RAMS (Railway Vehicle RAMS) index optimization method and system and storage medium

Abstract

The invention provides a railway vehicle RAMS index optimization method and system and a storage medium, and the method comprises the steps: 1, obtaining a performance parameter related to a railway vehicle RAMS index through a sensor; 2, performing data cleaning processing on the performance parameters to obtain performance parameters after data cleaning; 3, calculating the correlation between the performance parameters after data cleaning and the RAMS index of the railway vehicle; 4, ranking the performance parameters after data cleaning based on correlation to form an optimization sequence; and step 5, adjusting the performance parameters of the railway vehicle according to the optimization sequence so as to optimize the RAMS index of the railway vehicle. According to the method, through correlation analysis, the performance parameter which has the greatest influence on the RAMS index can be identified, so that the optimization target is clearer, and human resources can be concentrated for improvement.

Description

Technical Field

[0001] The present invention belongs to the technical field of RAMS index optimization, and in particular relates to a rail vehicle RAMS index optimization method, system and storage medium. Background Art

[0002] With the acceleration of urbanization and the continuous growth of population, urban rail transit, as an efficient and environmentally friendly public transportation mode, has become increasingly important. In rail transit systems, reliability, availability, maintainability and safety (RAMS) are important indicators for evaluating rail vehicle performance. Improving these indicators is not only related to the safety and comfort experience of passengers, but also directly affects operational efficiency and economic benefits. In practical applications, traditional RAMS index optimization methods often rely on experience, and the operating environment of modern rail vehicles is complex and changeable, involving multiple performance parameters, such as braking distance, maximum traction, maximum operating speed, acceleration, load, vibration intensity and noise intensity. The interaction between these parameters makes the optimization process more complicated and difficult. If there is a lack of systematic evaluation and analysis methods, the overall efficiency and safety of the vehicle may be affected due to the failure to discover performance bottlenecks in a timely manner. Summary of the invention

[0003] The present invention aims to solve the problem that traditional RAMS index optimization methods often rely on experience, and the operating environment of modern rail vehicles is complex and changeable, and the interaction between parameters makes the optimization process more complicated and difficult, thereby providing a rail vehicle RAMS index optimization method, system and storage medium to improve data quality, ensure the reliability of subsequent analysis, and obtain more reliable correlation results.

[0004] To achieve the above-mentioned object of the invention, the present invention provides a method for optimizing the RAMS index of a rail vehicle, comprising:

[0005] Step 1: Use sensors to obtain performance parameters related to the RAMS indicators of rail vehicles;

[0006] Step 2: Perform data cleaning on the performance parameters to obtain the performance parameters after data cleaning;

[0007] Step 3: Calculate the correlation between the performance parameters after data cleaning and the RAMS index of the rail vehicle;

[0008] Step 4: Rank the performance parameters after data cleaning based on relevance to form an optimized sequence;

[0009] Step 5: Adjust the performance parameters of the rail vehicle according to the optimization sequence to optimize the RAMS index of the rail vehicle.

[0010] Furthermore, step 2 specifically includes:

[0011] Step 2.1: Use different wavelet basis functions to decompose the performance parameters obtained by the sensor and obtain wavelet coefficients;

[0012] Step 2.2: Determine a threshold based on the wavelet coefficients;

[0013] Step 2.3: When the wavelet coefficient is greater than or equal to the threshold, the first threshold processing function is used to process the wavelet coefficient to generate the processed wavelet coefficient; wherein the first threshold processing function is:

[0014]

[0015] in, represents the processed wavelet coefficient, sgn represents the sign function, λ represents the threshold, ω j represents the jth original wavelet coefficient, a represents the first adjustment parameter, b represents the second adjustment parameter, and α represents the third adjustment parameter.

[0016] Step 2.4: When the wavelet coefficient is less than the threshold, the second threshold processing function is used to process the wavelet coefficient to generate the processed wavelet coefficient; wherein the second threshold processing function is:

[0017]

[0018] in, represents the processed wavelet coefficient, sgn represents the sign function, λ represents the threshold, ω j represents the jth original wavelet coefficient, a represents the first adjustment parameter, b represents the second adjustment parameter, and α represents the third adjustment parameter.

[0019] Step 2.5: reconstructing the processed wavelet coefficients to form performance parameters after data cleaning;

[0020] Furthermore, the step 2.2 determines the threshold value based on the wavelet coefficient, including:

[0021] Step 2.2.1: After squaring the wavelet coefficients, arrange them in ascending order to form an ascending set;

[0022] Step 2.2.2: Construct a threshold estimation function based on the ascending set;

[0023] Step 2.2.3: Calculate the minimum value of the threshold estimation function, and use the minimum value of the threshold estimation function as the threshold.

