A method of track switch management value

Abstract

The application discloses a kind of track turnout management values methods, the method includes: step S1) determining classification standard, according to the different classification statistics of track turnout;Step S2) according to the type of turnout track determines statistical dimension;Step S3) determines statistical angle, whether by K-S test method to determine to obey normal distribution;Step S4) data processing.The method provided by the application is by big data statistics analysis turnout part data, according to statistical interval, from TQI, acceleration maximum value and its RMS value several dimensions are statistically analyzed, and then the usual management value of track turnout is designed;It can be used for the basis of turnout routine maintenance in track industry.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of rail transit, and particularly relates to a method for managing the value of a rail turnout. BACKGROUND

[0002] High-speed turnouts are the most critical components in high-speed lines in China. High-speed turnouts are one of the basic connection devices in railway track lines. Not only are there complex turnout areas in station stopping sections, but there are also turnouts distributed in high-speed track lines. Due to the complex track relationship of the turnouts, the track cross section is complex and variable, which leads to the fact that the wheel-rail contact relationship becomes more complex when a high-speed train runs to the track turnout. Meanwhile, in the daily operation process of the track line, the turnout, as one of the three weak links of the railway track, has the characteristics of a large number, complex structure, short service life, and large maintenance investment. At present, there is no unified index parameter for managing the turnout. For the unevenness of the turnout part, the part is mainly excluded, and there is no unified standard for maintenance.

[0003] Kolmogorov-Smirnov test, K-Stest can use sample data to infer whether the sample comes from a population that obeys a certain theoretical distribution. It is a goodness-of-fit test method and is suitable for exploring the distribution of random variables. SUMMARY

[0004] In view of the deficiencies in the prior art, the application provides a method for managing the value of a rail turnout. Since the turnout widening or the track quality index of a section includes multiple sets of turnout conditions, it is difficult to better evaluate the track quality index of the turnout area on site, which is not conducive to the maintenance and management of the turnout. Therefore, a plurality of high-speed rail lines are selected, and based on the test data of the high-speed comprehensive detection train, the TQI, the maximum acceleration, and the maximum frame transverse displacement of the widened or non-widened turnout are analyzed from a statistical point of view, and then the track quality index of the turnout area is analyzed and a management recommendation value is given, thereby forming a standard for evaluating the overall quality state of the turnout.

[0005] In order to achieve the above purpose, the technical scheme of the application is as follows:

[0006] A method for managing the value of a rail turnout, comprising the following steps:

[0007] Step S1) determining a classification standard, and classifying and counting the rail turnouts according to different types;

[0008] Step S2) determining the statistical dimension according to the type of the turnout track;

[0009] Step S3) determining the statistical angle, and determining whether it obeys a normal distribution by a K-S test method;

[0010] Step S4) data processing.

[0011] Further, the step S1) comprises the following steps:

[0012] Step S101) According to whether the track gauge is widened, it is divided into two types: non-widened turnout and widened turnout. Using the dynamic detection data of high-speed railway, the spatial geometric characteristics of turnout area are statistically analyzed,

[0013] Step S102) According to the running speed of the car body under the turnout, it is divided into two grades: 200km / h-250km / h and 250km / h-350km / h.

[0014] Further, the step S2) comprises the following steps:

[0015] Step S201) For non-widened turnout: calculate the TQI of turnout track, the maximum and RMS value of lateral acceleration, the maximum and RMS value of vertical acceleration, the left frame transverse displacement and the right frame transverse displacement;

[0016] Step S202) For widened turnout: exclude the widened part of the TQI of the widened turnout track, and replace the remaining road section with the average value after excluding the corresponding value, that is, the average TQI; the maximum and RMS value of lateral acceleration, the maximum and RMS value of vertical acceleration, the left frame transverse displacement GL and the right frame transverse displacement GR;

[0017] Step S203) Determine the exclusion principle: based on the turnout account data, exclude the data with negative or 0 in the turnout frog tip mileage position, exclude the data with turnout frog tip mileage maintenance error, and exclude the data of outliers.

[0018] Further, the step S3) comprises the following steps:

[0019] Step S301) Determine the descriptive parameters of the data: mean, variance, minimum, maximum, peak-to-peak value, i.e. maximum-minimum, the multiple relationship of TQI before and after exclusion, the distribution of 80%, 85%, 90%, 95%,

[0020] Step S302) Determine whether it conforms to normal distribution by K-S test method.

