Method and system for predicting rail vehicle curve safety based on composite wheel-rail force

Abstract

The present invention discloses a method and system for predicting the safety of a rail vehicle curve based on composite wheel-rail force, and belongs to the technical field of rail vehicle safety detection. The method comprises: obtaining the vertical stiffness coefficient, lateral stiffness coefficient and lateral force influence coefficient of each detection point; obtaining the vertical displacement and lateral displacement occurring at each detection point; calculating the composite wheel-rail force of the detection point; calculating the derailment coefficient of the detection point using the composite wheel-rail force of each detection point; predicting the safety of the target curve based on the derailment coefficients of all detection points on the target curve. The present invention does not rely on a neural network, but uses an experimental method to find the relationship between displacement and composite wheel-rail force, and finally infers the composite wheel-rail force on the track through the displacement of the measured track. Compared with the prior art, the present invention is more accurate and does not require the collection of a large amount of data in advance, and is more suitable for engineering applications.

Description

Technical Field

[0001] The present invention belongs to the technical field of rail vehicle safety detection, and in particular relates to a rail vehicle curve safety prediction method and system based on composite wheel-rail force. Background Art

[0002] As the world's railways move towards high speed, heavy load, large capacity and high density, the safety of train operation has attracted much attention. Train derailment will cause significant loss of life and property. Therefore, the most basic requirement for safe train operation is to ensure that the train does not derail. Since the French scholar Nadal proposed the famous derailment coefficient criterion in 1908, the derailment coefficient has become an important indicator for studying train derailment problems. Therefore, obtaining the derailment coefficient has important theoretical and practical significance.

[0003] The basic indicator for judging whether a vehicle is derailed at home and abroad is the derailment coefficient Q / P, which is the ratio (Q / P) of the lateral force Q and the vertical force P (lateral force Q and vertical force P are the wheel-rail force) acting on the wheels at a certain moment.

[0004] Chinese patent application CN201110415711.1 discloses a method for predicting a derailment coefficient. First, a track inspection vehicle is used to collect data on the left track height unevenness, left track direction unevenness, right track height unevenness, and right track direction unevenness; then the collected data is simulated using professional software ADAMS / RAIL to obtain wheel-rail force data, namely vertical wheel-rail force and lateral wheel-rail force, and then the derailment coefficient is obtained, and the derailment coefficient is normalized; a training sample is selected to train a NARX neural network model; the trained NARX neural network prediction model is tested, and the derailment coefficient data after the test is output; the derailment coefficient data in the test sample and the derailment coefficient data obtained by the neural network after the test are analyzed to evaluate the performance of the NARX neural network prediction model.

[0005] The above method uses neural network to predict wheel-rail force and obtain derailment coefficient. The accuracy of this method is too dependent on the quality of the model itself and the data set. In actual engineering applications, there are not many trial and error opportunities. Summary of the invention

[0006] In view of this, the present invention provides a method and system for predicting the curve safety of a railway vehicle with composite wheel-rail force, which has high accuracy and is suitable for engineering applications.

[0007] In order to solve the above technical problems, the technical solution of the present invention is to adopt a rail vehicle curve safety prediction method based on composite wheel-rail force, comprising:

[0008] A plurality of detection points are set on the target curve on the track, and the vertical stiffness coefficient, the lateral stiffness coefficient and the lateral force influence coefficient of each detection point are obtained;

[0009] Obtain the vertical displacement and lateral displacement of each detection point when the rail vehicle passes through each detection point;

[0010] The composite wheel-rail force of each detection point is calculated using the vertical displacement and lateral displacement of each detection point and the vertical stiffness coefficient, lateral stiffness coefficient and lateral force influence coefficient corresponding to the detection point; the composite wheel-rail force includes vertical force and lateral force;

[0011] The derailment coefficient of each detection point is calculated using the composite wheel-rail force of the detection point;

[0012] The safety of the target curve is predicted based on the derailment coefficients of all detection points on the target curve.