[0024] Furthermore, the step 2.5 specifically includes:

[0025] Step 2.5.1: Evaluate the reconstructed wavelet coefficients to form a quality coefficient; the quality coefficient is calculated as follows:

[0026]

[0027] Among them, s(n) represents the original sensor signal, represents the reconstructed sensor signal, n represents the sampling point number, N represents the sampling length, SNR represents the first quality coefficient, and RMSE represents the second quality coefficient;

[0028] Step 2.5.2: Continuously change the adjustment parameters to keep the quality coefficient within the set range to form the performance parameters after data cleaning.

[0029] Furthermore, the step 3 specifically includes:

[0030] Step 3.1: Correspond each cleaned performance parameter with the rail vehicle RAMS index to form a data pair (X i ,Y i );

[0031] Step 3.2: X i Less than X j , and Y i Less than Y j The data pairs are taken as the number of consistent pairs;

[0032] Step 3.3: Place X i Less than X j , and Y i Greater than Y j The data pairs are taken as the number of inconsistent pairs;

[0033] Step 3.4: Calculate the correlation based on the number of consistent pairs and the number of inconsistent pairs; wherein the correlation calculation formula is:

[0034]

[0035] Where T represents the correlation, n c represents the number of consistent pairs, n d represents the number of inconsistent pairs, and n represents the total number of data pairs.

[0036] The present invention also provides a rail vehicle RAMS index optimization system, comprising:

[0037] A performance parameter acquisition module, used to acquire performance parameters related to the RAMS index of rail vehicles using sensors;

[0038] A data cleaning module is used to perform data cleaning on the performance parameters to obtain the performance parameters after data cleaning;

[0039] A correlation calculation module, used to calculate the correlation between the performance parameters after data cleaning and the RAMS index of the rail vehicle;

[0040] A ranking module, used for ranking the performance parameters after data cleaning based on the correlation to form an optimization sequence;

[0041] The performance parameter optimization module is used to adjust the performance parameters of the rail vehicle according to the optimization sequence to optimize the RAMS index of the rail vehicle.

[0042] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps in the method for optimizing the RAMS index of a rail vehicle described in any one of claims 1 to 5 are implemented.

[0043] The present invention provides a rail vehicle RAMS index optimization method, system and storage medium. Compared with the prior art, the present invention can identify the performance parameters that have the greatest impact on the RAMS index through correlation analysis, which makes the optimization target clearer and helps to concentrate human resources for improvement; in addition, the present invention can eliminate noise and outliers by performing data cleaning processing on the performance parameters, improve data quality, ensure the reliability of subsequent analysis, and thus obtain more reliable correlation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A flow chart of a method for optimizing RAMS indicators of a rail vehicle in an embodiment provided by the present invention;

[0045] Figure 2 A schematic diagram of a rail vehicle RAMS index optimization system in an embodiment provided by the present invention. DETAILED DESCRIPTION

[0046] In order to better understand the purpose, structure and function of the present invention, the following is a further detailed description of a rail vehicle RAMS index optimization method, system and storage medium of the present invention in conjunction with the accompanying drawings.

[0047] See also Figure 1 , a rail vehicle RAMS index optimization method, comprising:

[0048] Step 1: Use sensors to obtain performance parameters related to the RAMS index of rail vehicles; the performance parameters include: braking distance, maximum traction of the power system, maximum operating speed, maximum acceleration, rated load, vibration intensity in the car, and noise intensity during operation inside and outside the car;

[0049] Step 2: Perform data cleaning on the performance parameters to obtain the performance parameters after data cleaning;

[0050] Step 2 specifically includes:

[0051] Step 2.1: Use different wavelet basis functions to decompose the performance parameters obtained by the sensor and obtain wavelet coefficients;

[0052] In the embodiment of the present invention, Haar wavelet may be used to decompose the performance parameter signal acquired by the sensor and obtain wavelet coefficients.

[0053] Step 2.2: Determine a threshold based on the wavelet coefficients;

[0054] Wherein, step 2.2: determining a threshold based on the wavelet coefficients comprises:

[0055] Step 2.2.1: After squaring the wavelet coefficients, arrange them in ascending order to form an ascending set;

[0056] Step 2.2.2: Construct a threshold estimation function based on the ascending set;

[0057] Step 2.2.3: Calculate the minimum value of the threshold estimation function, and use the minimum value of the threshold estimation function as the threshold.

