘彬,蒋金水.轧机转速波动测量的扭振监测实验研究[J].计量学报,2007,28(3):272-275.
[134]丁康等.解调分析在故障诊断中应用的局限性研究[J].振动工程学报,1997,10(1):235-240.
[135]丁康等.传动箱齿轮和轴故障的振动诊断方法的研究[J].振动与冲击,1994,10(2):235-240.
[136]王志杰,丁康.汽车变速箱故障振动诊断的试验研究[J].汽车工程,1994,10(4):242-248.
[137]丁康,胡晓斌,谢明.解调分析的一种优化方法[J].振动工程学报,1998,11(2):235-240.
[138]马晓建等.机械故障诊断中常用解调方法比较及应用[J].东华大学学报,2001,27(5):105-108.
[139]朱利民.齿轮故障振动信号检波解调分析中的混频效应[J].振动工程学报,1999,12(3):332-335.
[140]马晓建等.碟式分离机横轴轴承的故障分析与诊断[J].化工机械,1999,26(6):320-325.
[141]金少先.齿轮振动的边带分布特征与故障诊断实例[J].振动工程学报,1999,12(3):429-432.
[142]马孝江,王凤利等.局域波时频分布在转子系统早期故障诊断中的应用研究[J].中国电机工程学报,2004,24(3):161-164.
[143]冯志鹏,褚福磊.基于Hilbert-Huang变换的水轮机非平稳压力脉动信号分析[J].中国电机工程学报,2005,25(10):111-115.
[144]赵文生,熊诗波.希尔伯特变换在扭振测量中的应用[J].太原理工大学学报,2006,37(2):187-189.
[145]张建勋,朱景春.振动信号处理技术在旋转机械振动监测与诊断中的应用和动向[J].昆明理工大学学报,2000,25(1):77-79
[146]祝海龙,屈梁生.自组织包络解调及其在轴承诊断中的应用[J].西安交通大学学报,2000,34(7):77-81.
[147]田涛等.希尔伯特变换及其在故障诊断中的应用[J].振动与冲击,1995:15(2):23-27
[148]王延春等.包络分析方法及其在齿轮故障诊断振动诊断中的应用[J].重庆大学学报,1995,18(1):87-91.
[149]马晓建,陈瑞琪,吴文英等.机械故障诊断中常用解调方法的比较及应用[J].东华大学学报,2001,27(5):105-108.
[150]徐春林,江志农,肖英.基于Hilbert变换的包络分析及其在机械故障诊断中的应用[J].机电工程技术,2004,33(7):114-115.
[2]曾庆元等.列车脱轨分析理论与应用[M].长沙:中南大学出版社,2006,12:38-73.
[3]俞展猷等.车轮脱轨及其评价[J].铁道学报,1999,21(3):33-38.
[4]张卫华等.车辆运动稳定性试验台试验及与线路试验间误差分析[J].铁道机车车辆,1996,19(4):69-72.
[5]张卫华等.单轮对脱轨实验及其理论分析[J].铁道学报,1998,20(1):39-44.
[6]杨国桢.铁道车辆脱轨安全性的设计(上).铁道车辆,1994,(11):8-27.
[7]杨国桢.铁道车辆脱轨安全性的设计(下).铁道车辆,1994,(12):12-32.
[8]向俊,周智辉.列车脱轨的研究的最新进展[J].铁道科学与工程学报,2005,5(2):1-8.
[9]曾庆元等,列车脱轨的力学机理与防治脱轨的理论[J].铁道科学与工程学报,2004(11):19-31.
[10]俞展猷等.防脱轨性能的评定[J].机车电传动,1999,(5):1-4.
[11]曾宇清.车辆脱轨安全评判的动态限度[J].中国铁道科学,1999,20(4):71-76.
[12]宫本吕幸.铁道车辆的脱轨机制[J].国外铁道车辆,1997,6:36-43.
[13]石田弘明.铁路车辆安全性评定标准[J].国外铁道车辆,1996,5:27-34.
[14]翟婉明,陈果.根据车轮抬升量评判车辆脱轨的方法与准则[J].铁道学报,2001,23(2):17-26.