[0013] As an improvement, the method for obtaining the vertical stiffness coefficient, lateral stiffness coefficient and lateral force influence coefficient of the detection point includes:

[0014] Apply a test vertical force to the detection point, obtain a first test vertical displacement of the detection point under the test vertical force, and calculate a vertical stiffness coefficient of the detection point using the first test vertical displacement and the test vertical force;

[0015] A test lateral force is applied to the detection point, and a test lateral displacement and a second test vertical displacement of the detection point under the test lateral force are obtained; the lateral stiffness coefficient and the lateral force influence coefficient of the detection point are calculated using the test lateral displacement, the second test vertical displacement and the test lateral force.

[0016] As a further improvement, the method for calculating the vertical stiffness coefficient of the detection point using the first test vertical displacement and the test vertical force is to use the formula

[0017] KPi=Pi / WPi

[0018] Calculate the vertical stiffness coefficient; where KPi is the vertical stiffness coefficient of the i-th detection point, Pi is the test vertical force of the i-th detection point, and WPi is the first test vertical displacement of the i-th detection point;

[0019] The method for calculating the lateral stiffness coefficient using the test lateral displacement and the test lateral force is to use the formula

[0020] KQi=Qi / WQi

[0021] Calculate the lateral stiffness coefficient; where KQi is the lateral stiffness coefficient of the i-th detection point, Qi is the test lateral force of the i-th detection point, and WQi is the test lateral displacement of the i-th detection point;

[0022] The method for calculating the lateral force influence coefficient using the second test vertical displacement and the test lateral force is to use the formula

[0023] Ui=Qi / WPi2

[0024] Calculate the lateral force influence coefficient; where Ui is the lateral force influence coefficient of the i-th detection point, Qi is the test lateral force of the i-th detection point, and WPi2 is the second test vertical displacement of the i-th detection point.

[0025] As another further improvement, a plurality of test vertical forces of different magnitudes are applied to the detection point, and a plurality of first test vertical displacements of the detection point under the action of the test vertical forces of different magnitudes are obtained; a vertical stiffness function is fitted using the plurality of first test vertical displacements and the plurality of test vertical forces as a vertical stiffness coefficient of the detection point;

[0026] Several test lateral forces of different magnitudes are applied to the detection point, and several test lateral displacements and several second test vertical displacements of the detection point under the action of the test lateral forces of different magnitudes are obtained. The lateral stiffness function and the lateral force influence function are fitted using the several test lateral displacements, the several second test vertical displacements and the several test lateral forces as the lateral stiffness coefficient and the lateral force influence coefficient of the detection point.

[0027] As an improvement, the calculation method of the composite wheel-rail force at the detection point includes:

[0028] Using the formula

[0029] Qit=WQit*KQi

[0030] Calculate the lateral force; where Qit is the lateral force at the i-th detection point, WQit is the lateral displacement at the i-th detection point, and KQi is the lateral stiffness coefficient at the i-th detection point.

[0031] Using the formula

[0032] Pit=[WPit-(Qit / Ui)]*KPi

[0033] Calculate the vertical force; where Pit is the vertical force at the i-th detection point, WPit is the vertical displacement at the i-th detection point, KPi is the vertical stiffness coefficient at the i-th detection point, Qit is the lateral force at the i-th detection point, and Ui is the lateral force influence coefficient at the i-th detection point.

[0034] As an improvement, the method for calculating the derailment coefficient of the detection point is to use the formula

[0035] D i =Qit / Pit

[0036] Calculate the derailment coefficient; where D i is the derailment coefficient of the ith detection point, Pit is the vertical force of the ith detection point, and Qit is the lateral force of the ith detection point.

[0037] As an improvement, the method for predicting the safety of the target curve based on the derailment coefficients of all detection points on the target curve includes:

[0038] Using the formula

[0039]

[0040] Calculate the safety index of the target curve; is the safety index, D i is the derailment coefficient of the ith detection point, w i is the weight of the i-th detection point, and N is the number of detection points;

[0041] The safety index is compared with the safety threshold to obtain the safety level of target completion.