[0021] Further, the step S4) comprises the following steps:

[0022] Step S401) Preprocessing of statistical unit: ensure that only one set of turnout is contained in the statistical range;

[0023] Step S402) Parameter calculation principle;

[0024] Step S403) Parameter calculation: based on the pretreatment of the statistical unit, calculate the statistical parameters such as non-rejected TQI, rejected TQI, rejected = mean TQI, acceleration maximum value, acceleration RMS, maximum value of frame transverse displacement, etc.

[0025] Further, the calculation principle is:

[0026] Principle one: if the sequence obeys normal distribution, select the normal value according to the 3σ principle;

[0027] Principle two: if it does not obey normal analysis, select two normal values according to the 90% and 95% confidence interval, compare the two values, and then select the corresponding normal value;

[0028] Principle three: based on the above standard, the data is verified by actual cases and fine-tuned.

[0029] Beneficial effects: the method provided by the application is to analyze the data of the turnout part through big data statistics, to perform statistical analysis on several dimensions such as TQI, the maximum value of acceleration and the RMS value thereof according to a 200m interval, and to design the normal management value of the track turnout; it can be used as the basis for daily maintenance of turnouts in the track industry. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 is a local feature distribution graph of a turnout space in a straight-to-turn turnout of the two kinds of turnouts of the application;

[0031] Figure 2 is a data processing flowchart of the application;

[0032] Figure 3 is a TQI time change trend graph in Example 3;

[0033] Figure 4 is an acceleration time change trend graph in Example 3;

[0034] Figure 5 is an acceleration RMS time change trend graph in Example 3;

[0035] Figure 6 is a frame transverse displacement time change trend graph in Example 3;

[0036] Figure 7 is an analysis content graph of a certain high-speed railway in Example 4. DETAILED DESCRIPTION

[0037] The application will be described below with reference to specific examples. Those skilled in the art can understand that these examples are only used to illustrate the application, and do not limit the scope of the application in any way.

[0038] A method for managing track turnout values ​​includes the following steps:

[0039] Step S1) Determine the classification criteria and classify and count the track switches according to their different types;

[0040] Step S101) Based on whether the track gauge is widened, turnouts are divided into two types: non-widened track gauge turnouts and widened track gauge turnouts. Using dynamic inspection data from high-speed railways, the spatial geometric characteristics of the turnout area are statistically analyzed.

[0041] Step S102) Based on the running speed of the train under the turnout, it is divided into two levels: 200km / h~250km / h and 250km / h~350km / h.

[0042] like Figure 1 As shown, the spatial geometric local features of the turnout area during straight-through turns in both types of turnouts are as follows:

[0043] (1) Non-widened gauge turnout

[0044] When a train passes through a straight turnout, the fluctuations in the left and right track directions in the turnout area are significantly larger than in the main line area, with extreme values ​​of ±2mm for the left track direction and ±3mm for the right track direction. There are no significant changes in left-right elevation or vehicle acceleration. Regarding track gauge, noticeable fluctuations occur in both the switch and frog areas, with a maximum value in the switch area, ranging from -1mm to 3mm. Regarding level, fluctuations occur in both the switch and frog areas, with a peak in the switch area, reaching an extreme value of 2mm. For the triangular joists, the fluctuations are mainly concentrated in the frog area, with extreme values ​​reaching ±3mm.

[0045] from Figure 1 As can be seen from the waveforms of two turnouts, a sharp abrupt change in gauge occurs at the tip of the switch rail, with the peak value ranging from -1mm to +3mm. Simultaneously, this gauge abrupt change at the switch rail tip also corresponds to a change in the track direction on one side, with a similar peak value appearing at the same mileage location. The main reason for this is the gauge widening caused by the concealed switch rail design, an inherent structural irregularity that facilitates accurate turnout location. When the track inspection vehicle passes through the switch rail tip area, it measures the gauge at a position 16mm below the rail top, which is the straight main rail after planing, resulting in a peak value for both the gauge and the track direction on one side. Furthermore, the distance from the tip of the switch rail to the tip of the frog rail at turnout No. 18 is 55m. At a mileage location approximately 55m away from the switch rail tip, a reverse gauge change occurs at the frog rail tip.

[0046] When the train passes through the turnout laterally, the track alignment and the lateral acceleration of the train body in the turnout area show obvious change characteristics. The track alignment shows obvious change characteristics mainly due to the detection principle when the detection vehicle turns, but the turnout position can be conveniently located; the lateral acceleration of the train body in the turnout area is obviously increased, and the fluctuation is mainly caused by the unbalanced centrifugal acceleration due to the absence of super-elevation in the turnout area.

