A train speed adjustment method

Abstract

A train speed adjustment method, when satellite positioning speed measurement is effective, satellite positioning speed measurement data is used to set the wheel / train speed ratio adjustment model parameters and the train radar speed adjustment model parameters; when satellite positioning speed measurement is not effective, the train radar speed adjustment model parameters obtained by setting before are used to calculate new train radar speed adjustment model parameters according to a given expression, the wheel / train speed ratio adjustment model parameters are set by the adjusted radar speed adjustment model parameters, and then the train speed, the creep rate change rate, the creep and the train wheel speed change rate and other train speed related quantities are calculated according to the wheel / train speed ratio adjustment model. The method combines the advantages of high accuracy of satellite positioning speed measurement and good real-time performance and long-term normal operation of radar speed measurement, and improves the accuracy and reliability of measuring various train speed related quantities.

Description

[0001] This invention patent application is a divisional application. The original application number is 202310475248.2, the application date is April 27, 2023, and the invention title is "A Train Maximum Traction Control System". Technical Field

[0002] This invention belongs to the field of train traction control technology, and in particular relates to a method for adjusting train speed. Background Technology

[0003] Train operation relies on the interaction between wheels and rails. Only when effective adhesion between wheels and rails is ensured can the power of the traction motor be further utilized. Wheel-rail adhesion characteristics are related not only to the train itself and the wheel and rail materials, but also to a series of uncertain factors that change with time and space, such as track conditions and rail surface cleanliness. If the traction force during locomotive operation exceeds the available adhesion between wheels and rails, the excess traction force will accelerate wheel spinning, rapidly increasing the relative sliding speed and quickly reducing the available adhesion. This will cause wear and even damage to the wheels and rails, increasing railway maintenance costs and threatening the safe operation of the locomotive. Because locomotive operating conditions are highly variable, changes in driver operation or deterioration of track conditions during traction can lead to slippage that cannot be completely avoided. Currently, domestic AC / DC locomotives mainly employ a combined correction method for slippage and anti-skid control. This method first assesses wheel acceleration; if acceleration exceeds a certain threshold, it indicates severe slippage, and the driving torque of the moving wheels is rapidly and deeply reduced, thus decreasing the locomotive's traction. If wheel acceleration does not exceed the threshold, creep speed is assessed; if creep speed exceeds the threshold, the driving torque is adjusted significantly; otherwise, it is considered normal operation. The current combined correction method uses two or more individual threshold conditions to determine slippage. When slippage has not occurred, it cannot comprehensively assess the risk of slippage; when slippage has already occurred, it cannot comprehensively assess the degree of slippage.

[0004] Due to creep, especially wheel spin, the train's wheelset speed does not match the actual train speed. Furthermore, when determining whether wheel spin has occurred and calculating creep rate and creep degree, the train's wheelset speed and train speed must be measured separately; the wheelset speed cannot be used to represent the train speed. Train speed is commonly measured using radar or satellite positioning. Satellite positioning speed measurement uses satellite positioning to track the train's speed and position in real time, transmitting this information to the train control system for processing to obtain the final train speed. Satellite positioning speed measurement can overcome errors caused by wheel spin and slippage, but its accuracy is greatly affected by weather and terrain, and it cannot achieve 100% speed measurement. Data transmission delays exist, and these delays are variable due to distance and ionospheric conditions, affecting the real-time performance of speed measurements. Radar speed measuring devices are typically installed under the train, with the radar antenna at a certain angle to the ground. The radar waves are emitted in the direction of the train. When the train moves relative to the ground, the received radar waves will experience a frequency shift. This shift is determined based on the radar wavelength, the amount of frequency shift, and the angle. The train speed can be calculated by solving data such as the radar installation height; however, the included angle... Data such as radar installation height may fluctuate over time, and the road conditions on trains are not consistent, so the radar installation height may also change with the road conditions, which will affect the accuracy of radar speed measurement. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by providing a method for adjusting train speed V. C (h) According to formula

[0006]

[0007] Adjustments and calculations are performed, and the train speed V is taken as the train speed V. C (h); where W(h) is the train radar speed, P W Radar speed adjustment coefficient; Train speed V C The adjustment calculation period for (h) and the sampling period for the train radar speed W(h) are both T. V .

[0008] The radar velocity adjustment coefficient P is obtained by iterative calculation. W The creep rate of change x1, creep x2, and train wheelset speed rate of change x3 are determined by the following method:

[0009] Step 1: Read the vehicle-mounted satellite positioning system speed U(k) and positioning status information X(k) at the time of the kth iteration calculation;

[0010] Step 2: Read the train wheel rotation speed V(k) and train radar speed W(k) collected at the synchronous acquisition time point of the vehicle-mounted satellite positioning system speed U(k);

[0011] Step 3: Determine if the vehicle-mounted satellite positioning system speed is valid; if the vehicle-mounted satellite positioning system speed is valid, proceed to step 4; if the vehicle-mounted satellite positioning system speed is invalid, proceed to step 5.

[0012] Step 4, according to the formula

[0013]

[0014] Adjust the current wheel / vehicle speed adjustment coefficient P V (k) and radar velocity ratio coefficient P W (k); Let the radar velocity adjustment coefficient P W equals P W (k), proceed to step 6;

[0015] Step 5, according to the formula

[0016]

[0017] Calculate the current radar velocity ratio coefficient P W (k); according to formula

[0018]

[0019] Calculate the radar velocity adjustment coefficient P W According to the formula

[0020]

[0021] Calculate the radar synchronization adjustment speed W * (k); according to formula

[0022]

[0023] Adjust the current wheel / vehicle speed adjustment coefficient P V (k); Proceed to step 6;

[0024] Step 6, for m points (k, P) V (k)), (k-1,P V (k-1)), ..., (k-m+1, P V (k-m+1) is used to perform linear fitting to obtain the first-order fitted line of the wheel / vehicle speed adjustment coefficient. The point (k, U) on the first-order fitted line of the wheel / vehicle speed adjustment coefficient is taken. V The value of (k)U V (k) is the wheel / vehicle speed ratio coefficient; according to the formula

[0025]

[0026] Calculate the creep coefficient x2; according to the formula

[0027]

[0028] Calculate the rate of change of creep x1; ​​according to the formula

[0029]

[0030] Calculate the train wheelset speed change rate x3, where V(h) is the train wheel rotation speed.

[0031] The sampling period of V(h) is T V The calculation period for the train wheelset speed change rate x3 is T. V The sampling period for the vehicle-mounted satellite positioning system's speed U(k) and positioning status information X(k) is T. U The calculation period for creep coefficient x2 and creep coefficient change rate x1 is T. U ;T V Less than T U m is an integer greater than or equal to 3.

[0032] Radar velocity weighting coefficients satisfy the formula

[0033]

[0034] The relationship is as follows: the value of each radar velocity weighting coefficient satisfies the relationship that the smaller i is, the larger the radar velocity weighting coefficient is.

[0035] The time point at which the train wheel rotation speed is collected τ times before the sampling time of the vehicle-mounted satellite positioning system's speed U(k) is the synchronous acquisition time point of the vehicle-mounted satellite positioning system's speed U(k); τ is the number of delay interval periods, which is calculated by satisfying the following...

[0036]

[0037] Given the relationship and that the speed of the vehicle-mounted satellite positioning system has been valid in the most recent m1 consecutive judgments, the number of delay interval periods τ is calculated; where β(ki) is the rate of change of train acceleration in the most recent m1 judgments, and ε is the acceleration change threshold greater than 0. The rate of change of train acceleration is calculated according to the formula...

[0038]

[0039] The calculation is performed; where α(k) is the most recently collected train acceleration, and α(k-1) is the previously collected train acceleration. Train acceleration is calculated according to the formula...

[0040]

[0041] The calculation is performed; where U(k-1) is the speed of the vehicle-mounted satellite positioning system in the previous acquisition of U(k).

[0042] Let the parameter to be optimized be the number of delay intervals, τ. * Radar speed adjustment coefficient p W * and rotational speed adjustment coefficient p V * The number of delay intervals is τ. * At that time, the train wheel rotation speed collected at the synchronous acquisition time point corresponding to U(ki) is V. * (ki), the train radar speed acquired at the synchronous acquisition time point corresponding to U(ki) is W. * (ki), the minimum optimization objective function is

[0043]

[0044] Take the number of delay intervals τ that satisfy the optimal value Q. * Let τ be the number of delay interval periods; * The value range is greater than 0 and less than 2 / T V integers, p W * and p V * The values ​​of m1 are all greater than or equal to 0.8 and less than or equal to 1.2; m1 is greater than or equal to 10.