[0042] As an improvement, using the formula

[0043]

[0044] Calculate the weight of the detection point; where w i is the weight of the i-th detection point, is the local curvature of the ith detection point, is the maximum curvature of the target curve, k is the adjustment coefficient, and n is the adjustment index.

[0045] As an improvement, When ≤0.5Dt, the target curve is low risk;

[0046] At 0.5Dt< When ≤Dt, the target curve is of medium risk;

[0047] exist >Dt, the target curve is high risk; is the safety index, and Dt is the safety threshold.

[0048] The present invention also provides a rail vehicle curve safety prediction system based on composite wheel-rail force, comprising:

[0049] A coefficient acquisition module is used to set a number of detection points on the target curve on the track and obtain the vertical stiffness coefficient, lateral stiffness coefficient and lateral force influence coefficient of each detection point;

[0050] A displacement acquisition module is used to acquire the vertical displacement and lateral displacement of each detection point when the rail vehicle passes through each detection point;

[0051] A composite wheel-rail force acquisition module, used to calculate the composite wheel-rail force of each detection point by using the vertical displacement and lateral displacement of each detection point and the vertical stiffness coefficient, lateral stiffness coefficient and lateral force influence coefficient corresponding to the detection point; the composite wheel-rail force includes vertical force and lateral force;

[0052] A derailment coefficient acquisition module is used to calculate the derailment coefficient of each detection point using the composite wheel-rail force of the detection point;

[0053] The prediction module is used to predict the safety of the target curve based on the derailment coefficients of all detection points on the target curve.

[0054] The present invention is beneficial in that:

[0055] The present invention sets several detection points on the target curve and obtains the vertical stiffness coefficient, lateral stiffness coefficient and lateral force influence coefficient of the detection points. Then, the composite wheel-rail force at the detection point is inferred by the vertical displacement and lateral displacement of the track at the detection point when the rail vehicle passes through the detection point. Finally, the derailment coefficient of the detection point is calculated using the composite wheel-rail force, and the safety of the entire target curve is predicted and evaluated using the derailment coefficient.

[0056] The present invention does not rely on a neural network, but instead uses an experimental method to find the relationship between displacement and composite wheel-rail force, and finally infers the composite wheel-rail force exerted on the track by measuring the displacement of the track. Compared with the prior art, the present invention is more accurate and does not require the collection of a large amount of data in advance, and is more suitable for engineering applications.

[0057] The present invention introduces a lateral force influence coefficient, which reveals the influence of the lateral force on the vertical displacement, thereby making the final detection result closer to the actual situation. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is a flow chart of Example 1 of the present invention.

[0059] Figure 2 This is a structural diagram of embodiment 2 of the present invention. DETAILED DESCRIPTION

[0060] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is further described in detail below in conjunction with specific implementation methods.

[0061] Example 1

[0062] like Figure 1As shown, the present invention provides a method for predicting the safety of a railway vehicle on a curve based on a composite wheel-rail force, comprising:

[0063] S1. Set a number of detection points on the target curve on the track, and obtain the vertical stiffness coefficient, lateral stiffness coefficient and lateral force influence coefficient of each detection point.

[0064] The setting of the detection points can be adjusted according to the actual situation. It is foreseeable that the denser the detection points are, the more accurate the final result will be, but of course the cost will be higher. In this embodiment, five detection points are set between the sleepers, which are respectively distributed at two sleepers, the midpoint between two sleepers, and 1 / 4 and 3 / 4. The above five detection points are representative and can truly reflect the stress condition of the track.

[0065] The calculation method of the vertical stiffness coefficient in this embodiment includes:

[0066] A test vertical force is applied to the detection point, and a first test vertical displacement of the detection point generated by the test vertical force is obtained. The vertical stiffness coefficient of the detection point is calculated using the first test vertical displacement and the test vertical force. Specifically, the formula is used:

[0067] KPi=Pi / WPi

[0068] Calculate the vertical stiffness coefficient; where KPi is the vertical stiffness coefficient of the i-th detection point, Pi is the test vertical force of the i-th detection point, and WPi is the first test vertical displacement of the i-th detection point.