[0047] (2) Track-gauge widened turnout

[0048] When the train passes through the turnout straightly, the track alignment and the track gauge in the turnout area show obvious change characteristics. The track gauge is affected by the outward bending of the basic track alignment on one side of the switch area, and the peak value reaches 15-17 mm; the track alignment of the widened side also shows a sudden change, and the peak value is about 7-9 mm. The high-low and triangular pits do not show obvious change characteristics in the turnout area; the level shows an "M" type distribution in the entire turnout area, and the two positive peaks are located at the tail end of the switch area and the middle position of the frog area, respectively, and the amplitude is about 3 mm.

[0049] When the train passes through the turnout laterally, the track alignment and the track gauge in the turnout area show obvious change characteristics. The track gauge in the switch area has a peak value of 15-20 mm, which is caused by the structural characteristics of the widened side of the track-gauge widened turnout; the waveform characteristics of the track alignment on the non-widened side are caused by the detection principle, and the waveform is consistent with the characteristics of the track-gauge non-widened turnout; the track alignment on the widened side is affected by the widened track gauge, and a peak value of 3-4 mm appears in the switch area.

[0050] Step S2) determining the statistical dimension according to the type of the turnout track:

[0051] Step S201) for non-widened turnout: calculating the TQI of the turnout track, the maximum value and the RMS value of the lateral acceleration, the maximum value and the RMS value of the vertical acceleration, the left frame lateral displacement and the right frame lateral displacement;

[0052] Step S202) for widened turnout: excluding the widened part of the TQI of the widened turnout track, and replacing the remaining road section with the average value after excluding the corresponding value, i.e. the average TQI; the maximum value of the lateral acceleration, i.e. the lateral-G_Lacc_1(L) and the RMS value, i.e. the lateral-G_Lacc_1(L).1, the maximum value of the vertical acceleration, i.e. the vertical-G_Vacc(V) and the RMS value, i.e. the vertical-G_Vacc(V).1, the left frame lateral displacement GL(mm) and the right frame lateral displacement GR(mm);

[0053] Step S203) determining the exclusion principle: based on the turnout account data, excluding the data with negative or 0 in the turnout frog tip mileage position, excluding the data with maintenance error in the turnout frog tip mileage, and excluding the data of outliers.

[0054] Step S3) Determine the statistical angle, determine whether to comply with the normal distribution by K-S test method;

[0055] Step S301) Determine the descriptive parameters of the data: mean, variance, minimum, maximum, peak-to-peak value, i.e. maximum - minimum, the multiple relationship of TQI before and after rejection, 80%, 85%, 90%, 95% distribution,

[0056] Step S302) Determine whether to comply with the normal distribution by K-S test method.

[0057] Assume that the sample size of two samples is n1 and n2, respectively, and F1(x) and F2(x) represent the cumulative empirical distribution functions of the two samples, D = max(F1(x)-F2(x)), the test statistic is approximately normally distributed, and expression (1) is

[0058]

[0059] H0: The two data distributions are consistent or the data comply with the theoretical distribution;

[0060] H1: The two data distributions are inconsistent or the data do not comply with a certain theoretical distribution;

[0061] Determination basis: In the solving process, two data will be output, the first data is the statistic, and the second data is the P value, the closer the statistic is to 0, the better the data fits the standard normal distribution, if the P value is greater than the significance level, usually 0.05, also use this value as the judgment basis, accept the null hypothesis, then determine that the population of the sample complies with the normal distribution.

[0062] Step S4) Data processing, such as Figure 2 as shown:

[0063] Step S401) Preprocessing of the statistical unit:

[0064] For widened turnouts, since there may be 1 or 2 groups of turnouts within a 200m statistical range, the data are processed according to the following two cases, and then it is ensured that there is only one group of turnouts within the 200m section range. Assuming that the two turnout point mileages are x(i), x(j) (j < i), if the turnout width is L, then

[0065] (1) If x(j)-x(i)>L+100, then i is judged as a single turnout; the statistical principle is to take the turnout point mileage position corresponding to the single turnout in the table as the starting point, and take 200m forward and backward as the statistical unit;

[0066] (2) If the first condition is not met, the two switches at i and j are taken as a set unit of double switches, and the statistical principle is that the center point of the distance between the two switch rail tip mileposts is taken as the starting point, the statistical unit of the previous switch is taken as 200 m forward, and the statistical unit of the latter switch is taken as 200 m backward.