[0045] The method for determining whether the speed of the vehicle-mounted satellite positioning system is valid is as follows: when the positioning status in the positioning status information X(k) is valid and the number of satellites using the position calculation in the positioning status information X(k) is greater than or equal to δ, the speed of the vehicle-mounted satellite positioning system is valid; otherwise, the speed of the vehicle-mounted satellite positioning system is invalid; the value of δ is required to be greater than or equal to 4.

[0046] The creep rate of change x1, creep rate of change x2, and train wheelset speed rate of change x3 output by the train speed adjustment method are used for train wheelset slip traction control. The method is as follows:

[0047]

[0048] Calculate the idling risk value E; where θ1 is the creep rate of change threshold, θ2 is the creep rate of change threshold, θ3 is the wheelset speed rate of change threshold; τ is the nonlinear weighting exponent, γ1, γ2, and γ3 are nonlinear weighting factors, and τ≥2, γ1≥1, γ2≥1, and γ3≥1.

[0049] The method for determining whether a train wheelset is spinning is as follows: when the spinning risk value E is greater than or equal to 1, the train wheelset is spinning. Spinning traction control is achieved by controlling the spinning traction control ratio θ. The spinning traction control ratio θ is the ratio between the train traction force output by the spinning traction control module and the input train traction force, and 0 ≤ θ ≤ 1. The spinning traction control process is as follows:

[0050] Process I, the process of reducing idling traction, begins when the idling risk value E is greater than or equal to 1 and continues to increase, and ends when the idling risk value E changes from continuously increasing to starting to continuously decreasing; in Process I, the control θ begins to decrease with a slope d1, and the value of θ at the end of Process I is the minimum maintenance value; the minimum maintenance value of θ is not less than 0.

[0051] Process II, the process of maintaining the minimum traction force during idling, starts from the end of Process I and ends when the idling risk value E is less than 1; during Process II, the idling risk value E continuously decreases, and the control θ is equal to the minimum maintenance value;

[0052] Process III, the idling traction recovery process, starts from the end of Process II and ends when θ increases to equal 1; in Process III, the idling traction control module controls θ to increase with a slope d2 until θ equals 1; the rate of decrease of slope d1 is greater than the rate of increase of slope d2.

[0053] The train speed V output by the aforementioned train speed adjustment method is used for upper limit control of train traction force. The method involves establishing a maximum traction force limiting neural network model through two rounds of training. The input to the maximum traction force limiting neural network model is the train speed V and the track state C1, and the output is the maximum traction force limit value F. μm .

[0054] The method for the first round of training and modeling is to set the input track state to dry, and to limit the neural network model to a function based on the maximum traction force.

[0055]

[0056] To approximate the target, we perform training to obtain an empirical formula approximation model, where V is the train speed input to the model, and F... μm P is the maximum traction force limit value output by the model. μ To calculate the adhesive weight, a1, a2, a3, a4, and a5 are empirical formula parameters for calculating the adhesion coefficient.

[0057] The second round of training and modeling involves using collected sample data of train speed, track conditions, and corresponding maximum traction force limits to train the model based on empirical formulas. After training, the model parameters are fixed to obtain the maximum traction force limit neural network model.

[0058] The F output of the neural network model is limited by the maximum traction force. μm The train's traction force is subject to upper limit and amplitude control.

[0059]

[0060] Among them, F1 is the train traction force before the upper limit control, and F2 is the train traction force after the upper limit control.

[0061] The train speed adjustment method is used in the train maximum traction force control system. The train maximum traction force control system includes a maximum traction force limiting neural network model, a traction force limiting module, and an idle traction force control module. The train traction force output by the train speed controller is controlled by the maximum traction force limiting value output by the maximum traction force limiting neural network model through the traction force limiting module to control the upper limit amplitude. Then, the idle traction force control module controls the train traction force to perform idle traction force control, that is, to perform train traction force load reduction control.

[0062] The train maximum traction control system also includes a track surface monitoring module for classifying, identifying, and outputting track conditions. The track surface monitoring module includes a track surface image acquisition unit and a track surface image recognition unit. The track surface image recognition unit processes the real-time track surface images acquired by the track surface image acquisition unit and outputs the current track condition. Track conditions include three states: dry, wet, and sandy. The train maximum traction control system also includes a train speed adjustment processing module, which implements the train speed adjustment method. In the train speed adjustment processing module, the sampled train wheel rotation speed is filtered to obtain the acquired train wheel rotation speed; the sampled train radar speed is filtered to obtain the acquired train radar speed; and the sampled onboard satellite positioning system speed is filtered to obtain the acquired onboard satellite positioning system speed. Before acquiring the first onboard satellite positioning system speed, let...

[0063]

[0064] Where i = 1, 2, ..., m-1.

[0065] The beneficial effects of this invention are as follows: The maximum traction force limiting neural network model is modeled in a two-round manner. Based on an empirical formula for calculating the adhesion coefficient derived from extensive experimental data, this model takes into account both the actual operating sections and changing road conditions, allowing the maximum traction force limit to change in real time with changes in road conditions, thus traction is performed as much as possible without wheelset slippage. When adhesion conditions worsen, even with upper limit limiting, if the locomotive traction force (wheel circumferential tangential force) on the wheel axle is still greater than the wheel-rail adhesion force, and wheelset slippage cannot be avoided, this invention uses a nonlinear mathematical model to calculate the slippage risk value E to quickly restore normal train traction. It integrates multiple individual threshold judgment conditions for traditional wheelset slippage and weighted judgment conditions when none of the individual threshold conditions are met into a single whole, simplifying the judgment criteria. Furthermore, when none of the individual threshold conditions are met, multiple factors are quantified and weighted for calculation, achieving a comprehensive judgment of multiple factors, making slippage judgment more comprehensive and accurate. The selection of a nonlinear mathematical model can minimize the possibility of misjudgment in the weighted judgment conditions when none of the individual threshold conditions are met. Furthermore, the magnitude of the weighted judgment conditions and the relative magnitude of each weighted term can be set and adjusted through parameters, making the train wheelset slip judgment method normalized by this nonlinear mathematical model applicable to different locomotive types and operating conditions. Attached Figure Description

[0066] Figure 1 This is a schematic diagram of the maximum traction force control system of a train.

[0067] Figure 2 Flowchart of the method for establishing a train traction force limiting model;

[0068] Figure 3 This diagram illustrates the traction control function of the traction control module when train wheelsets experience idling. Figure 1 ;

[0069] Figure 4 This diagram illustrates the traction control function of the traction control module when train wheelsets experience idling. Figure 2 ;

[0070] Figure 5 This is a schematic diagram of the train speed adjustment processing module.

[0071] Figure 6 Here is a flowchart of the train speed adjustment method;

[0072] Figure 7 A schematic diagram of the first-order fitted straight line for the wheel / vehicle speed ratio coefficient;

[0073] Figure 8A flowchart for calculating the number of delay intervals;

[0074] Figure 9 A schematic diagram illustrating the speed acquisition delay, train acceleration, and rate of change of train acceleration in a vehicle-mounted satellite positioning system.

[0075] Figure 10 A schematic diagram showing the time points at which the train wheel rotation speed and train radar speed are synchronously collected, representing the speed of the onboard satellite positioning system. Detailed Implementation

[0076] The invention will be further described below with reference to the accompanying drawings. The train is a train pulled by an electric locomotive.

[0077] Figure 1 This is a schematic diagram of the train's maximum traction force control system, including a maximum traction force limiting neural network model 10, a traction force limiting module 11, an idle traction force control module 12, a train speed adjustment and processing module 13, and a track surface monitoring module 14. F1 represents the train traction force output from the train speed controller (i.e., the locomotive speed controller). μm To limit the maximum traction force output by the neural network model, the traction force limiting module is based on F. μm The train traction force F1 before the upper limit control is applied is then subjected to upper limit control, and the output is the train traction force F2 after the upper limit control is applied. Specifically, there are...

[0078]

[0079] The maximum traction force limit neural network model takes the train speed V and track state C1 as inputs and outputs the maximum traction force limit F of the train. μm The method for establishing a train traction force limiting model, i.e., a maximum traction force limiting neural network model, is to use two rounds of training and modeling.

[0080] Figure 2 The flowchart for establishing a train traction force limiting model is as follows. The first step is to determine the neural network model structure. The track state output by the track surface monitoring module can be a single-parameter vector, where C1 is a scalar. For the identified dry, wet, and sandy states, C1 is equal to three constants respectively; for example, let C1 correspond to 2, 1, and 3 for dry, wet, and sandy states respectively. In this case, the maximum traction force limiting neural network model has two input parameters: C1 and V, and the output is F. μmThe maximum traction force limiting neural network model can be either a 2-5-1 three-layer BP neural network structure or a 2-7-1 three-layer BP neural network structure. During the first round of training and modeling, C1 is set to the dry state, i.e., C1 = 2. When the track state is a single-parameter vector, C1 can also be set to other constant values ​​for the dry, wet, and sandy states of the recognition results. For example, C1 can be set to 0, 1, and -1 for the dry, wet, and sandy states respectively. In this case, during the first round of training and modeling, C1 = 0.