[0069] In this embodiment, a force-applying device is used to apply a test vertical force at the detection point, and the vertical displacement of the rail at the detection point is recorded, so as to calculate the vertical stiffness coefficient of the detection point.

[0070] The above method has high accuracy and low cost when the vertical force and vertical displacement are relatively proportional. For the case where the vertical force and vertical displacement are not proportional, in this embodiment, a plurality of different test vertical forces are applied to the detection point, and a plurality of first test vertical displacements generated by the detection point under the action of the different test vertical forces are obtained; a plurality of first test vertical displacements and a plurality of test vertical forces are used to fit a vertical stiffness function as the vertical stiffness coefficient of the detection point.

[0071] The relationship between vertical force and vertical displacement is obtained by fitting function, and the result is more accurate, but the acquisition cost is higher. In engineering, the above two methods can be selected according to specific circumstances.

[0072] In addition, a test lateral force is applied to the detection point, and the test lateral displacement and the second test vertical displacement of the detection point under the test lateral force are obtained. The lateral stiffness coefficient and the lateral force influence coefficient of the detection point are calculated using the test lateral displacement, the second test vertical displacement and the test lateral force.

[0073] In this embodiment, the calculation method of the lateral stiffness coefficient and the lateral force influence coefficient includes:

[0074] Apply a test lateral force to the detection point, and obtain the test lateral displacement and the second test vertical displacement of the detection point under the test lateral force, and calculate the lateral stiffness coefficient and lateral force influence coefficient of the detection point using the test lateral displacement and the second test vertical displacement as well as the test lateral force. Specifically, use the formula

[0075] KQi=Qi / WQi

[0076] Calculate the lateral stiffness coefficient; where KQi is the lateral stiffness coefficient of the i-th detection point, Qi is the test lateral force of the i-th detection point, and WQi is the test lateral displacement of the i-th detection point;

[0077] The method for calculating the lateral force influence coefficient using the second test vertical displacement and the test lateral force is to use the formula

[0078] Ui=Qi / WPi2

[0079] Calculate the lateral force influence coefficient; where Ui is the lateral force influence coefficient of the i-th detection point, Qi is the test lateral force of the i-th detection point, and WPi2 is the second test vertical displacement of the i-th detection point.

[0080] In this embodiment, a force-applying device is used to apply a test lateral force at the detection point, and the lateral displacement and vertical displacement of the rail at the detection point are recorded, so as to calculate the lateral stiffness coefficient and lateral force influence coefficient of the detection point. It is worth noting that in the process of applying a lateral force to the track detection point, the track will not only produce a lateral displacement, but also a vertical displacement. Therefore, the concept of the lateral force influence coefficient is introduced in the present invention, so that the final detection result is closer to the actual situation.

[0081] Similar to the vertical stiffness coefficient, the above method is applicable to the case where the lateral force is proportional to the lateral displacement and the vertical displacement under the influence of the lateral force.

[0082] When the lateral force is not proportional to the lateral displacement and the vertical displacement under the influence of the lateral force, several test lateral forces of different magnitudes are applied to the detection point, and several test lateral displacements and several second test vertical displacements of the detection point under the action of test lateral forces of different magnitudes are obtained. The lateral stiffness function and the lateral force influence function are fitted using the several test lateral displacements, several second test vertical displacements and several test lateral forces as the lateral stiffness coefficient and lateral force influence coefficient of the detection point.

[0083] The relationship between the lateral force and the lateral displacement as well as the vertical displacement under the influence of the lateral force is also obtained by the fitting function method.

[0084] S2. Obtain the vertical displacement and lateral displacement of each detection point when the rail vehicle passes through each detection point.

[0085] After obtaining the vertical stiffness coefficient, lateral stiffness coefficient and lateral force influence coefficient of each detection point, the vertical displacement and lateral displacement of each detection point when the rail vehicle passes by are recorded.