[0067] Step S402) Parameter calculation principle:

[0068] Principle one: if the sequence obeys normal distribution, the normal value is selected according to the 3σ principle;

[0069] Principle two: if it does not obey normal analysis, two normal values are selected according to 90% and 95% confidence interval, the two values are compared, and then the corresponding normal value is selected;

[0070] Principle three: based on the above criteria, the data is verified by actual cases and fine-tuned.

[0071] Step S403) Parameter calculation:

[0072] Based on the preprocessing of the statistical unit, the statistical parameters such as un-removed TQI, removed TQI, removed = mean TQI, maximum acceleration, acceleration RMS, maximum frame transverse displacement, etc. are calculated; the specific calculation is as follows:

[0073] (1) Un-removed TQI

[0074] According to the traditional calculation method of TQI, the statistical formulas (2), (3) are:

[0075]

[0076]

[0077] Where: σ i = left high-low, right high-low, left track, right track, track gauge, horizontal, single standard deviation of triangular pit; N is the number of 200m calculation unit sampling points; x ij The arithmetic mean value, high-low and track are calculated using 1.5m-42m wavelength data;

[0078] (2) Removed TQI

[0079] Based on the switch rail tip milepost position in the account table, the nearest previous sampling point from the switch rail tip milepost is determined as the basis, and the points L / 0.25 / 2 points before and after the point are removed, and then the TQI value and the average value of the remaining points are calculated (as the replacement value of the removed TQI point in (3)).

[0080] (3) Removed = mean TQI

[0081] The processing method is the same as ②, the only difference is to find the removed point position, the value of the removed point position is replaced by the average value calculated in ②, and then the TQI value of all points in the section is calculated again;

[0082] (4) Maximum acceleration

[0083] Directly count the maximum value of the horizontal and vertical acceleration of 200 units;

[0084] (5) Acceleration RMS

[0085] If the 200m statistical unit contains n sample points, the calculation formula (4) of RMS is as follows:

[0086]

[0087] Example 1: Analysis of TQI general management value of gauge widening turnout

[0088] 1.1 Process analysis

[0089] (1) In the speed range of 200-250km / h, a total of 607 widened turnouts were counted, and the statistical indicators, parameter correlation indicators, and three-dimensional analysis and processing of normal distribution were analyzed, and the results are as follows:

[0090] From the statistical indicators, the unremoved TQI is about twice the removed TQI (or removed = average TQI); the removed TQI is basically the same as the removed = average TQI; from the correlation index results, the correlation between the removed TQI and the removed = average TQI is 1; according to the size of P value and 0.05, it is known that it conforms to the normal distribution, see Table 1:

[0091] Table 1 Normal distribution table

[0092] Statistical dimension Statistical value P value Whether to obey normal distribution Unremoved TQI 0.08247 0.00049 No Removed TQI 0.08585 0.00024 No Removed = mean TQI 0.08219 0.00052 No Horizontal-G_Lacc_1(L) 0.08825 0.00015 No Vertical-G_Vacc(V) 0.08682 0.00020 No Horizontal-G_Lacc_1(L).1 0.11212 0.00000 No Vertical-G_Vacc(V).1 0.16369 0.00000 No GL (mm) 0.16821 0.00000 No GR (mm) 0.05764 0.03402 No

[0093] (2) In the speed range of 250-350km / h, a total of 1682 widened turnouts were counted, from the statistical indicators, the unremoved TQI is about twice the removed TQI (or removed = average TQI); the removed TQI is basically the same as the removed = average TQI; from the correlation index results, the correlation between the removed TQI and the removed = average TQI is 1; according to the size of P value and 0.05, it is known that it conforms to the normal distribution, see Table 2:

[0094] Table 2 Normal distribution table

[0095]

[0096] 1.2 General value analysis

[0097] The general value analysis is made according to the principle one and the principle two, and the details are as follows:

[0098] (1) The speed range is 200-250 km / h

[0099] 1) The TQI general value analysis is shown in Table 3:

[0100] Table 3 The TQI general value analysis table

[0101]

[0102] After rounding up the statistical results, it is known from the table that the general values determined by 90% and 95% are the same, that is, the general value of the TQI without elimination is 9. Similarly, the general value of the TQI after elimination is 5. The general value of the TQI after elimination of the mean value is 5. If the unique general value, it can be 5.