[15]Weber H.H.Zur Ermmilung der kraftor Zwischen Rad und Schiene[J].Eigenerlag des Institues fur Flugzeugustatick und Leichtbau der ETH Zurich,1968.
[16]A.RonardPocklington.The B.R.Load Measuring Wheel[J].proceedingds of international conference on Wheel/rail load and Displacement Measurement Techniques,1981.
[17]H.Kanehara.Measuring technique of dynamic characteristics of railway vehicles,Japan National Railways,1977(121):113-114.
[18]H.Kanehara.Study on online measurement of longitudinal creep force of railway vehicle[J].Proceedinds of the conference J-Rail' 98.
[19]Modransky.Instrumented Locomotive Wheels for Continuous Measurements of Vertical and Lateral Loads[J].ASME Publication,Paper 79-RT-8,Feb.1980.
[20]Kesler J K.Measurement techniques for onboard wheel/rail loads[J]. Proceeeds of International Conference on Wheel/Rail Load and Displacement Techniques,1981:19-20.
[21]Johnson M R,Development and Use of an instrumented wheelset for the measurement of wheel/rail interaction forces[J].Proceedings of international conference on wheel/rail load and displacement measurement techniques,1981,1:19-20.
[22]Olson P E,Johnson S.Lateral forces between wheels and rail,an experimental investigation[J].ASME Publication.Paper No.79-RT-8,1980.
[23]Anderson,T.J.Calibration guidelines and equipments,important characteristics and error types for instrumented wheelsets[J]Proceedings of the International Conference on Wheel/Rail and Displacement Measurement Techniques,1981(6):1-26.
[24]Bakken,G.B.Development of a Wheel/Rail Measurement System-From Concept to Implemententation[J].Proceedings of the International Conference on Wheel/Rail and Displacement Measurement Techniques,1981(2):1-30.
[25]Boyd,P.Measurement Techniques for Onboard Wheel/Rail Loads[J].Proceedings of the International Conference on Wheel/Rail and Displacement Measurement Techniques,1981(1):1-36.
[26]Pocklinton,A.R.Improved Data from Load-Measuring Wheels[J].Railway Engineer,1977,2(4):37-43.
[27]Swenson,C.A.Development and Use of Instrumented Locomotive Wheelsets[J].Proceedings of the International Conference on Wheel/Rail and Displacement Measurement Techniques,1981(5):1-10.
[28]Zuck,H.H.Determination of Wheel/Rail Forces by Forces by Means of Measuring Wheelsets on Deutsche Bundesbahn[J].Proceedings of the International Conference on Wheel/Rail and Displacement Measurement Techniques,1981(1):1-22.
[29]李富达等.机车与线路间垂向力和横向力的测试[J].中国铁道科学.1986,7(1):92-102.
[30]刘尚举等.提高测力轮对测量精度的研究[J].铁道车辆,1998,36(2):32-34.
[31]吴铎等.测力轮对轮/轨力数据处理迭代方法[J],中国铁道科学,1995,16(3):49-55.
[32]曾宇清等.测力轮对连续测量的理论与实践[J].铁道学报,1998,20(6):28-34.
[33]张立民,金学松.基于测力轮对的轮轨瞬态作用力仿真计算[J].铁道学报,1998,20(6):28-134.
[34]Hiroaki Ishida.轮轨接触力新的连续测量法[J].国外铁道车辆,1998,2:19-24
[35]Otter D E.A Design for Next Generation Load Measuring Wheelsets[J].IEEE/ASME,May 1991.
[36]Higgns R L.High Accuracy Load Measuring Wheelset.1992.
[37]Hiromichi Kanehara.Measuring rail/wheel contact points of running railway vehicles[J].Wear,253(2002):275-283.
[38]陈成元等.连续测量轮轨力的测力轮对技术[J].西南交通大学学报,1994,29(2):263-267.
[39]张绍华等.辐板式测力轮对的研制[J],铁道学报,1992,14(2):103-109.