[0086] S3. Calculate the composite wheel-rail force of each detection point using the vertical displacement and lateral displacement of the detection point and the vertical stiffness coefficient, lateral stiffness coefficient and lateral force influence coefficient corresponding to the detection point; the composite wheel-rail force includes vertical force and lateral force.

[0087] In this step, the vertical displacement and lateral displacement of each detection point when the rail vehicle passes by, as well as the vertical stiffness coefficient, lateral stiffness coefficient and lateral force influence coefficient of the detection point, are used to reversely infer the force conditions at the track detection point when the rail vehicle passes by.

[0088] Specifically, in this embodiment, the calculation method of the composite wheel-rail force at the detection point includes:

[0089] Using the formula

[0090] Qit=WQit*KQi

[0091] Calculate the lateral force; where Qit is the lateral force at the i-th detection point, WQit is the lateral displacement at the i-th detection point, and KQi is the lateral stiffness coefficient at the i-th detection point.

[0092] Using the formula

[0093] Pit=[WPit-(Qit / Ui)]*KPi

[0094] Calculate the vertical force; where Pit is the vertical force at the i-th detection point, WPit is the vertical displacement at the i-th detection point, KPi is the vertical stiffness coefficient at the i-th detection point, Qit is the lateral force at the i-th detection point, and Ui is the lateral force influence coefficient at the i-th detection point.

[0095] In the case of fitting functions, the composite wheel-rail force can also be calculated by substituting the lateral displacement and vertical displacement of the detection point into the corresponding function.

[0096] S4. Calculate the derailment coefficient of each detection point using the composite wheel-rail force of the detection point.

[0097] After obtaining the composite wheel-rail force at each test point, the formula

[0098] D i =Qit / Pit

[0099] Calculate the derailment coefficient; where D i is the derailment coefficient of the ith detection point, Pit is the vertical force of the ith detection point, and Qit is the lateral force of the ith detection point.

[0100] S5. Predict the safety of the target curve based on the derailment coefficients of all detection points on the target curve.

[0101] In this embodiment, firstly use the formula

[0102]

[0103] Calculate the safety index of the target curve; is the safety index, D i is the derailment coefficient of the ith detection point, w i is the weight of the i-th detection point, and N is the number of detection points.

[0104] In a curve, the greater the curvature, the higher the probability of derailment. Therefore, the weight of each detection point in the safety evaluation of the entire curve is different. It can be understood that the greater the curvature of the detection point, the higher its weight.

[0105] In this embodiment, the formula

[0106]

[0107] Calculate the weight of the detection point; where wi is the weight of the i-th detection point, is the local curvature of the ith detection point, is the maximum curvature of the target curve, k is the adjustment coefficient, and n is the adjustment index. k and n can be adjusted according to actual conditions. k is used to adjust the overall scale of the weight, and n is used to adjust the degree of influence of curvature on the weight.

[0108] In the above formula, the local curvature of the i-th detection point is defined as for

[0109]

[0110] Among them, L i is the arc length of the i-th detection point, R i is the radius of the i-th detection point.

[0111] is the maximum curvature in the entire target curve. It is necessary to measure the curvature of all detection points in the target curve and select the largest curvature as .

[0112] Secondly, the safety index is compared with the safety threshold to obtain the safety level of the target completion. The method for evaluating the safety level in this embodiment is:

[0113] exist When ≤0.5Dt, the target curve is low risk;

[0114] At 0.5Dt< When ≤Dt, the target curve is of medium risk;

[0115] exist >Dt, the target curve is high risk; is the safety index, and Dt is the safety threshold.

[0116] Example 2

[0117] like Figure 2 As shown, the present invention also provides a rail vehicle curve safety prediction system based on composite wheel-rail force, comprising:

[0118] A coefficient acquisition module is used to set a number of detection points on the target curve on the track and obtain the vertical stiffness coefficient, lateral stiffness coefficient and lateral force influence coefficient of each detection point;

[0119] A displacement acquisition module is used to acquire the vertical displacement and lateral displacement of each detection point when the rail vehicle passes through each detection point;

[0120] A composite wheel-rail force acquisition module, used to calculate the composite wheel-rail force of each detection point by using the vertical displacement and lateral displacement of each detection point and the vertical stiffness coefficient, lateral stiffness coefficient and lateral force influence coefficient corresponding to the detection point; the composite wheel-rail force includes vertical force and lateral force;

[0121] A derailment coefficient acquisition module is used to calculate the derailment coefficient of each detection point using the composite wheel-rail force of the detection point;

[0122] The prediction module is used to predict the safety of the target curve based on the derailment coefficients of all detection points on the target curve.