[0103] 2) The acceleration maximum value and RMS are shown in Table 4:

[0104] Table 4 The TQI acceleration analysis result table

[0105]

[0106] The general value of the lateral acceleration is determined in the rounding way. It is known from the table that the general values corresponding to 90% and 95% are 0.05 and 0.06. The general value of the vertical acceleration maximum value is 0.05. The general value of the lateral acceleration RMS can be determined as 0.003. The general value of the vertical acceleration RMS can be determined as 0.003 and 0.004.

[0107] 3) The frame lateral displacement is shown in Table 5:

[0108] Table 5 The frame lateral displacement analysis result table

[0109]

[0110] After rounding up the statistical results, it is known from the table that the general values determined by 90% and 95% are the same, that is, the general value of the TQI without elimination is 9. Similarly, the general value of the TQI after elimination is 5. The general value of the TQI after elimination of the mean value is 5. If the unique general value, it can be 5.

[0111] (2) The speed range is 250-350 km / h

[0112] 1) The TQI threshold value analysis is shown in Table 6:

[0113] Table 6 The TQI general value analysis result table

[0114]

[0115] The statistical results are rounded up, and from the table, the normal value of 90% and 95% is 10, i.e. the normal value of TQI without elimination is 10. Similarly, the normal value of TQI after elimination is 6, and the normal value of elimination-average TQI is 5 and 6.

[0116] 2) Maximum value and RMS of acceleration, see Table 7:

[0117] Table 7: Analysis results table of acceleration

[0118]

[0119] The normal value of lateral acceleration is determined in a rounding manner, and from the table, the normal value of 90% and 95% is 0.05 and 0.06; the normal value of maximum value of vertical acceleration is 0.06; the normal value of RMS of lateral acceleration can be determined as 0.003 and 0.004; and the normal value of RMS of vertical acceleration can be determined as 0.004 and 0.005.

[0120] 3) Amount of lateral displacement of the frame, see Table 8:

[0121] Table 8: Analysis results table of lateral displacement of the frame

[0122]

[0123] The statistical results are rounded up, and from the table, the normal value of 90% and 95% of the maximum value of left track gauge can be determined as 9; and the normal value of the maximum value of right track gauge can be determined as 12 and 13.

[0124] Example 2: Analysis of normal management values of TQI of non-widened turnout

[0125] 2.1 Process analysis

[0126] (1) Speed range 200-250 km / h

[0127] A total of 1681 sets of non-widened turnouts are statistically analyzed, and from the statistical data, for non-widened turnouts of the speed grade of 200-250 km / h, the TQI of 95% of turnout areas is less than 6.5; from the correlation index results, the correlation of elimination TQI and elimination-average TQI is 1; according to the size of P value and 0.05; from the correlation coefficient table, TQI has a certain correlation with lateral and vertical accelerations, and the correlation is between 0.3 and 0.6.

[0128] (2) Speed range 250-350 km / h

[0129] The widened turnout 1411 groups are counted together. From the statistical data, for the non-widened turnout of 200-250 km / h speed level, 95% of the turnout area TQI is less than 5. From the correlation index results, the correlation between TQI and TQI is 1. According to the size of P value and 0.05; From the correlation coefficient table, TQI has certain correlation with lateral and vertical acceleration, and the correlation is between 0.3-0.6.

[0130] 2.2 General value analysis

[0131] (1) Speed range 200-250 km / h

[0132] 1) TQI threshold analysis: after rounding up the statistical results, it is known that the general values determined by 90% and 95% are the same, that is, the general value of TQI is 5, 7.

[0133] 2) Acceleration maximum and RMS: the general value of lateral acceleration is determined according to the rounding method, and it is known that the general values corresponding to 90% and 95% are 0.05; The general value of vertical acceleration maximum is 0.05; The general value of lateral acceleration RMS can be determined as 0.003 and 0.004; The general value of vertical acceleration RMS can be determined as 0.003 and 0.004.

[0134] 3) Frame lateral displacement: after rounding up the statistical results, it is known that the general value of left track gauge maximum can be determined as 9, 10 according to 90% and 95%; The general value of right track gauge maximum can be determined as 11, 12;

[0135] (2) Speed range 250-350 km / h 1) TQI general value analysis: after rounding up the statistical results, it is known that the general values determined by 90% and 95% are the same, that is, the general value of TQI is 5.