[40]徐春明,姜海波,于建江.一种新的加权主分量分析方法[J].佳木斯大学学报,2007,25(6):821-822.
[41]胡煜.主分量分析方法和二次判别分析方法应用于基因芯片数据分析[J].广东技术师范学院学报,2007,10:25-27.
[42]王展青,刘小双.基于PCA与ICA的人脸识别算法研究[J].华中师范大学学报,2007,41(3):373-376.
[43]杨红,周伟.主成分分析法在土壤腐蚀性评价中的应用[J].油气储运 2006,26(10):43:46.
[44]刘直芳,游志胜,王运琼.基于PCA和ICA的人脸识别[J].激光技术,2004,28(1):78-81.
[45]杨竹青,李勇,胡德文.独立成分分析综述[J].自动化学报,2002,28(5):762-772.
[46]江泓.一种基于K-L变换和支持向量机的图像分割算法[J].电子测量技术,2006.29(5):81-83.
[47]Erwin E,Obermeyer K,Schulten K.Self-organizing maps:ordering,convergence properties and energy functions[J].Biol Cybern,1992,67(1):47-55.
[48]Cichocki A.Neural Networks for Optimization and Signal Processing[J].New York:John Wiley & Sons Inc,1993:38-84.
[49]Huang N E,et al.The empirical mode decomposition and t he Hilbert spect rum for nonlinear and nonstationary time series nalysis[J].Proceedings of the Royal Society,1998,903-995.
[50]Kwok H K,Jones D L.Improved Instantaneous Frequency Estimation Using an Adaptive Short-time Fourier Transform[J].IEEE Trans.on SP,2000,48(10):2964.
[51]Mallat S,Zhang Z.Matching Pursuits with Time-frequency Dictionaries[J].IEEE Trans.Signal Processing,1993,41(12):3397.
[52]Mann S,Haykin S.The Chirplet Transform-A Generalization of Gabors Logon Transform[J].Vision Interface'91,1991,3:6.
[53]Mann S,Haykin S.Chirplets and Warblers:Novel Time-frequency Representations.Electron.Lett.1992,28(2):114.
[54]Martin J G,Rasheed K.Using Singular Value Decomposition to Improve a Genetic Algorithm's Performance[J].The Congresson Evolutionary Computation,2003,3(8-12):612-617.
[55]Kirby M,Sirovich Application of Karhuner-Loeve Procedure For the Characterization of Human Faces[J]. IEEE Trans. 1990, 12(1):103-108.
[56] Algorithm for Time-varying Fourier[J]. Transform IEEE. Trans. SP. , 1993, 41(7).
[57] Yeung, K. Y. and Ruzzo, W. L. Principal compo-nent analysis for clustering gene expression data. 2001, 17(9):763-774.
[58] Erikssons J, Koivunen V. Complex random vectors and ICA models: Identifiability, uniqueness, and separability [J]. IEEE Trans on Information Theory, 2006, 52 (3):1017-1029.
[59] Bell A,Sejnowski T. An information maximaization approach to blind separation and blind deconvolution[J]. Neural Computation, 1995, 7:1129-1159.
[60] Weldon T P, Higgins W E. An algorithm for designing multiple gabor filters for segmenting multi-textured images [C]. In: Proc. Of IEEE Int'l. Conf. of ImageProc. , Chicago, IL, 1998,10.
[61] GirgisAA, PetersonWL. AdaPtive estintion of Powersystem frequency deviation and its rate change for calculating sudden powersystem overload[J]. IEEE Transactions on Power Delivery,1990,5(4):21-25.
[62] David W. P. Thomas, Malecolm. Evaluation of Frequency tracting methods [J]. IEEE Transctions on Power Delivery, 2001, 16(3):367-371.
[63] Taolin, Minel Tsuji,Eiji Yamada. A wavelet approach to real time estimation of power system frequeney[C], IEEE Digital object Identifier. 2001:58-65.
[64] T. S. Sidhu. Accurate measurement of Power system frequency using a digital signal Processing teehnique[J]. IEEE Transactions on Instrumenttation and Measumentl, 1999,48(1):75-80.