[0123] The above are only preferred embodiments of the present invention. It should be noted that the above preferred embodiments should not be regarded as limiting the present invention, and the protection scope of the present invention should be based on the scope defined by the claims. For ordinary technicians in this technical field, several improvements and modifications can be made without departing from the spirit and scope of the present invention, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. A method for predicting the safety of railway vehicles on curves based on composite wheel-rail forces, characterized in that include: A plurality of detection points are set on the target curve on the track, and the vertical stiffness coefficient, the lateral stiffness coefficient and the lateral force influence coefficient of each detection point are obtained; Obtain the vertical displacement and lateral displacement of each detection point when the rail vehicle passes through each detection point; The composite wheel-rail force of each detection point is calculated using the vertical displacement and lateral displacement of each detection point and the vertical stiffness coefficient, lateral stiffness coefficient and lateral force influence coefficient corresponding to the detection point; The composite wheel-rail force includes vertical force and lateral force; The calculation method of the composite wheel-rail force at the detection point includes: Using the formula Qit=WQit*KQi Calculate the lateral force; where Qit is the lateral force at the i-th detection point, WQit is the lateral displacement at the i-th detection point, and KQi is the lateral stiffness coefficient at the i-th detection point; Using the formula Pit=[WPit-(Qit / Ui)]*KPi Calculate the vertical force; where Pit is the vertical force at the i-th detection point, WPit is the vertical displacement at the i-th detection point, KPi is the vertical stiffness coefficient at the i-th detection point, Qit is the lateral force at the i-th detection point, and Ui is the lateral force influence coefficient at the i-th detection point; The derailment coefficient of each detection point is calculated using the composite wheel-rail force of the detection point; The safety of the target curve is predicted based on the derailment coefficients of all detection points on the target curve.

2. A method for predicting the safety of a railway vehicle on a curve based on composite wheel-rail force according to claim 1, characterized in that Methods for obtaining the vertical stiffness coefficient, lateral stiffness coefficient and lateral force influence coefficient of the detection point include: Apply a test vertical force to the detection point, obtain a first test vertical displacement of the detection point under the test vertical force, and calculate a vertical stiffness coefficient of the detection point using the first test vertical displacement and the test vertical force; A test lateral force is applied to the detection point, and a test lateral displacement and a second test vertical displacement of the detection point under the test lateral force are obtained; the lateral stiffness coefficient and the lateral force influence coefficient of the detection point are calculated using the test lateral displacement, the second test vertical displacement and the test lateral force.

3. A method for predicting the safety of a railway vehicle on a curve based on composite wheel-rail force according to claim 2, characterized in that: The method for calculating the vertical stiffness coefficient of the test point using the first test vertical displacement and the test vertical force is to use the formula KPi=Pi / WPi Calculate the vertical stiffness coefficient; where KPi is the vertical stiffness coefficient of the i-th detection point, Pi is the test vertical force of the i-th detection point, and WPi is the first test vertical displacement of the i-th detection point; The method for calculating the lateral stiffness coefficient using the test lateral displacement and the test lateral force is to use the formula KQi=Qi / WQi Calculate the lateral stiffness coefficient; where KQi is the lateral stiffness coefficient of the i-th detection point, Qi is the test lateral force of the i-th detection point, and WQi is the test lateral displacement of the i-th detection point; The method for calculating the lateral force influence coefficient using the second test vertical displacement and the test lateral force is to use the formula Ui=Qi / WPi2 Calculate the lateral force influence coefficient; where Ui is the lateral force influence coefficient of the i-th detection point, Qi is the test lateral force of the i-th detection point, and WPi2 is the second test vertical displacement of the i-th detection point.