[0136] 2) Acceleration maximum and RMS: the general value of lateral acceleration is determined according to the rounding method, and it is known that the general values corresponding to 90% and 95% are 0.04; The general value of vertical acceleration maximum is 0.06; The general value of lateral acceleration RMS can be determined as 0.003 and 0.004; The general value of vertical acceleration RMS can be determined as 0.004 and 0.005.

[0137] 3) Frame lateral displacement: after rounding up the statistical results, according to the principle of 90% and 95%, it is known that the general value of left track gauge maximum can be determined as 10, 12; The general value of right track gauge maximum can be determined as 11, 12.

[0138] 2.3 General value conclusion

[0139] (1) For track gauge widened turnouts, the widened part of the data needs to be removed and recalculated. It is more beneficial for the evaluation and management of turnouts to include only one set of turnouts within a 200m range.

[0140] (2) The original TQI of the widened gauge turnout is close to twice the TQI of the remaining sample points after removing the widened part;

[0141] (3) Determine the typical value range of turnouts according to the 90% and 95% cumulative distribution law. The typical values ​​determined according to the maximum value principle are shown in Tables 9 and 10:

[0142] Table 9. Typical Management Values ​​for Widened Gauge Turnouts

[0143]

[0144]

[0145] Table 10 Typical Management Values ​​for Non-Wide Gauge Turnouts

[0146]

[0147] As can be seen from the table: for turnouts with a speed rating of 200km / h to 250km / h, the recommended Track Quality Index (TQI) management value is 6.5; for turnouts with a speed rating of 250km / h to 350km / h, the recommended TQI management value is 5.0.

[0148] Example 3: Analysis of the Time Dimension Characteristics of TQI in the Branch Zone

[0149] For widened and non-widened turnouts, the changing trends of the turnout components are analyzed from a time perspective. The analysis parameters are as follows: For widened turnouts, TQI (excluding TQI), TQI equal to the mean, lateral -G_Lacc_1(L) and RMS values ​​(lateral -G_Lacc_1(L).1), vertical -G_Vacc(V) and RMS values ​​(vertical -G_Vacc(V).1), and frame lateral displacement (left frame (GL(mm)) and right frame (GR(mm)); For non-widened turnouts, TQI, maximum and RMS values ​​of lateral acceleration, maximum and RMS values ​​of vertical acceleration, and frame lateral displacement (left and right frames).

[0150] Data was selected from the Beijing-Shanghai High-Speed ​​Railway (K0-K847) and the Beijing-Guangzhou High-Speed ​​Railway (K781-K1225). The selected data points were: K211+623 on the Beijing-Shanghai High-Speed ​​Railway and K1273+027 on the Beijing-Guangzhou High-Speed ​​Railway.

[0151] Based on the inspection data of the Beijing-Shanghai and Beijing-Guangzhou high-speed railways, the development trend of TQI (Traffic Quality Indicator) in the turnout area is analyzed. An analysis of the widening mileage point K211+623 on the Beijing-Shanghai high-speed railway yields the following results. Figures 3 to 6The figure shows that the value of TQI presents a continuous increasing trend, and other parameters have no obvious trend, which can indicate that the geometric state of turnout area has deteriorated. Similarly, the analysis of the widening mileage point K1273+027 of Beijing-Guangzhou data shows that the value of TQI is relatively stable, and other parameters have no obvious trend, which can indicate that the geometric state of turnout area is relatively stable.

[0152] Example 4: Correlation analysis of turnout TQI and vehicle dynamic response

[0153] Randomly select 200-250 km / h widening line Beijing-Shanghai high-speed railway, Xicheng passenger dedicated line, and non-widening line Xicheng passenger dedicated line, Daxi high-speed railway, Tianjin-Qinhuangdao passenger dedicated line, Xulanheng high-speed railway, Shendan passenger dedicated line; 250-350 km / h widening line Hefu high-speed railway, Beijing-Shanghai high-speed railway, Beijing-Guangzhou high-speed railway and Shanghai-Kunming high-speed railway and non-widening turnout line Ninghang high-speed railway, Shenda high-speed railway, Xulanheng high-speed railway, Hefu high-speed railway, Tianjin-Qinhuangdao passenger dedicated line data, and perform correlation analysis on the parameters.