[65] Gusev V G, Yanover B I. The complex representation of a discrete signal on the basis of its Hilbert transform. Radio. Eng. and Electron Phys. , 1983, 28(1):83-89.
[66] Leung Y, Guo Y. Degree of Population Diversity—A Perspective on Premature Convergence in Genetic Algorithms and its Markov Chain Analysis[J]. IEEE Trans. On Neural Networks. 1997, 8 (5):1165-1176.
[67] Srinivas M, Patnaik L M. Adaptive Probabilities of Crossover and Mutation in Genetic Algorithms[J]. IEEE Trans on System. 1994, 24 (4):656-665.
[68] R. B RANDLL, LUO DEYANG , Hilber Transform Techniques for Torsional Vibration Anallysis[C]. Vibration and Noiseconference, 1990:15-25
[69] Mallat S,Hwang W L. Singularity detection and processing with wavelets[J]. IEEE Trans. on Information Theory, 1992 (2):617-643.
[70] Hammond J, White P. The analysis of non-stationary signals using time-frequency methods[J]. Journal of Sound and Vibration, 1996, 190:419-447.
[71]Huang N E,Shen Z,Long S R,et al.The empirical mode decomposition and the Hilbert spectrum for nonlinear and nonstationary time series analysis[J].Proceedings of the Royal Society of London,Series A,1998,454:903-995.
[72]Olhede S,Walden A T.The Hilbert spectrum via wavelet projections[J].Proceedings of the Royal Society of London,Series A,2004,460:955-975.
[73]Olhede S,Walden A T.A generalized demodulation approach to time-frequency projections for multicomponent signals[J].Proceedings of the Royal Society of London,Series A,2005,461:2159-2179.
[74]Nielsen M.On the construction and frequency localization of finite orthogonal quadrature filters[J].Journal of Approximation Theory,2001,108:36-52.
[75]Fort J C.Solving a combinatorial problem via self-organizing process:an application of the Kohonen algorithm to the travelling salesman problem[J].Biol Cybern,1988,59(1):33-40.
[76]杨培杰,印兴耀,张广智.希尔伯特-黄变换地震信号时频分析与属性提取[J].地球物理学进展,2007,22(5):1585-1590.
[77]邹红星等.时频分析:回溯与前瞻[J].电子学报,2000,28(9):78-84.
[78]刘庆云,李志舜,刘朝晖.时频分析技术及研究现状[J].计算机工程,2004,30(1):171-173.
[79]金雷,李月,杨宝俊.用时频峰值滤波方法消减地震勘探资料中随机噪声的初步研究[J].地球物理学进展,2005,20(3):724-728.
[80]彭勇,周轼尘,希尔伯特变换及解调方法的应用[J].武汉交通科技大学学报,1996,20(5):544-548.
[81]黄迪山,陆乃炎.应用希尔伯特变换提取相位信号[J].振动与动态测试,1988:68-71.
[82]党志敏,贺振华,黄德济.广义希尔伯特变换在含噪信号边缘检测中的应用效果分析[J].油气地球物 2007,5(2):25-27.
[83]苏峰,曲毅,陈军.线性时频分析在弱信号检测中的应用[J].广西民族学院学报2004,10(3):63-65.
[84]周丹 蔡坤宝.基于短时傅立叶变换的脉象信号的模式识别方法[J].重庆科技学院学报,2007,9(3):49-52.
[85]王胜文,陈伏兵.一种基于奇异值分解的特征抽取方法[J].电子与信息学报,2005,27(2):294-297.
[86]何正嘉,警艳阳.基于Hermtiina小波的信号奇异性识别[J].工程数学学报,2001,18(5):37-43.
[87]盖强,张海勇.Hilbert-Huang变换的自适应频率多分辨率分析研究[J].电子学报,2005,33(3):563-566.
[88]陈章位,潘宏.渐进信号瞬态频率的提取[J].Journal of vibration Engmeering,1997,10(4):451-457.
[89]庞浩,李东霞等.基于Hiibert移相滤波的全数字锁相环[J].电网技术,2003,27(11):55-59.