[0081] The track status output by the track surface monitoring module can also be a 3-parameter vector, i.e., C1 = (C 11 C 12 C 13 ), 3 parameters C 11 C 12 C 13 The identification results for dry, damp, and sand-covered conditions are described separately, with each result being a binary value of 1 or 0. For example, if the binary value is 1 or 0, then when the track surface condition identification result is dry, C... 11 C 12 C 13 The values ​​are 1, 0, and 0 respectively; when the rail surface condition identification result is wet, C 11 C 12 C 13 The values ​​are 0, 1, and 0 respectively; when the track surface condition identification result is sand spreading, C 11 C 12 C 13 The values ​​are 0, 0, and 1 respectively; or, the two values ​​are set to 1 and -1. When the rail surface condition identification result is dry, C 11 C 12 C 13 The values ​​are 1, -1, and -1 respectively; when the rail surface condition identification result is wet, C 11 C 12 C 13 The values ​​are -1, 1, and -1 respectively; when the track surface condition identification result is sand spreading, C 11 C 12 C 13 The values ​​are -1, -1, 1, etc. The maximum traction force limiting neural network model can choose a 4-7-1 3-layer BP neural network structure, a 4-9-1 3-layer BP neural network structure, or a 4-11-1 3-layer BP neural network structure. During the first round of training and modeling, C1 is set to a dry state; for example, if the two values ​​are set to 1 and 0, then let C... 11 C 12 C 13 The values ​​are 1, 0, and 0 respectively.

[0082] When the track state output by the track surface monitoring module is a single-parameter vector, the maximum traction force limiting neural network model can also be a 3-layer radial basis function neural network with a structure of 2-3-1, 2-4-1, or 2-5-1. When the track state output by the track surface monitoring module is a 3-parameter vector, the maximum traction force limiting neural network model can also be a 3-layer radial basis function neural network with a structure of 4-5-1, 4-6-1, or 4-7-1. The track surface image recognition unit classifies and outputs the real-time track surface image for dry, wet, and sand-covered states, which is a conventional image processing technique.

[0083] The second step is to determine the empirical formula to be approximated. The empirical formula is:

[0084]

[0085] Where V is the train speed input to the model; F μm It is the maximum traction force limit value output by the model; P μ The calculated adhesive weight of the train traction locomotive is a constant after the locomotive model of the traction train is determined; the expression in parentheses is the adhesion coefficient μ. j ,Right now

[0086]

[0087] a1, a2, a3, a4, and a5 are empirical formula parameters for calculating the adhesion coefficient. Their values ​​are related to the electric locomotive model. For example, for various domestically produced electric locomotives, a1 = 0.24, a2 = 12, a3 = 100, a4 = 8, and a5 = 0 respectively; for the 6K type electric locomotive, a1 = 0.189, a2 = 8.86, a3 = 44, a4 ​​= 1, and a5 = 0 respectively; for the 8G type electric locomotive, a1 = 0.28, a2 = 4, a3 = 50, a4 = 6, and a5 = -0.0006 respectively; and so on. The unit of train speed V is km / h; P μ and each traction force F μm The units for F1, F2, and F3 are kN; when needed, the traction force can also be converted into the corresponding torque.

[0088] The third step is to conduct the first round of training and modeling. The input track states are dry, wet, and sandy. The input track state is set to dry. The maximum traction force limits the neural network model to approximate the function of equation (2) as the target, and the empirical formula approximation model is obtained.

[0089] Fourth, a second round of training and modeling is conducted. Sample data is collected for different types of electric locomotives, including train speed, track conditions, and the corresponding maximum traction force limits. When a1, a2, a3, a4, and a5 from the first round of training and modeling are empirical formula parameters for various types of domestically produced electric locomotives, the collected sample data of train speed, track conditions, and corresponding maximum traction force limits for these domestically produced electric locomotives are used. The model is trained based on the empirical formula approximation model. After training, the model parameters are fixed, resulting in a neural network model for the maximum traction force limit of various types of domestically produced electric locomotives. Similarly, when a1, a2, a3, a4, and a5 from the first round of training and modeling are empirical formula parameters for the 6K type electric locomotive, the collected sample data of train speed, track conditions, and corresponding maximum traction force limits for the 6K type electric locomotive are used. The model is trained based on the empirical formula approximation model. After training, the model parameters are fixed, resulting in a neural network model for the maximum traction force limit of the 6K type electric locomotive. When the parameters a1, a2, a3, a4, and a5 used in the first round of training modeling are empirical formula parameters for the 8G electric locomotive, the collected sample data of the 8G electric locomotive's train speed, track condition, and corresponding maximum traction force limit values ​​are used. Training is then conducted based on the empirical formula approximation model. After training, the model parameters are fixed, resulting in the maximum traction force limit neural network model for the 8G electric locomotive. After training and modeling, the maximum traction force limit neural network model is input into the train speed adjustment processing module and the track surface monitoring module, which collect real-time data on the train speed and track condition. It outputs the maximum traction force limit value, which the traction force limit module then uses to control the upper limit of the train's traction force.

[0090] The idling traction control module uses an established nonlinear mathematical model to calculate the idling risk value E, which is calculated according to the formula...

[0091]

[0092] Calculations are performed. In equation (4), x1 is the creep rate of change, θ1 is the creep rate of change threshold; x2 is the creep, θ2 is the creep threshold; τ is the nonlinear weighting exponent, γ1 and γ2 are nonlinear weighting factors, and τ≥2, γ1≥1, and γ2≥1. The creep rate of change x1 and the creep degree x2 are both non-negative values. The idling judgment condition is that when E≥1, the train (locomotive) wheelset is judged to have idled. Combining equation (4) and the idling judgment condition, the idling judgment logic obtained by decomposition is: there are 3 situations (or one of the 3 conditions) that can be judged as the train (locomotive) wheelset having idled, namely, ① when the creep rate of change x1 is greater than or equal to the threshold θ1; ② or, when the creep degree x2 is greater than or equal to the threshold θ2; ③ or, when the creep rate of change x1 is less than the threshold θ1 and the creep degree x2 is less than the threshold θ2 and the idling risk value E is greater than or equal to 1. The first two conditions, ① and ②, are single-item threshold conditions. That is, when either x1 ≥ θ1 or x2 ≥ θ2, E is greater than or equal to 1, thus satisfying the conditions for the idle judgment. Condition ③ is a weighted judgment condition when neither of the single-item threshold conditions is met. The larger the value of τ, the greater the influence of the single-item threshold judgment, and the smaller the effect of the weighted judgment. When τ is sufficiently large, the main body of the idle judgment logic is the single-item threshold conditions ① and ②, and the effect of condition ③ is very small, or almost non-existent. For example, when γ1 and γ2 are both 1, and τ equals 200, 0.99... 200 Even if x1 / θ1 and x2 / θ2 are both 0.99, the idle risk value E is 0.268, which is less than 1 and cannot meet the idle judgment condition. The weighted judgment has almost no effect. The smaller the value of τ, the greater the weighted effect of condition ③. For example, when γ1 and γ2 are both 1 and τ is 2, if x1 / θ1 is 0.8, then when x2 / θ2 is 0.6, E is 1, which already meets the idle judgment condition. The nonlinear weighting factors γ1 and γ2 are used to determine the relative magnitude of the effects between the weighting terms and do not affect the judgment conditions for each individual term exceeding the threshold. The larger the values ​​of γ1 and γ2, the smaller the weighting effect of the corresponding judgment term; conversely, the smaller the values ​​of γ1 and γ2, the larger the weighting effect of the corresponding judgment term. For example, if γ1 is small and γ2 is large, then in the calculation of the idle risk value E in condition ③, the term x1 / θ1 plays a larger role in the weighted calculation than the term x2 / θ2, but the effects of the individual threshold conditions ① and ② remain unchanged. As long as either ① or ② reaches or exceeds the threshold, the idle judgment condition is still satisfied.