[0058] In step 2.2.3, the threshold estimation function is:

[0059]

[0060] Among them, P represents an ascending set, r 2 (i) represents the value of the i-th number in the ascending set, 0≤i≤n, K represents the length of the sensor signal, and λ represents the threshold;

[0061] Step 2.3: When the wavelet coefficient is greater than or equal to the threshold, the first threshold processing function is used to process the wavelet coefficient to generate the processed wavelet coefficient; wherein the first threshold processing function is:

[0062]

[0063] in, represents the processed wavelet coefficient, sgn represents the sign function, λ represents the threshold, ω j represents the jth original wavelet coefficient, a represents the first adjustment parameter, b represents the second adjustment parameter, and α represents the third adjustment parameter.

[0064] Step 2.4: When the wavelet coefficient is less than the threshold, the second threshold processing function is used to process the wavelet coefficient to generate the processed wavelet coefficient; wherein the second threshold processing function is:

[0065]

[0066] in, represents the processed wavelet coefficient, sgn represents the sign function, λ represents the threshold, ω j represents the jth original wavelet coefficient, a represents the first adjustment parameter, b represents the second adjustment parameter, and α represents the third adjustment parameter.

[0067] The present invention can adjust the degree of contraction of the threshold processing function by changing the value of the adjustment parameter, so that the threshold processing function can be converted between a conventional hard threshold function and a soft threshold function. In this way, different processing methods can be selected according to different performance parameter processing requirements. For example, when the noise level is high, the soft threshold function can be used to better reduce the impact of noise on the signal, while when the signal is strong or the noise is small, the hard threshold function can be used to retain more original signal features.

[0068] Step 2.5: reconstructing the processed wavelet coefficients to form performance parameters after data cleaning;

[0069] Step 2.5 specifically includes:

[0070] Step 2.5.1: Evaluate the reconstructed wavelet coefficients to form a quality coefficient; the quality coefficient is calculated as follows:

[0071]

[0072] Among them, s(n) represents the original sensor signal, represents the reconstructed sensor signal, n represents the sampling point number, N represents the sampling length, SNR represents the first quality coefficient, and RMSE represents the second quality coefficient;

[0073] Step 2.5.2: Continuously change the adjustment parameters to keep the quality coefficient within the set range to form the performance parameters after data cleaning.

[0074] The present invention guides the strategy of noise removal by evaluating the reconstructed wavelet coefficients, thereby improving the denoising efficiency.

[0075] Step 3: Calculate the correlation between the performance parameters after data cleaning and the RAMS index of the rail vehicle;

[0076] Furthermore, step 3 specifically includes:

[0077] Step 3.1: Correspond each cleaned performance parameter with the rail vehicle RAMS index to form a data pair (X i ,Y i );

[0078] Step 3.2: X i Less than Xj , and Y i Less than Y j The data pairs are taken as the number of consistent pairs;

[0079] Step 3.3: Place X i Less than X j , and Y i Greater than Y j The data pairs are taken as the number of inconsistent pairs;

[0080] Step 3.4: Calculate the correlation based on the number of consistent pairs and the number of inconsistent pairs; wherein the correlation calculation formula is:

[0081]

[0082] Where T represents the correlation, n c represents the number of consistent pairs, n d represents the number of inconsistent pairs, and n represents the total number of data pairs.

[0083] Step 4: ranking the performance parameters after data cleaning based on the correlation to form an optimization sequence;

[0084] Step 5: Adjust the performance parameters of the rail vehicle according to the optimization sequence to optimize the RAMS index of the rail vehicle.

[0085] By performing correlation analysis on the performance parameters of the rail vehicle, the present invention can better understand which performance parameters affect the RAMS index of the rail vehicle, so that the staff can focus more on important performance tuning.

[0086] See also Figure 2 The present invention also provides a rail vehicle RAMS index optimization system, comprising:

[0087] A performance parameter acquisition module is used to use sensors to acquire performance parameters related to the RAMS index of rail vehicles; the performance parameters include: braking distance, maximum traction of the power system, maximum operating speed, maximum acceleration, rated load, vibration intensity in the car, and noise intensity during operation inside and outside the car;

[0088] A data cleaning module is used to perform data cleaning on the performance parameters to obtain the performance parameters after data cleaning;

[0089] A correlation calculation module, used to calculate the correlation between the performance parameters after data cleaning and the RAMS index of the rail vehicle;

[0090] A ranking module, used for ranking the performance parameters after data cleaning based on the correlation to form an optimization sequence;

[0091] The performance parameter optimization module is used to adjust the performance parameters of the rail vehicle according to the optimization sequence to optimize the RAMS index of the rail vehicle.

[0092] Compared with the prior art, the beneficial effects of a rail vehicle RAMS index optimization system provided by the present invention are the same as the beneficial effects of a rail vehicle RAMS index optimization method described in the above technical solution, which will not be elaborated here.