[90]何仁贵,黄登山.一种改进的正弦波频率谱估计方法[J].西安工业大学学报,2004.24(4):328-331.
[91]李钢虎等.Winger-HOugh变换在水下目标信号检测中的应用[J].兵工学报,2006,27(1):121-125.
[92]严光洪,束滋德.包络分析法在某航行体振动信号处理中的应用[J].振动、测试与诊断,2000,20(3):202-205.
[93]丁康等.复解析带通滤波器及其在解调分析中的应用[J].振动工程学报,2000,13(3):385-391.
[94]何岭松,杨叔子.包络检波的数字滤波算法[J].振动工程学报,1997,10(3):362-367.
[95]张建勋 俞智昆.希尔伯特变换原理在扭转振动分析中应用的理论研究[J].昆明理工大学学报,1998,23(1):90-94.
[96]黄迪山,陆乃炎.应用希尔伯特变换提取相位信号[J].振动与动态测试,1988:68-71.
[97]李生红,王树勋.利用希尔伯特变换进行谐波分析[J].电子学报,1997,25(1):98-101.
[98]王兆旗,丘斌煌.希尔伯特变换子波反褶积及其改进方案[J].山西建筑,2007,33(7):117-118.
[99]徐伯勋,林贵标.希尔伯特变换子波反褶积[J].石油物探,1987,26(1):84-85
[110]饶建中.子波反褶积程序[J].物探化探计算技术,1993,15(2):77-78.
[101]刘雯林.子波处理方法的研究[J].石油地球物理勘探,1985,20(2):64-65.
[102]周洁玲,成世琦.反褶积与信噪比的关系研究[J].西安石油学院学报,2002,17(4):21-22.
[103]张南平,黄延禾.低信噪比地震资料的子波处理方法[J].石油物探,1989,28(1):90-91.
[104]赵学智,符红羽.小波分析住信号包络提取中的应用[J].广东工业大学学报,2000.17(3):36-39.
[105]冯志鹏,李学军.基于平稳小波包分解的水轮机非平稳振动信号希尔伯特谱分析[J].中国电机工程学报,2006,26(12):79-84.
[106]彭志科,何永勇等.用小波时频分析方法研究发电机碰摩故障特征[J].中国电机工程学报,2003,23(5):75-79.
[107]杨世锡,胡劲松等.旋转机械振动信号基于EMD的希尔伯特变换和小波变换时频分析比较[J].中国电机工程学报,2003,23(6):102-107.
[108]张绪省等.信号包络提取-从希尔伯特变换到小波变换[J].电子科学学刊,1997,19(1):120-123.
[109]皮骏等.基于小波原理的振动信号包络提取法[J].中国民航学院学报,2005,23(2):12-15.
[110]王汉平等.基于遗传算法进化神经网络的潜射导弹筒盖压力预测[J].北京理 工大学学报,2006,26(1):23-26.
[111]郑志军,郑守淇.用基于实数编码的自适应遗传算法进化神经网络[J].计算机工程与应用,2000,15(9):36-37.
[112]袁慧梅.具有自适应交换率和变异率的遗传算法[J].首都师范大学学报,2000,21(3):14-20.
[113]陈明.基于进化遗传算法的优化计算[J].软件学报,1998,9(11):876-879.
[114]辛长欣等.BP神经网络在回归分析中的应用[J].西安工业学院学报,2002,22(2):129-135.
[115]雷英杰,张善文,李续武等.MATLAB遗传算法工具箱及应用[M].西安:西安电子科技大学出版社.2004.
[116]L.科恩.时-频分析:理论与应用[M].西安交通大学出版社,1998.
[117]张贤达,保铮.非平稳信号分析与处理[M].北京:国防工业出版社,2001.
[118]胡广月.数字信号处理-理论、算法与实现[M].清华大学出版社,1996.
[119]阎平凡,张长水.人工神经网络与模拟进化计算[M].北京:清华大学出版社,2001.
[120]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
[121]黄绪德.反褶积与地震道反演[M].北京:石油工业出版社,1992.