[0154] 4.1 Speed range 200-250 km / h

[0155] (1) Widening turnout

[0156] The method used is pearson, and the correlation heat map is used for representation. In the figure, TQI1 corresponds to the parameter excluding TQI, TQI2 corresponds to the parameter excluding = mean TQI, L-max corresponds to the parameter horizontal-G_Lacc_1(L), V-Max corresponds to the parameter vertical-G_Vacc(V), L-RMS corresponds to the parameter horizontal-G_Lacc_1(L), V-RMS corresponds to the parameter vertical-G_Vacc(V), GL corresponds to the parameter GL(mm), and GR corresponds to the parameter GR(mm).

[0157] The correlation coefficient of Beijing-Shanghai high-speed railway is as shown in Figure 7 The results show that the two dimensions with strong linear correlation are:

[0158] 1) The second dimension: TQI2 and the first dimension: TQI1, the correlation coefficient of the two is: 0.999666;

[0159] 2) The fifth dimension: L-RMS and the third dimension: L-Max, the correlation coefficient of the two is: 0.836979;

[0160] 3) The third dimension: L-Max and the seventh dimension: GL, the correlation coefficient of the two is: 0.7045;

[0161] 4) The correlation of TQI and V-Max is stronger than that of L-Max.

[0162] (2) Non-widening turnout

[0163] The two dimensions with strong linear correlation of statistical characteristics of Datong-Xi'an high-speed railway are:

[0164] 1) The 4th dimension: L-RMS and the 2nd dimension: L-Max, the correlation coefficient of which is 0.912116;

[0165] 2) The 7th dimension: GR and the 2nd dimension: L-Max, the correlation coefficient of which is 0.849103;

[0166] 3) The 5th dimension: V-RMS and the 3rd dimension: V-Max, the correlation coefficient of which is 0.833971;

[0167] The two dimensions with strong linear correlation of statistical characteristics of Xi'an-Chengdu high-speed railway are:

[0168] 1) The 4th dimension: L-RMS and the 2nd dimension: L-Max, the correlation coefficient of which is 0.836655;

[0169] 2) The 5th dimension: V-RMS and the 3rd dimension: V-Max, the correlation coefficient of which is 0.895590;

[0170] 3) The correlation of TQI and V-Max is stronger than that of L-Max.

[0171] 4.2 Speed range 250-350 km / h

[0172] (1) Widened turnout

[0173] The two dimensions with strong linear correlation of statistical characteristics of Hefei-Fuzhou high-speed railway are:

[0174] 1) The 2nd dimension: TQI2 and the 1st dimension: TQI1, the correlation coefficient of which is 0.999626;

[0175] 2) The 5th dimension: L-RMS and the 3rd dimension: L-Max, the correlation coefficient of which is 0.841060;

[0176] 3) The 6th dimension: V-RMS and the 4th dimension: V-Max, the correlation coefficient of which is 0.881593;

[0177] The two dimensions with strong linear correlation of statistical characteristics of Beijing-Shanghai high-speed railway are:

[0178] 1) The 2nd dimension: TQI2 and the 1st dimension: TQI1, the correlation coefficient of which is 0.999787;

[0179] 2) The correlation of TQI and V-Max is stronger than that of L-Max.

[0180] (2) No widening of the turnout

[0181] The two dimensions of strong linear correlation of the Ninghang high-speed railway are:

[0182] 1) The 5th dimension: V-RMS and the 3rd dimension: V-Max, the correlation coefficient is: 0.832237;

[0183] 2) The difference in correlation between TQI and V-Max and L-Max is not large;

[0184] Similarly, analyze Shenda Expressway, Xulan Expressway, and Jinqin Passenger Dedicated Line. Based on the above results, the correlation between TQI and vertical and horizontal acceleration in different line turnout areas is not consistent. The correlation between TQI and vertical acceleration in some line turnout areas is better, and the correlation between TQI and horizontal acceleration in some line turnout areas is better. In general, the correlation between TQI and vertical and horizontal acceleration in turnout areas is between 0.2 and 0.6, indicating that TQI has a certain influence on acceleration.

[0185] 4.3 Conclusion:

[0186] (1) Based on the statistical analysis of long-term track geometry detection data of more than 20 high-speed railway lines such as Beijing-Shanghai and Beijing-Guangzhou, 18 rail width widened turnouts, and non-widened turnouts, it is recommended to exclude the left and right track directions and track width data within a 200m section range. It is beneficial to include only one set of turnout in the 200m range for turnout evaluation and management.