4. The method for predicting the safety of a railway vehicle on a curve based on composite wheel-rail force according to claim 1, characterized in that: Applying a plurality of test vertical forces of different magnitudes to the detection point, and obtaining a plurality of first test vertical displacements of the detection point under the action of the test vertical forces of different magnitudes; fitting a vertical stiffness function using the plurality of first test vertical displacements and the plurality of test vertical forces as a vertical stiffness coefficient of the detection point; Several test lateral forces of different magnitudes are applied to the detection point, and several test lateral displacements and several second test vertical displacements of the detection point under the action of the test lateral forces of different magnitudes are obtained. The lateral stiffness function and the lateral force influence function are fitted using the several test lateral displacements, the several second test vertical displacements and the several test lateral forces as the lateral stiffness coefficient and the lateral force influence coefficient of the detection point.

5. The method for predicting the safety of a railway vehicle on a curve based on composite wheel-rail force according to claim 1 is characterized in that The method for calculating the derailment coefficient of the detection point is to use the formula D i =Pit / Pit Calculate the derailment coefficient; where D i is the derailment coefficient of the ith detection point, Pit is the vertical force of the ith detection point, and Qit is the lateral force of the ith detection point.

6. The method for predicting the safety of a railway vehicle on a curve based on composite wheel-rail force according to claim 1 is characterized in that The method for predicting the safety of the target curve based on the derailment coefficients of all detection points on the target curve includes: Using the formula Calculate the safety index of the target curve; is the safety index, D i is the derailment coefficient of the ith detection point, w i is the weight of the i-th detection point, and N is the number of detection points; The safety index is compared with the safety threshold to obtain the safety level of target completion.

7. A method for predicting the safety of a railway vehicle on a curve based on composite wheel-rail force according to claim 6, characterized in that Using the formula Calculate the weight of the detection point; where w i is the weight of the i-th detection point, is the local curvature of the ith detection point, is the maximum curvature of the target curve, k is the adjustment coefficient, and n is the adjustment index.

8. The method for predicting the safety of a railway vehicle on a curve based on composite wheel-rail force according to claim 6, characterized in that: exist When ≤0.5Dt, the target curve is low risk; At 0.5Dt< When ≤Dt, the target curve is of medium risk; exist >Dt, the target curve is high risk; is the safety index, and Dt is the safety threshold.

9. A rail vehicle curve safety prediction system based on composite wheel-rail force, characterized in that include: A coefficient acquisition module is used to set a number of detection points on the target curve on the track and obtain the vertical stiffness coefficient, lateral stiffness coefficient and lateral force influence coefficient of each detection point; A displacement acquisition module is used to acquire the vertical displacement and lateral displacement of each detection point when the rail vehicle passes through each detection point; A composite wheel-rail force acquisition module is used to calculate the composite wheel-rail force of each detection point by using the vertical displacement and lateral displacement of each detection point and the vertical stiffness coefficient, lateral stiffness coefficient and lateral force influence coefficient corresponding to the detection point; The composite wheel-rail force includes vertical force and lateral force; The calculation method of the composite wheel-rail force at the detection point includes: Using the formula Qit=WQit*KQi Calculate the lateral force; where Qit is the lateral force at the i-th detection point, WQit is the lateral displacement at the i-th detection point, and KQi is the lateral stiffness coefficient at the i-th detection point; Using the formula Pit=[WPit-(Qit / Ui)]*KPi Calculate the vertical force; where Pit is the vertical force at the i-th detection point, WPit is the vertical displacement at the i-th detection point, KPi is the vertical stiffness coefficient at the i-th detection point, Qit is the lateral force at the i-th detection point, and Ui is the lateral force influence coefficient at the i-th detection point; A derailment coefficient acquisition module is used to calculate the derailment coefficient of each detection point using the composite wheel-rail force of the detection point; The prediction module is used to predict the safety of the target curve based on the derailment coefficients of all detection points on the target curve.

Images