[0093] The idling risk value E or according to the formula

[0094]

[0095] Calculations are performed. In equation (5), x1 is the rate of change of creep, θ1 is the threshold of the rate of change of creep; x2 is the creep, θ2 is the creep threshold; x3 is the rate of change of train wheelset speed, θ3 is the threshold of the rate of change of wheelset speed; τ is the nonlinear weighting exponent, γ1, γ2, and γ3 are nonlinear weighting factors, and τ≥2, γ1≥1, γ2≥1, and γ3≥1. The rates of change of creep x1, creep x2, and train wheelset speed x3 are all non-negative values. The condition for judging slippage is that when E≥1, the train (locomotive) wheelset is judged to have slipped. Combining equation (5) and the idling judgment condition, the decomposed logic for judging train (locomotive) wheelset idling is as follows: There are four situations (or one of the four conditions must be met) that can be judged as idling, namely: ① when the creep rate of change x1 is greater than or equal to the threshold θ1; ② or when the creep x2 is greater than or equal to the threshold θ2; ③ or when the train wheelset speed rate of change x3 is greater than or equal to the threshold θ3; ④ or when the creep rate of change x1 is less than the threshold θ1, the creep x2 is less than the threshold θ2, the train wheelset speed rate of change x3 is less than the threshold θ3, and the idling risk value E is greater than or equal to 1. The first three conditions ①②③ are single-item threshold conditions, that is, when a single item satisfies x1≥θ1, or x2≥θ2, or x3≥θ3, E is greater than or equal to 1, which satisfies the idling judgment condition. Condition ④ is a weighted judgment condition when none of the individual threshold conditions are met; the larger the value of τ, the greater the influence of the individual threshold judgment, and the smaller the effect of the weighted judgment; when τ is large enough, the main body of the idle judgment logic is the individual threshold conditions ①②③, and the effect of condition ④ is very small, or almost non-existent; for example, when γ1, γ2, and γ3 are all 1, and τ equals 200, 0.99 200 Even if the ratio of each of the three judgment items to the corresponding threshold is 0.99, the idle risk value E is only 0.402, which is less than 1, and therefore cannot meet the idle judgment condition. When the value of τ is small, the weighting effect of condition ④ is greater. For example, when γ1, γ2, and γ3 are all 1 and τ equals 2, if x1 / θ1 equals 0.8, x2 / θ2 equals 0.6, and x3 / θ3 equals 0, the idle judgment condition can be met. The nonlinear weighting factors γ1, γ2, and γ3 are used to determine the relative magnitude of the effects between the weighting items and do not affect the judgment condition of each individual item exceeding the threshold. The larger the values ​​of γ1, γ2, and γ3, the smaller the weighting effect of the corresponding judgment item; conversely, the smaller the values ​​of γ1, γ2, and γ3, the greater the weighting effect of the corresponding judgment item. For example, if γ1 is small and γ2 and γ3 are large, then in the calculation of the idle risk value E in condition ④, the single term x1 / θ1 plays a greater role in the weighted calculation than terms x2 / θ2 and x3 / θ3. However, the role of the single threshold conditions ①②③ remains unchanged. As long as any one of ①②③ reaches or exceeds the threshold, the idle judgment condition is still satisfied.

[0096] The aforementioned values ​​for θ2 range from 0.005 to 0.05; θ1 ranges from 0.0001 / s to 0.005 / s; and θ3 ranges from 3 m / s. 2 ~30m / s 2 Between. The units of x1, x2, and x3 are the same as the units of θ1, θ2, and θ3, respectively.

[0097] The two judgment terms in equation (4), or the three judgment terms in equation (5), are all power functions with an exponent greater than or equal to 2. Their characteristic is that when the value to be judged does not exceed the corresponding threshold, the slope of the curve increases as the value to be judged increases. That is, the closer the relevant value is to the corresponding threshold, the greater the impact of its value change on the function term. For example, comparing x1 and θ1, the closer x1 is to θ1, the smaller the change in x1 can cause a larger change in the judgment term. This characteristic of the power function amplifies the effect of changes in the values ​​to be judged (i.e., x1, x2, x3) near the threshold, making it more sensitive near the threshold. Conversely, when the value to be judged is much smaller than the threshold, the sensitivity is reduced to minimize the possibility of misjudgment of the weighted judgment condition when none of the individual threshold conditions are met. In equations (4) and (5), the larger the value of τ, the greater the change in the slope of the curve when the value to be judged does not exceed the corresponding threshold, meaning it is more sensitive near the threshold and less sensitive further away from the threshold.

[0098] The nonlinear mathematical models (4) and (5) for calculating the risk value of wheel spin E both contain creep rate of change and creep term. Creep is the relative difference between the train wheelset speed and the train speed. Its value directly reflects how far the train wheelset is from wheel spin, or how far it has already wheeled. Creep rate of change is the speed at which creep changes. This value is related to both the train wheelset speed rate of change and the train speed rate of change. The larger the value, the higher the risk of wheel spin. Equation (5) also contains a train wheelset speed rate of change term. This term is similar to the creep rate of change. The larger the value, the higher the risk of wheel spin. However, the train wheelset speed rate of change is unrelated to the train speed change. The addition of this term can increase the predictability of wheel spin when the train speed is high. When calculating the idling risk value E, you can choose either formula (4) or formula (5) as needed. When choosing formula (5), since the effects of the creep rate of change and the train wheelset speed rate of change are similar, the magnitudes of γ1 and γ3 should be taken into consideration.

[0099] The nonlinear mathematical model for calculating the idling risk value E, namely Equation (4) or Equation (5), and the corresponding idling judgment conditions, integrate the traditional single-threshold judgment conditions for multiple wheelset idling and the weighted judgment conditions when none of the single-threshold conditions are met into a whole, simplifying the judgment basis. Furthermore, when none of the single-threshold conditions are met, multiple factors are quantified and weighted to achieve a comprehensive judgment of multiple factors, making the idling judgment more comprehensive and accurate. The selection of the nonlinear mathematical model can minimize the possibility of misjudgment by the weighted judgment conditions when none of the single-threshold conditions are met. At the same time, the magnitude of the weighted judgment conditions can be set and adjusted by parameters, and the relative magnitude of each weighted term can also be set and adjusted by parameters, making the train wheelset idling judgment method normalized by this nonlinear mathematical model applicable to different locomotive types and operating conditions.

[0100] Figure 3 This diagram illustrates the traction control function of the traction control module when train wheelsets experience idling. Figure 1 The traction control ratio θ is the ratio between the train traction force output by the traction control module and the input train traction force. In other words, the traction control ratio θ is the ratio between the train traction force F3 after traction control and the train traction force F2 before traction control. F3 and the input traction force F2 satisfy the following condition:

[0101] F3=θ·F2 0≤θ≤1 (6)

[0102] The relationship. Figure 3 Before t1, the slip risk value E is less than 1, the train wheelset does not slip, and the slip traction control ratio θ is equal to 1. The slip traction control process of the slip traction control module is as follows:

[0103] Process I, the process of decreasing traction force during idling; starting from when the idling risk value E is greater than or equal to 1 and continues to increase, until the idling risk value E changes from continuously increasing to beginning to decrease, that is, from... Figure 3 The process begins at time t1 and ends at time t2. During process I, the idling traction control module controls θ to decrease at a slope d1, and the value of θ at the end of process I is the minimum maintenance value. The minimum maintenance value of θ is not less than 0.

[0104] Process II involves maintaining the minimum idling traction force value; starting from the end of Process I, the idling risk value E continuously decreases until it falls below 1, at which point the process ends. Figure 3 The process begins at time t2 and ends at time t3; during process II, the idling traction control module controls θ to be equal to the minimum maintenance value.

[0105] Process III, the traction recovery process during idling; it begins at the end of Process II and ends when θ increases to equal 1, i.e., from... Figure 3 The process begins at time t3 and ends at time t4. In process III, the idling traction control module controls θ to increase with a slope d2 until θ equals 1. The rate of decrease of slope d1 is selected between 0.3 / s and 2 / s. For example, if the rate of decrease of slope d1 is selected as 0.5 / s, then θ will decrease by 50% in 1 second, which could be from 100% to 50% in 1 second, or from 80% to 30% in 1 second, and so on. The rate of increase of slope d2 is selected between 0.05 / s and 0.5 / s. For example, if the rate of increase of slope d2 is selected as 0.2 / s, then θ will increase by 20% in 1 second, which could be from 40% to 60% in 1 second, or from 50% to 70% in 1 second, and so on. When determining d1 and d2, the rate of decrease (absolute value) of slope d1 is greater than the rate of increase (absolute value) of slope d2.

[0106] When the idling risk value E increases from less than 1 to greater than or equal to 1, the condition that the idling risk value E is greater than or equal to 1 and continues to increase is met. When θ equals 1 and the idling risk value E is continuously less than 1, the idling traction control module does not implement idling traction control.