[0093] The present invention also provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps in the above-mentioned method for optimizing the RAMS index of a rail vehicle are implemented. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present invention are the same as the beneficial effects of the method for optimizing the RAMS index of a rail vehicle described in the above-mentioned technical solution, which will not be repeated here.

Claims

1. A method for optimizing RAMS index of rail vehicles, characterized by: include: Step 1: Use sensors to obtain performance parameters related to the RAMS indicators of rail vehicles; Step 2: Perform data cleaning on the performance parameters to obtain the performance parameters after data cleaning; Step 3: Calculate the correlation between the performance parameters after data cleaning and the RAMS index of the rail vehicle; Step 4: Rank the performance parameters after data cleaning based on relevance to form an optimized sequence; Step 5: Adjust the performance parameters of the rail vehicle according to the optimization sequence to optimize the RAMS index of the rail vehicle.

2. A rail vehicle RAMS index optimization method according to claim 1, characterized in that: Step 2 specifically includes: Step 2.1: Use different wavelet basis functions to decompose the performance parameters obtained by the sensor and obtain wavelet coefficients; Step 2.2: Determine a threshold based on the wavelet coefficients; Step 2.3: When the wavelet coefficient is greater than or equal to the threshold, the first threshold processing function is used to process the wavelet coefficient to generate the processed wavelet coefficient; wherein the first threshold processing function is: in, represents the processed wavelet coefficient, sgn represents the sign function, λ represents the threshold, ω j represents the jth original wavelet coefficient, a represents the first adjustment parameter, b represents the second adjustment parameter, and α represents the third adjustment parameter. Step 2.4: When the wavelet coefficient is less than the threshold, the second threshold processing function is used to process the wavelet coefficient to generate the processed wavelet coefficient; wherein the second threshold processing function is: in, represents the processed wavelet coefficient, sgn represents the sign function, λ represents the threshold, ω j represents the jth original wavelet coefficient, a represents the first adjustment parameter, b represents the second adjustment parameter, and α represents the third adjustment parameter. Step 2.5: Reconstruct the processed wavelet coefficients to form performance parameters after data cleaning.

3. A rail vehicle RAMS index optimization method according to claim 2, characterized in that: The step 2.2 determines a threshold value based on the wavelet coefficient, including: Step 2.2.1: After squaring the wavelet coefficients, arrange them in ascending order to form an ascending set; Step 2.2.2: Construct a threshold estimation function based on the ascending set; Step 2.2.3: Calculate the minimum value of the threshold estimation function, and use the minimum value of the threshold estimation function as the threshold.

4. A rail vehicle RAMS index optimization method according to claim 2, characterized in that: The step 2.5 specifically includes: Step 2.5.1: Evaluate the reconstructed wavelet coefficients to form a quality coefficient; the quality coefficient is calculated as follows: Where s(n) represents the original sensor signal, represents the reconstructed sensor signal, n represents the sampling point number, N represents the sampling length, SNR represents the first quality coefficient, and RMSE represents the second quality coefficient; Step 2.5.2: Continuously change the adjustment parameters to keep the quality coefficient within the set range to form the performance parameters after data cleaning.

5. A rail vehicle RAMS index optimization method according to claim 2, characterized in that: The step 3 specifically includes: Step 3.1: Correspond each cleaned performance parameter with the rail vehicle RAMS index to form a data pair (X i , Y i ); Step 3.2: X i Less than X j , and Y i Less than Y j The data pairs are taken as the number of consistent pairs; Step 3.3: Place X i Less than X j , and Y i Greater than Y j The data pairs are taken as the number of inconsistent pairs; Step 3.4: Calculate the correlation based on the number of consistent pairs and the number of inconsistent pairs; wherein the correlation calculation formula is: Where T represents the correlation, n c represents the number of consistent pairs, n d represents the number of inconsistent pairs, and n represents the total number of data pairs.

6. A rail vehicle RAMS index optimization system, characterized in that: include: A performance parameter acquisition module, used to acquire performance parameters related to the RAMS index of rail vehicles using sensors; A data cleaning module is used to perform data cleaning on the performance parameters to obtain the performance parameters after data cleaning; A correlation calculation module, used to calculate the correlation between the performance parameters after data cleaning and the RAMS index of the rail vehicle; A ranking module, used for ranking the performance parameters after data cleaning based on the correlation to form an optimization sequence; The performance parameter optimization module is used to adjust the performance parameters of the rail vehicle according to the optimization sequence to optimize the RAMS index of the rail vehicle.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in the method for optimizing the RAMS index of a rail vehicle described in any one of claims 1 to 5 are implemented.

Images