[122]秦前清,杨宗凯.实用小波分析[M].西安:西安电子科技大学出版社.1995.
[123]Thiran P,Hasler M.Self-organisation of a one dimensional Kohonen network with quantized weights and inputs[J].Neural Networks,1994,7(9):1427-1439.
[124]邵玉萍等.BP神经网络在测力轮对横垂向桥解耦中的应用研究[J],中国测试技术.2003,5(10):17-19.
[125]赵华敏,陈开周.用神经网络解非线性方程组[J].西安电子科技大学学报,2001,28(1):35-38.
[126]焦李成.神经网络系统理论[J].西安:西安电子科技大学出版社,1996:1-25.
[127]李春宇等.遗传算法在微弱GPS信号捕获方法中的应用[J].航空学报,2007,28(6):1433-1437.
[128]杨慧娟,曲喜强.基于小波变换的声波信号包络提取[J].华北工学院学报,2004,25(4):300-302.
[129]梅璐璐,林京.基于相移小波的信号包络提取方法研究[J].仪器仪表学报,2007,28(6):1122-1128.
[130]李智,陈祥初.基于小波分析的解调分析方法[J].华北工学院测试技术学报,1999,13(3):165-169.
[131]丁康,胡晓斌,谢明.解调分析的一种优化方法[J].振动工程学报,1998,11(2):235-240.
[132]Walden A T,Contreras Cristan A.The phase-corrected undecimated discrete wavelet packet transform and its application to interpreting the timing of events[J].Proceedings of the Royal Society of London,Series A,1998,454:2243-2266.
[133]刘彬,蒋金水.轧机转速波动测量的扭振监测实验研究[J].计量学报,2007,28(3):272-275.
[134]丁康等.解调分析在故障诊断中应用的局限性研究[J].振动工程学报,1997,10(1):235-240.
[135]丁康等.传动箱齿轮和轴故障的振动诊断方法的研究[J].振动与冲击,1994,10(2):235-240.
[136]王志杰,丁康.汽车变速箱故障振动诊断的试验研究[J].汽车工程,1994,10(4):242-248.
[137]丁康,胡晓斌,谢明.解调分析的一种优化方法[J].振动工程学报,1998,11(2):235-240.
[138]马晓建等.机械故障诊断中常用解调方法比较及应用[J].东华大学学报,2001,27(5):105-108.
[139]朱利民.齿轮故障振动信号检波解调分析中的混频效应[J].振动工程学报,1999,12(3):332-335.
[140]马晓建等.碟式分离机横轴轴承的故障分析与诊断[J].化工机械,1999,26(6):320-325.
[141]金少先.齿轮振动的边带分布特征与故障诊断实例[J].振动工程学报,1999,12(3):429-432.
[142]马孝江,王凤利等.局域波时频分布在转子系统早期故障诊断中的应用研究[J].中国电机工程学报,2004,24(3):161-164.
[143]冯志鹏,褚福磊.基于Hilbert-Huang变换的水轮机非平稳压力脉动信号分析[J].中国电机工程学报,2005,25(10):111-115.
[144]赵文生,熊诗波.希尔伯特变换在扭振测量中的应用[J].太原理工大学学报,2006,37(2):187-189.
[145]张建勋,朱景春.振动信号处理技术在旋转机械振动监测与诊断中的应用和动向[J].昆明理工大学学报,2000,25(1):77-79
[146]祝海龙,屈梁生.自组织包络解调及其在轴承诊断中的应用[J].西安交通大学学报,2000,34(7):77-81.
[147]田涛等.希尔伯特变换及其在故障诊断中的应用[J].振动与冲击,1995:15(2):23-27
[148]王延春等.包络分析方法及其在齿轮故障诊断振动诊断中的应用[J].重庆大学学报,1995,18(1):87-91.
[149]马晓建,陈瑞琪,吴文英等.机械故障诊断中常用解调方法的比较及应用[J].东华大学学报,2001,27(5):105-108.
[150]徐春林,江志农,肖英.基于Hilbert变换的包络分析及其在机械故障诊断中的应用[J].机电工程技术,2004,33(7):114-115.