[0107] Figure 4 This diagram illustrates the traction control function of the traction control module when train wheelsets experience idling. Figure 2 In process II, if the idling risk value E changes from continuously decreasing to continuously increasing, then return to process I for idling traction control; if... Figure 4 In process t5, the idling risk value E changes from continuously decreasing to continuously increasing, and the idling traction control module immediately returns from process II to process I. In process III, if the idling risk value E increases again to a level greater than or equal to 1, it returns to process I for idling traction control; if... Figure 4 At time t6, the idling risk value E increases again to greater than or equal to 1, and the idling traction control module immediately returns from process III to process I.

[0108] Currently, commonly used combined correction methods in China employ a fixed torque unloading strategy regardless of the degree of idling, failing to consider the wheel-rail adhesion state during unloading. This leads to several problems: first, insufficient unloading depth results in incomplete suppression of idling; second, excessive unloading depth causes locomotive traction loss; and third, unloading only stops when acceleration or creep rate falls below a set threshold, potentially resulting in excessive unloading depth. The idling traction control module of this invention controls idling traction based on an idling risk value that incorporates multi-factor comprehensive judgment. Both the degree and process of locomotive traction reduction are controlled by the idling risk value, reflecting the wheel-rail adhesion state. This effectively avoids situations where insufficient unloading depth leads to incomplete suppression of idling, or excessive unloading depth results in locomotive traction loss. Unloading stops when the idling risk value changes from increasing to decreasing, further mitigating the consequences of excessive unloading depth. The non-linear characteristics of the idling risk value allow judgments with higher risks to have a more significant control effect.

[0109] Figure 5 This diagram illustrates the structure of the train speed adjustment processing module, or the train speed adjustment system, to implement the train speed adjustment method. The train wheel rotation speed acquisition unit 101 outputs the acquired train wheel rotation speed V(h) (including V(k)) to the speed adjustment calculation unit 104; the train radar speed acquisition unit 103 outputs the acquired train radar speed W(h) (including W(k)) to the speed adjustment calculation unit 104; the onboard satellite positioning system speed acquisition unit 102 acquires and outputs the onboard satellite positioning system speed U(k) and positioning status information X(k) to the speed adjustment calculation unit 104; the speed adjustment calculation unit 104, based on the input information, adjusts and calculates the wheel / vehicle speed ratio adjustment model parameters and the train radar speed adjustment model parameters, and outputs the train speed, creep, creep rate of change, and train wheelset speed rate of change. Specifically, the combination switch SW1 in the speed adjustment calculation unit 104 is controlled by the positioning status information X(k) input from terminal 5. When it is determined that the vehicle-mounted satellite positioning system speed is valid based on X(k), terminal 1 of the control combination switch SW1 is connected to terminals 2 and 3, and the parameters of the wheel / vehicle speed ratio adjustment model and the train radar speed adjustment model are adjusted by the vehicle-mounted satellite positioning system speed U(k). Terminal 4 is left floating, and the radar synchronization adjustment speed W output by the train radar speed adjustment model is... * (k) Not used at this time, i.e., W * (k) has no effect at this time. When it is determined that the speed of the vehicle-mounted satellite positioning system is invalid based on X(k), terminal 4 of control SW1 is connected to terminal 2. The train radar speed adjustment model recursively derives the parameters of the train radar speed adjustment model according to the given method. The train radar speed adjustment model adjusts the train radar speed value W(k) at the synchronous acquisition time point in the train radar speed value W(h) to obtain the radar synchronous adjustment speed W.* (k), speed W is adjusted synchronously by radar. * (k) To tune the parameters of the wheel / vehicle speed ratio adjustment model; terminals 1 and 3 are suspended, meaning that the vehicle-mounted satellite positioning system speed U(k) is not used (or is invalid) at this time, and the parameters of the train radar speed adjustment model are not tuned by external signals. The wheel / vehicle speed ratio adjustment model is adjusted and calculated based on the input train wheel rotation speed V(k) and train radar speed W(k), and outputs the train speed V to the maximum traction force limiting neural network model 10 and the train speed related quantities C2 to the idle traction control module 12. When the idle risk value E is calculated according to formula (4), the train speed related quantities C2 include the creep change rate x1 and creep x2; when the idle risk value E is calculated according to formula (5), the train speed related quantities C2 include the creep change rate x1, creep x2 and train wheelset speed change rate x3. Figure 5 The combination switch SW1 in the diagram is a schematic switch, which means that the direction of the signal flow is controlled according to x(k). In digital control, it is usually implemented by program branching.

[0110] In this embodiment of the train speed adjustment system, the acquisition period T of the train wheel rotation speed acquisition unit is... V The acquisition period T of the vehicle-mounted satellite positioning system's speed acquisition unit is 32ms. UFor 1 second, m equals 4. When outputting the train wheel rotation speed V(h) and train radar speed W(h), the corresponding speed acquisition unit has already performed corresponding filtering processing according to the specific situation in the speed sampling and data processing stage; for example, if the train wheel rotation speed V(h) is sampled using a pulse speed sensor (encoder), the jitter interference of the pulse edge and the high-frequency interference in the pulse transmission process are filtered out accordingly; if the train wheel rotation speed V(h) and train radar speed W(h) are directly output as analog or digital quantities, low-pass filtering, smoothing filtering, Kalman filtering, and other filtering methods can be used alone or in combination to filter out high-frequency interference, random interference, white noise interference, etc. The vehicle-mounted satellite positioning system speed acquisition unit includes one or more GNSS (Global Navigation Satellite System) receiving terminals, such as GPS, BeiDou, Galileo, and GLONASS system receiving terminals, and a corresponding receiving and processing module. The receiving and processing module receives information such as the number of satellites currently calculating the position, ground velocity (vehicle-mounted satellite positioning system speed), and the validity of the positioning status from one or more receiving terminals. Alternatively, it may also receive information such as longitude, latitude, UTC time, and altitude from one or more receiving terminals and calculate the vehicle-mounted satellite positioning system speed accordingly. The technical means employed in the train wheel rotation speed acquisition unit, train radar speed acquisition unit, and vehicle-mounted satellite positioning system speed acquisition unit are conventional techniques in this field.

[0111] Figure 6 This is a flowchart illustrating the train speed adjustment method in a train speed adjustment system, which involves setting and calculating the parameters of the wheel / vehicle speed ratio adjustment model, the train radar speed adjustment model, and the wheel / vehicle speed ratio coefficient and various train speed-related quantities. The iterative calculation cycle is the same as the acquisition cycle of the onboard satellite positioning system's speed acquisition unit. The specific steps for each iterative calculation are as follows:

[0112] Step 1, read the calculation results of the kth iteration (equivalent to kT). U The data from the vehicle-mounted satellite positioning system at the sampling time includes the vehicle-mounted satellite positioning system speed U(k) and positioning status information X(k);

[0113] Step 2: Read the train wheel rotation speed V(k) and train radar speed W(k) collected at the synchronous acquisition time point of the vehicle-mounted satellite positioning system speed U(k);

[0114] Step 3: Determine if the vehicle-mounted satellite positioning system speed is valid; if the vehicle-mounted satellite positioning system speed is valid, proceed to step 4; if the vehicle-mounted satellite positioning system speed is invalid, proceed to step 5.

[0115] Step 4: Adjust the model parameters based on the wheel / vehicle speed ratio and the radar speed, i.e., according to the formula...

[0116]

[0117] Adjust the current wheel / vehicle speed adjustment coefficient P V (k) and radar velocity ratio coefficient P W (k); Let the radar velocity adjustment coefficient P W equals P W (k), proceed to step 6;

[0118] Step 5: Calculate and adjust the radar velocity model parameters, i.e., according to the formula...

[0119]

[0120] Calculate the current radar velocity ratio coefficient P W (k); and according to formula

[0121]

[0122] Calculate the radar velocity adjustment coefficient P W According to the formula

[0123]

[0124] Calculate the radar synchronization adjustment speed W * (k); according to W * (k) Adjust the wheel / vehicle speed ratio model parameters, i.e., according to the formula

[0125]

[0126] Adjust the current wheel / vehicle speed adjustment coefficient P V (k); Proceed to step 6;

[0127] Step 6: Calculate the wheel / car speed ratio coefficient and calculate the speed-related quantities for each train. Calculate the wheel / car speed ratio coefficient U. V (k) There are two implementation examples; calculate the wheel / vehicle speed ratio coefficient U. V Example 1 of (k), according to formula

[0128]

[0129] Calculate the wheel / vehicle speed ratio coefficient U V (k). Calculate the wheel / vehicle speed ratio coefficient U. V (k) Example 2, for m points (k, P) V (k)), (k-1,P V(k-1)), ..., (k-m+1, P V (k-m+1) is used to perform linear fitting to obtain the first-order fitted line of the wheel / vehicle speed adjustment coefficient. The point (k, U) on the first-order fitted line of the wheel / vehicle speed adjustment coefficient is taken. V The value of (k)U V (k) is the wheel / vehicle speed ratio coefficient U V (k). Figure 7 This is a schematic diagram of the first-order fitted straight line for the wheel / vehicle speed adjustment coefficient. Figure 7 In the diagram, m equals 4, and the four "+" points from left to right are points (k-3, P). V (k-3)), (k-2,P) V (k-2)), (k-1,P V (k-1)), (k,P) V (k)), the point "o" on the first-order fitted line of the wheel / vehicle speed adjustment coefficient is the point (k, U). V (k)). Figure 7 For illustrative purposes, the coefficient values ​​of the four "+" points are not actual data. To make the illustration clearer, the errors are intentionally marked as larger, and the slope of the first-order fitted line is also intentionally marked as larger.