[0187] (2) Based on the long-term track geometry detection data mining of more than 20 high-speed railway lines such as Beijing-Guangzhou, rail width widened turnouts, and non-widened turnouts, the usual management values of track overall quality index and vehicle dynamic response are obtained. In addition, it is recommended to use vehicle body acceleration and frame lateral displacement to assist in evaluation, as shown in Table 11.

[0188] Table 11 Recommended usual management values of track overall quality index and vehicle dynamic response

[0189]

Claims

1. A method of designing a track switch management value, characterized by, It comprises the following steps: Step S1) determining classification criteria, classifying and counting track switches according to different categories; Step S2) determining statistical dimensions according to the types of switch tracks; Step S3) determining statistical angles, determining whether to conform to normal distribution by K-S test method; Step S4) data processing; The step S1) comprises the following steps: Step S101) dividing into two types of non-widened track gauge switch and widened track gauge switch according to whether the track gauge is widened, and using high-speed rail dynamic inspection data to statistically analyze the spatial geometric characteristics of switch area, Step S102) dividing into two grades according to the running speed of the car body under the switch: 200km / h-250km / h and 250km / h-350km / h; The step S3) comprises the following steps: Step S301) determining the descriptive parameters of data: mean, variance, minimum value, maximum value, peak-to-peak value, i.e. maximum value-minimum value, the multiple relationship of TQI before and after elimination, the distribution of 80%, 85%, 90%, and 95%; Step S302) determining whether to conform to normal distribution by K-S test method; The step S4) comprises the following steps: Step S401) preprocessing of statistical unit: ensuring that only one set of switches is contained in the statistical range; Step S402) parameter calculation principle: principle one: if the sequence conforms to normal distribution, then select the usual value according to the 3σ principle; principle two: if it does not conform to normal analysis, select two usual values according to the 90% and 95% confidence intervals, compare them, and determine the corresponding usual value according to the maximum value principle; principle three: based on the above standards, verify the data with actual cases and make fine adjustments; Step S403) parameter calculation: based on the preprocessing of the statistical unit, calculate the statistical parameters: uneliminated TQI, eliminated TQI, eliminated mean TQI, maximum acceleration, acceleration RMS, maximum frame lateral displacement; Eliminated TQI: based on the switch point rail point location in the account table, taking the nearest previous sampling point to the point location as the basis, eliminating each L / 0.25 / 2 points before and after the point, and then calculating the TQI value and average value of the remaining points; Eliminated mean TQI: based on the switch point rail point location in the account table, taking the nearest previous sampling point to the point location as the basis, eliminating each L / 0.25 / 2 points before and after the point, finding the eliminated point, replacing the value of the eliminated point with the average value of the remaining points, and then calculating the TQI value of all point locations in the section.

2. The method for designing a rail turnout management value according to claim 1, wherein, The step S2) comprises the following steps: step S201) for non-widening turnout: calculating the TQI of the turnout track, the maximum and RMS values of lateral acceleration, the maximum and RMS values of vertical acceleration, the left frame lateral displacement and the right frame lateral displacement; step S202) for widening turnout: excluding the widening part of the TQI of the widening turnout track, replacing the mean value of the remaining road section after excluding the corresponding value, that is, the mean TQI; the maximum and RMS values of lateral acceleration, the maximum and RMS values of vertical acceleration, the left frame lateral displacement GL and the right frame lateral displacement GR; step S203) determining the exclusion principle: based on the turnout account data, excluding the data with negative or 0 in the turnout frog tip mileage position, excluding the data with turnout frog tip mileage maintenance error, and excluding the data of outliers.

3. The method for designing a rail turnout management value according to claim 1, wherein, The calculation and judgment of normal distribution are as follows: assuming that the sample sizes of two samples are n1 and n2, F1(x) and F2(x) represent the cumulative empirical distribution functions of the two samples respectively, D = max(F1(x)-F2(x)), the test statistic is approximately normally distributed, and expression (1) is ； H0: the two data distributions are consistent or the data conforms to the theoretical distribution; H1: the two data distributions are inconsistent or the data does not conform to a certain theoretical distribution; the judgment basis: in the solving process, two data will be output, the first data is the statistic, and the second data is the P value, the closer the statistic is to 0, the better the data fits the standard normal distribution, if the P value is greater than the significance level, the significance level is set to 0.05, which is also used as the judgment basis, the original hypothesis is accepted, and it is judged that the sample population obeys the normal distribution.

Images