[0130] The positioning status information X(k) includes whether the positioning status is valid or invalid, and the number of satellites currently using the position calculation. In step 3 of the train speed adjustment method, method 1 for determining the validity of the onboard satellite positioning system speed is: if the positioning status in the positioning status information X(k) is valid, the onboard satellite positioning system speed is valid; otherwise, the onboard satellite positioning system speed is invalid. Alternatively, method 2 for determining the validity of the onboard satellite positioning system speed is: if the positioning status in both positioning status information X(k) and X(k-1) is valid, the onboard satellite positioning system speed is valid; otherwise, the onboard satellite positioning system speed is invalid. Alternatively, method 3 for determining the validity of the onboard satellite positioning system speed is: if the positioning status in the positioning status information X(k) is valid, and the number of satellites currently using the position calculation in the positioning status information X(k) is greater than or equal to δ, the onboard satellite positioning system speed is valid; otherwise, the onboard satellite positioning system speed is invalid. Alternatively, method 4 for determining whether the speed of the vehicle-mounted satellite positioning system is valid is as follows: when the positioning status in both positioning status information X(k) and X(k-1) is valid, and the number of satellites currently calculating the position in both positioning status information X(k) and X(k-1) is greater than or equal to δ, the speed of the vehicle-mounted satellite positioning system is valid; otherwise, the speed of the vehicle-mounted satellite positioning system is invalid.

[0131] X(k-1) is the vehicle-mounted satellite positioning system data read at time k-1 during the previous iteration calculation. In this embodiment, the vehicle-mounted satellite positioning system speed acquisition unit includes a GPS system receiving terminal and a corresponding receiving and processing module. Method 3 is used to determine whether the vehicle-mounted satellite positioning system speed is valid, with δ set to 5. Typically, δ is required to be greater than or equal to 4.

[0132] P in steps 4-6 V (k), or P when i equals 0 V (ki) represents the current wheel / vehicle speed adjustment coefficient. P is defined as follows: when i equals 1, 2, ..., m-1. V (k-1), P V (k-2), ..., P V (k-m+1) represents the wheel / vehicle speed adjustment coefficients obtained during the first m-1 iterations. P in steps 4-5... W (k), or P when i equals 0 W (ki) represents the current radar velocity ratio coefficient. P is defined as follows: when i equals 1, 2, ..., m. W (k-1), P W (k-2), ..., P W (km), representing the radar velocity ratio coefficients obtained in the first m iterations. μ W (k), μ W (k-1), ..., μ W (k-m+1), for P W (k), P W (k-1), ..., P W The radar velocity weighting coefficients corresponding to (k-m+1) satisfy the equation

[0133]

[0134] The relationship between μ and μ, from largest to smallest. W (k), μ W (k-1), ..., μ W The value is taken from (k-m+1), for example, when m equals 4, μ W (k), μ W (k-1), μ W (k-2), μ W (k-3) is equal to 0.4, 0.3, 0.2, 0.1 respectively, or equal to 0.55, 0.27, 0.13, 0.05 respectively, and so on.

[0135] Step 6: Calculate the wheel / vehicle speed ratio coefficient U V In Example 1 of (k), μ V (k-1), μV (k-2), ..., μ V (k-m+1) is related to P V (k-1), P V (k-2), ..., P V The corresponding variable weighting coefficients (k-m+1) satisfy the equation

[0136]

[0137] The relationship between μ and μ, from largest to smallest. V (k-1), μ V (k-2), ..., μ V The value is taken from (k-m+1), for example, when m equals 4, μ V (k), μ V (k-1), μ V (k-2), μ V (k-3) is equal to 0.4, 0.3, 0.2, 0.1 respectively, or equal to 0.55, 0.27, 0.13, 0.05 respectively, and so on.

[0138] In step 6, the train speed-related quantities include train speed V, creep rate of change x1, creep x2, and train wheelset speed rate of change x3. Current train speed V C (h) According to formula

[0139]

[0140] Adjustments are made to the calculation period and sampling period T. V Same. V(h), V(k), W(h), W(k), U(k), W * (k), V C The unit of (h) is m / s; T V T U The unit is seconds (s). The train speed V is taken as the current train speed V. C (h); The unit of train speed V is km / h. After converting the unit m / s to km / h, the value of train speed V is equal to V0. C 3.6 times the (h) value.

[0141] U V (k) reflects the ratio between the train wheelset speed and the train speed, therefore the creep coefficient x2 can be calculated according to the formula...

[0142]

[0143] Perform calculations, calculation period and sampling period T U Same. Or, according to the formula.

[0144]

[0145] Calculate the current creep degree x2(h), calculation period and sampling period T V Similarly, take the creep degree x2 as equal to the current creep degree x2(h).

[0146] The rate of change of creep x1 is calculated according to the formula

[0147]

[0148] Perform calculations, calculation period and sampling period T U Same. U V (k-1) is the wheel / vehicle speed ratio coefficient obtained from the previous iterative calculation using the train speed adjustment method. Alternatively, according to formula...

[0149]

[0150] Calculate the current rate of change of creep x1, calculation period and sampling period T. V Same. x2(h-1) is the previous sampling period T. V The current creep degree obtained when calculating creep degree.

[0151] The train wheelset speed change rate x3 is calculated according to the formula

[0152]

[0153] Perform calculations, calculation period and sampling period T V Same. V(h-1) is the previous sample value of V(h).

[0154] The creep degree x2 and creep degree change rate x1 are calculated using equations (16) and (18). To ensure a rapid response when calculating the idling risk value E, it is recommended to use equation (5) to calculate the idling risk value E. When the creep degree x2 and creep degree change rate x1 are calculated using equations (17) and (19), equation (4) or equation (5) can be selected to calculate the idling risk value E as needed.

[0155] The sampling time of the τth train wheel rotation speed V(h) before the sampling time of the vehicle-mounted satellite positioning system speed U(k) is the synchronous sampling time point of the vehicle-mounted satellite positioning system speed U(k), where τ is the number of delay intervals. The train wheel rotation speed sampled at this point is V(k). Similarly, the sampling time of the τth train radar speed before the sampling time of the vehicle-mounted satellite positioning system speed U(k) is the synchronous sampling time point of U(k), and the train radar speed sampled at this point is W(k). The synchronous sampling time points of the train radar speed W(k) and the train wheel rotation speed V(k) are consistent. When the vehicle-mounted satellite positioning system speed U(k) is invalid, its sampling time still exists, that is, the synchronous sampling time point of U(k) still exists. Figure 8 The flowchart illustrates a method for calculating the number of delay interval cycles in an embodiment of a train speed adjustment system. The calculation cycle is the same as the acquisition cycle of the speed acquisition unit in the onboard satellite positioning system. This calculation can be performed before or after the iterative calculation of the train speed adjustment method. Specifically:

[0156] Step ①: Obtain the current time, i.e., time k (i.e., kT). U The rate of change of train (locomotive) acceleration β(k) at the sampling time;

[0157] Step ②: Determine whether the conditions for calculating the number of delay interval periods are met. The equation is satisfied.

[0158]

[0159] If the relationship is such that the speed of the vehicle-mounted satellite positioning system has been valid for the most recent m1 consecutive checks, proceed to step ③; otherwise, exit; m1 is greater than or equal to 10. The acceleration change threshold ε can be selected based on the train's acceleration capability and in conjunction with experiments. The value of ε can be determined by... to Choose from within the range of values. This represents the average acceleration of the train upon startup. In the embodiment, T... U Given a time interval of 1 second and m1 equals 20, the average acceleration of a train from 0 to 200 meters can typically reach 0.4 m / s². 2 Then, the value of ε can be selected in the range of 0.4 to 2.4. For example, ε can be taken as 0.75. In equation (21), β(ki) when i equals 0 is the train acceleration change rate β(k) at the current moment; β(ki) when i equals 1 is the train acceleration change rate obtained when calculating the number of delay interval cycles (i.e., iteratively calculating the wheel / vehicle speed ratio coefficient) in the previous calculation; and so on, β(ki) when i equals 1 to m1-1 are the train acceleration change rates obtained when calculating the number of delay interval cycles in the previous m1-1 calculations. The most recent consecutive m1 judgments that the vehicle satellite positioning system speed is valid refer to the following: Figure 6In the iterative calculation of the train speed adjustment method, in the most recent consecutive m1 iterations, step 3 determined that the speed of the on-board satellite positioning system was valid.

[0160] Step ③: Obtain the number of hysteresis intervals τ. The method is to set the parameter to be optimized as the number of hysteresis intervals τ. * Radar speed proportionality coefficient p W * And wheelset speed ratio coefficient p V * ;τ * The value of τ is selected within the range that ensures the delay interval is no greater than 2 seconds. * Greater than 0, less than 2 / T V Integer; in the example, T V Equals 32ms, or 0.032s; 2 / T V Equals 62.5, therefore τ * The value of p is greater than 0 and less than or equal to 62. W * and p V * The values ​​of p are all greater than or equal to 0.8 and less than or equal to 1.2. W * and p V * Only used in this optimization process. The number of delay interval periods is τ. * At that time, the train wheel rotation speed collected at the synchronous acquisition time point corresponding to U(ki) is V. * (ki), the train radar speed acquired at the synchronous acquisition time point corresponding to U(ki) is W. * (ki), the minimum optimization objective function is

[0161]

[0162] Optimization can employ various algorithms such as genetic algorithms and particle swarm optimization, selecting the number of delay intervals τ that satisfies the optimal (minimum) value Q. * Let τ be the number of delay intervals.

[0163] In step ①, obtain kT U The method for calculating the rate of change of train acceleration β(k) at the sampling time is as follows:

[0164]

[0165] The calculation is performed, where α(k) is the currently collected train (locomotive) acceleration, and α(k-1) is the train acceleration collected in the previous iteration. In this embodiment, the currently collected train acceleration α(k) is calculated according to the formula...

[0166]

[0167] The calculation is performed, where U(k) is the currently acquired vehicle satellite positioning system speed, and U(k-1) is the previously acquired vehicle satellite positioning system speed. The train acceleration α(k) can also be measured using an accelerometer. The unit of α(k) is m / s². 2 The unit of β(k) is m / s. 3 .

[0168] Figure 9 This diagram illustrates the speed acquisition hysteresis, train acceleration, and rate of change of train acceleration for an onboard satellite positioning system. Here, V(t) is the train wheel rotation speed obtained by making V(h) continuous, W(t) is the train radar speed obtained by making W(h) continuous, and U(t) is the satellite positioning system speed obtained by making U(k) continuous. τ The lag time between the vehicle-mounted satellite positioning system's speed acquisition time and the train wheel rotation speed acquisition time is given. Points k-7 to k represent the various sampling times (k-7)T for the vehicle-mounted satellite positioning system's speed. U To kT U α(k) and β(k) are the train acceleration and the rate of change of train acceleration, respectively.

[0169] Figure 10 This diagram illustrates the synchronized data acquisition time points of the train wheel rotation speed and train radar speed for the onboard satellite positioning system. The sampling time at which U(k) occurs (i.e., kT) is... U The sampling times of V(h-τ), V(h-τ+1), ..., V(h-3), V(h-2), V(h-1), V(h), etc., are the sampling times of the train wheel rotation speed. For example, the time of V(h) is its sampling time hT. V Due to ionospheric delay and other factors, for the acquisition of train speed (including speed from the onboard satellite positioning system and train radar speed) and train wheel rotation speed at the same moment, the acquisition time of the onboard satellite positioning system speed lags behind the acquisition time of the train wheel rotation speed and train radar speed, with a time lag value of T. τ The delay interval τ is the acquisition period T relative to the train wheel rotation speed. V The number of cycles, i.e., the number of hysteresis interval cycles τ, is the time lag value of the acquisition time of the vehicle-mounted satellite positioning system lagging behind the acquisition time of the train wheel rotation speed and the train radar speed, converted into the acquisition period T. V Multiple values. Figure 10 In the above, V(h-τ) is located at the sampling time (h-τ)T. VLet V(k) be the synchronous acquisition time point of the vehicle-mounted satellite positioning system speed U(k). The train wheel rotation speed V(h-τ) acquired at this point is V(k). Specifically, the τ-th train wheel rotation speed acquisition time point (which is also the train radar speed acquisition time point) before the vehicle-mounted satellite positioning system speed U(k) sampling time point is the synchronous acquisition time point of U(k). The acquisition period and time point of the train radar speed and the train wheel rotation speed are the same, and the mutual delay between them is negligible. Therefore, the sampling times of the train radar speeds W(h-τ), W(h-τ+1), ..., W(h-3), W(h-2), W(h-1), and W(h) are the same as the sampling times of the train wheel rotation speeds V(h-τ), V(h-τ+1), ..., V(h-3), V(h-2), V(h-1), and V(h), respectively. The sampling time (h-τ)T of V(h-τ) is... V The sampling time of W(h-τ) is also the synchronous acquisition time point of the vehicle-mounted satellite positioning system speed U(k). The train radar speed W(h-τ) acquired at this point is W(k).

[0170] Similarly, with Figure 10 For example, when performing the optimization calculation of the number of delay interval periods τ, if τ * If V(h-1) equals 1, then the sampling point where V(h-1) is located is its corresponding synchronous acquisition time point, and its V * (k) equals V(h-1), W * (k) equals W(h-1); if τ * If V(h-2) equals 2, then the sampling point where V(h-2) is located is its corresponding synchronous acquisition time point, and its V * (k) equals V(h-2), W * (k) equals W(h-2); and so on. Note that, for example, τ * Equal to 1, V * (k) equals V(h-1), and V * (k-1) is not V(h-2); in the embodiment, the vehicle-mounted satellite positioning system samples the speed once, and the train wheel rotation speed is sampled an average of 31.25 times. Therefore, if τ * Equal to 1, V * (k) equals V(h-1), then V * (k-1) could be V(h-32) or V(h-33). When implementing the method for calculating the number of delay interval cycles, the most recent m1·T is collected when various speed data of the train are acquired. UWithin a given timeframe, the train wheelsets must not experience free spin during train operation. When the train wheelsets do not experience free spin, the proportional relationship between the train wheelset speed and the speed U(k) displayed by the onboard satellite positioning system, which reflects the train speed, remains relatively stable. In other words, the proportionality coefficient between the actual wheelset speed and the train speed during operation is relatively stable, or the wheelset speed proportionality coefficient p... V * Relatively stable.

[0171] Due to creep, especially wheel spin, the train's wheelset speed does not match the actual train speed. Furthermore, when determining whether wheel spin has occurred and calculating creep rate and creep degree, the train's wheelset speed and train speed must be measured separately; the wheelset speed cannot be used to represent the train speed. Train speed is commonly measured using radar or satellite positioning. Satellite positioning speed measurement uses satellite positioning to track the train's speed and position in real time, transmitting this information to the train control system for processing to obtain the final train speed. Satellite positioning speed measurement can overcome errors caused by wheel spin and slippage, but its accuracy is greatly affected by weather and terrain, and it cannot achieve 100% speed measurement. Data transmission delays exist, and these delays are variable due to distance and ionospheric conditions, affecting the real-time performance of speed measurements. Radar speed measuring devices are typically installed under the train, with the radar antenna at a certain angle to the ground. The radar waves are emitted in the direction of the train. When the train moves relative to the ground, the received radar waves will experience a frequency shift. This shift is determined based on the radar wavelength, the amount of frequency shift, and the angle. The train speed can be calculated by solving data such as the radar installation height; however, the included angle... Data such as radar installation height may fluctuate over time, and road conditions vary, potentially affecting the accuracy of radar speed measurement. In the aforementioned train speed adjustment system, when satellite positioning speed measurement is effective, the wheel / vehicle speed ratio adjustment model parameters and train radar speed adjustment model parameters are adjusted and calculated using the satellite positioning speed measurement data. When satellite positioning speed measurement is ineffective, new train radar speed adjustment model parameters are calculated from the previously adjusted parameters using a given expression or by employing a first-order linear fitting method. These adjusted radar speed adjustment model parameters are then used to adjust and calculate the wheel / vehicle speed ratio adjustment model parameters. Finally, based on the wheel / vehicle speed ratio adjustment model, various train speed-related quantities, such as train speed, creep rate of change, creep, and train wheelset speed rate of change, are calculated. This method combines the high accuracy of satellite positioning speed measurement with the real-time performance and long-term operational capability of radar speed measurement, improving the accuracy and reliability of measuring various train speed-related quantities. The train speed adjustment method also uses the method of determining whether the train is in a speed change state. If it is in a speed change state, the information obtained from radar speed measurement, satellite positioning speed measurement, and train wheelset speed measurement after the train speed change is collected to optimize the satellite positioning data transmission time, i.e., the number of delay intervals, so as to obtain the accurate real-time satellite positioning data transmission delay time (i.e., the number of delay intervals). This further ensures the accuracy and reliability of the relevant speed data calculated by the aforementioned train speed adjustment method.

Claims

1. A method for adjusting train speed, characterized in that, Train speed V C ( h According to the formula Adjustments were made and calculations were performed to determine the train speed. V For train speed V C ( h );in, W(h) For train radar speed, P W Radar speed adjustment factor; train speed V C ( h The adjustment calculation cycle of the train radar speed W(h) The sampling period is T V ; The radar velocity adjustment coefficient is obtained by iterative calculation. P W Creep change rate x 1. Creep x 2 and the rate of change of train wheelset speed x 3. The method is: Step 1, read the first k The speed of the vehicle-mounted satellite positioning system during the next iteration calculation U ( k and location status information X ( k ); Step 2, read the speed from the vehicle's onboard satellite positioning system. U ( k Train wheel rotation speeds collected synchronously at specific time points V ( k ) and train radar speed W ( k ); Step 3: Determine if the vehicle-mounted satellite positioning system speed is valid; if the vehicle-mounted satellite positioning system speed is valid, proceed to step 4; if the vehicle-mounted satellite positioning system speed is invalid, proceed to step 5. Step 4, according to the formula Adjust the current wheel / vehicle speed adjustment coefficient P V ( k and radar speed ratio coefficient P W ( k ); Adjust the radar speed coefficient P W equal P W ( k Proceed to step 6; Step 5, according to the formula Calculate the current radar velocity ratio coefficient P W ( k According to the formula Calculate radar speed adjustment coefficient P W According to the formula Calculate radar synchronization adjustment speed W * ( k According to the formula Adjust the current wheel / vehicle speed adjustment coefficient P V ( k Proceed to step 6; Step 6, for m points ( k , P V ( k )), ( k -1, P V ( k -1)), ..., ( k - m +1, P V ( k - m +1)) Perform linear fitting to obtain the first-order fitted line of the wheel / vehicle speed adjustment coefficient, and take the point (k,) on the first-order fitted line of the wheel / vehicle speed adjustment coefficient. U V ( k The value of )) U V ( k ) is the wheel / vehicle speed ratio coefficient; according to formula Calculate creep x 2; according to formula Calculate the rate of change of creep x 1; According to the formula Calculate the rate of change of train wheelset speed x 3, V ( h () represents the rotational speed of the train wheels; V ( h The sampling period is T V Train wheelset speed variation rate x The calculation period for 3 is T V Vehicle-mounted satellite positioning system speed U ( k and location status information X ( k The sampling period is T U creep x 2. Creep change rate x The calculation period for 1 is T U ; T V Less than T U ; m It is an integer greater than or equal to 3.

2. The train speed adjustment method as described in claim 1, characterized in that, Radar velocity weighting coefficients satisfy the formula The relationship; the values ​​of the velocity weighting coefficients for each radar satisfy... i The smaller the value, the larger the radar velocity weighting coefficient.

3. The train speed adjustment method as described in claim 2, characterized in that, Vehicle-mounted satellite positioning system speed U ( k The first sampling time before τ The rotational speed of each train wheel was collected at the time of the onboard satellite positioning system. U ( k Synchronous data collection time points; τ It is the number of delay interval periods, and its calculation method is to satisfy the following condition: The relationship and the most recent consecutive m When the speed of the vehicle-mounted satellite positioning system is determined to be valid in one test, calculate the number of delay interval periods. τ ; in, β ( k -i) is the nearest m Rate of change of train acceleration ε The threshold for acceleration change is greater than 0; the train acceleration change rate is calculated according to the formula... Perform calculations; among which, α ( k (This refers to the most recently collected train acceleration data.) α ( k -1) represents the train acceleration collected in the previous data collection; the train acceleration is calculated according to the formula... Perform calculations; among which, U ( k -1) For data collection U ( k The speed of the vehicle-mounted satellite positioning system in the previous data collection; Let the parameter to be optimized be the number of delay intervals. τ * Radar speed adjustment coefficient p W * and rotational speed adjustment coefficient p V * The number of delay intervals is τ * At that time, and U ( k -i) The train wheel rotation speed collected at the corresponding synchronous acquisition time point is V * ( k -i), and U ( k -i) The train radar speed collected at the corresponding synchronous acquisition time point is W * ( k -i), the minimum optimization objective function is ； Take the optimal value Q Number of delay intervals τ * Number of delay intervals τ ; τ * The value range is greater than 0 and less than 2 / T V integers, p W * and p V * The values ​​are all greater than or equal to 0.8 and less than or equal to 1.2; m 1 is greater than or equal to 10.

4. The train speed adjustment method as described in claim 3, characterized in that, The method to determine whether the speed of a vehicle-mounted satellite positioning system is effective is to check the positioning status information. X ( k The location status in the image is valid, and the location status information is... X ( k The number of satellites using the position calculation in the system is greater than or equal to 1. δ At any given time, the vehicle's satellite positioning system speed is valid; otherwise, the vehicle's satellite positioning system speed is invalid. δ The value is greater than or equal to 4.

5. The train speed adjustment method as described in any one of claims 1-4, characterized in that, Output creep rate of change x 1. Creep x 2 and the rate of change of train wheelset speed x 3. For train wheelset slip traction control, the method is as follows: Calculate the risk value of idling E ; in, θ 1 represents the threshold for the rate of change of creep. θ 2 represents the creep threshold. θ 3 represents the threshold for the rate of change of wheelset speed; τ It is a non-linear weighted exponent. γ 1. γ 2. γ 3 is a non-linear weighting factor, and τ ≥2、 γ 1≥1、 γ 2≥1、 γ 3≥1; To determine whether a train wheelset is spinning, the method is to determine when the spinning risk value is... E When the value is greater than or equal to 1, the train wheelset will spin; the spinning is controlled by controlling the spinning traction force ratio. θ To achieve idling traction control, idling traction control ratio θ This is the ratio between the train traction force output by the idling traction force control module and the input train traction force, and 0 ≤ 0. θ ≤1; The idling traction control process is: Process I, the process of decreasing traction during idling, from the risk value of idling. E Starting from a value greater than or equal to 1 and continuously increasing, up to the idling risk value. E The process ends when the continuous increase changes to a continuous decrease; control in process I. θ With slope d 1 begins to decrease, and when process I ends... θ The value is the minimum maintenance value; θ The minimum maintenance value is not less than 0; Process II, the process of maintaining the minimum traction value during idling, begins at the end of Process I and continues until the idling risk value is reached. E The process ends when the value is less than 1; during process II, the risk value for idling is... E Continuously reduce, control θ Equal to the minimum maintenance value; Process III, the traction recovery process during idling, begins at the end of Process II and continues until... θ The process ends when the value increases to 1; during process III, the idling traction control module controls... θ With slope d 2 begins to increase until θ Equal to 1; slope d The rate of descent of 1 is greater than the slope. d The rate of increase of 2.

6. The train speed adjustment method as described in claim 5, characterized in that, Train speed V For controlling the upper limit of train traction force, the method involves establishing a maximum traction force limiting neural network model through two rounds of training. The input to the maximum traction force limiting neural network model is the train speed. V and orbital status C 1. The output is the maximum traction force limit of the train. F μm ; The method for the first round of training and modeling is to set the input track state to dry, and to limit the neural network model to a function based on the maximum traction force. To approximate the target, we perform training to obtain an empirical formula approximation model, where... V It is the train speed input to the model. F μm This is the maximum traction force limit value output by the model. P μ To calculate the adhesive weight, a 1. a 2. a 3. a 4. a 5 represents the parameters in the empirical formula for calculating the adhesion coefficient; The second round of training and modeling involves using the collected sample data of train speed, track condition, and corresponding maximum traction force limit values ​​to train the model based on the empirical formula approximation. After training, the model parameters are fixed to obtain the maximum traction force limit neural network model. The output of the neural network model is limited by the maximum traction force. F μm The train's traction force is subject to upper limit and amplitude control. in, F 1 represents the train traction force before the upper limit and amplitude control. F 2 represents the train traction force after the upper limit and amplitude control.

Images