Tooth-entering switching control optimization method for rack rail train

Abstract

The invention provides a toothed rail train tooth-entering switching control optimization method, which comprises the following steps of: regularizing test smooth control parameters during toothed rail train tooth-entering switching to obtain corresponding normalized quantized values, distributing a corresponding initial weight for each normalized quantized value, and calculating the initial weight of each normalized quantized value according to the initial weight of each normalized quantized value; inputting the normalized quantized value carrying the initial weight into an initial tooth entering control model; iteration is conducted on the initial tooth entering control model according to a preset control strategy, iteration is stopped when the train comfort degree reaches the maximum value, and a universal tooth entering control model is obtained; and to-be-optimized smooth control parameters during toothed rail train tooth-entering switching are substituted into the universal tooth-entering control model to obtain an iterative evaluation result under the control strategy, and the iterative evaluation result is used for optimizing the to-be-optimized smooth control parameters. Through the train tooth entering control method and device, the problem that in the prior art, when train tooth entering is manually recognized and train tooth entering is controlled, errors exist due to the control opportunity, time delay and position recognition accuracy, and the train tooth entering smoothness is affected is solved.

Description

Technical Field

[0001] The present invention relates to the field of rail transportation technology, and in particular to a method for optimizing the control of gear switching of a rack railway train. Background Art

[0002] Rack trains are typically used in medium- and low-volume scenarios with steep slopes, with a maximum climbing gradient of up to 480‰. Rack train control systems can be adapted and modified from urban rail transit train control systems such as the communication-based CBTC, track-circuit-based TBTC, or the national railway train control system CTCS.

[0003] The most distinctive safety control function in the rack-and-pinion train control system is the train's in-and-out rack switching control. This controllable switching allows the train to alternate between operating on friction between the train's wheels and rails and generating power through meshing of the train's gears with the ground rack.

[0004] Currently, there are no commercially operated rack railway lines in China. The only rack line under construction is still undergoing testing and debugging for its entry and exit switching. This line uses manual identification of the entry position and manual control to initiate the entry switching process. Due to uncontrollable errors in timing, delay, and position recognition accuracy, manual control often results in "tooth bumping" during the train's entry process, seriously affecting the smoothness of the train's entry. Summary of the Invention

[0005] The present invention provides a method for optimizing the switching control of the tooth entry of a rack railway train, which is used to solve the problem in the prior art that errors in control timing, delay, and position recognition accuracy occur during manual identification and control of the train tooth entry, thus affecting the smoothness of the train tooth entry.

[0006] The present invention provides a method for optimizing the switching control of a rack train, the method comprising: Regularizing the experimental smoothing control parameters of the rack train during gear switching to obtain corresponding normalized quantized values, wherein the experimental smoothing control parameters include the train comfort during gear switching; Assigning a corresponding initial weight to each of the normalized quantized values, and inputting the normalized quantized values ​​carrying the initial weights into an initial gear entry control model; Iterating the initial tooth entry control model according to a preset control strategy, and stopping the iteration when the train comfort reaches a maximum value, to obtain a universal tooth entry control model; The smooth control parameters to be optimized during the gear entry switching of the rack railway train are substituted into the universal gear entry control model to obtain an iterative evaluation result under the control strategy, and the iterative evaluation result is used to optimize the smooth control parameters to be optimized.

[0007] In some embodiments, the step of regularizing the test smoothing control parameters during the gear switching of the rack train to obtain corresponding normalized quantized values ​​includes: For the test smoothing control parameters of the rack train during gear switching, other test parameters except train comfort are determined, and the actual upper and lower limits of the other test parameters during gear switching are normalized and quantized to obtain corresponding normalized quantized values, wherein the other test parameters include at least one of the following: the gear switching speed of the rack train; the impact force borne by the gear switching device; the traction load of the gear switching train; With respect to the train comfort during gear switching, the train comfort is normalized and quantized to obtain a corresponding normalized quantized value.

[0008] In some embodiments, the control strategy includes dynamic weight iteration and fixed weight iteration, and the iterative evaluation results under the control strategy include weight distribution evaluation results and parameter deviation evaluation results.

[0009] In some embodiments, the initial tooth entry control model includes a maximum likelihood function based on maximum train comfort, and iterating the initial tooth entry control model according to a preset control strategy includes: When the control strategy is dynamic weight iteration, iterative calculation is performed based on the maximum likelihood function of the input normalized quantized value, and the target weight of each normalized quantized value when the train comfort reaches the maximum value is determined; When the control strategy is fixed weight iteration, iterative calculation is performed based on the maximum likelihood function of the input normalized quantized value, and the target normalized quantized value when the train comfort reaches the maximum value is determined.

[0010] In some embodiments, substituting the smoothing control parameters to be optimized during the gear entry switching of the rack train into the universal gear entry control model to obtain an iterative evaluation result under the control strategy includes: Determine the true normalized quantized value corresponding to the smoothing control parameter to be optimized when the rack train switches into the gear; When the control strategy is dynamic weight iteration, in the universal gear entry control model, a target weight corresponding to each true normalized quantized value is determined, and the target weight is determined as a weight allocation evaluation result; When the control strategy is fixed weight iteration, in the universal gear entry control model, the parameter difference value between each true normalized quantized value and the target normalized quantized value is determined respectively, and the parameter difference value is used as the parameter offset evaluation result.

[0011] In some embodiments, the method further comprises: When the iterative evaluation result is a weight distribution evaluation result, the distribution weight of the smoothing control parameter to be optimized is updated according to the target weight included in the weight distribution evaluation result.

[0012] When the iterative evaluation result is a parameter offset evaluation result, the smoothing control parameter to be optimized is updated using the parameter difference value included in the parameter offset evaluation result.

[0013] The present invention also provides a rack train gear switching control optimization device, the device comprising: a regularization module, configured to regularize the experimental smoothing control parameters of the rack train during the gear switching operation to obtain corresponding normalized quantized values, wherein the experimental smoothing control parameters include the train comfort during the gear switching operation; an allocation module, configured to allocate a corresponding initial weight to each of the normalized quantized values, and input the normalized quantized values ​​carrying the initial weights into an initial gear entry control model; an iteration module, configured to iterate the initial tooth entry control model according to a preset control strategy, and stop the iteration when the train comfort reaches a maximum value, thereby obtaining a universal tooth entry control model; The optimization module is used to substitute the smooth control parameters to be optimized when the rack train switches into the universal rack-entry control model to obtain an iterative evaluation result under the control strategy, and the iterative evaluation result is used to optimize the smooth control parameters to be optimized.

[0014] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for optimizing the control of tooth switching of a rack train as described above is implemented.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described rack train tooth switching control optimization methods.

[0016] The present invention also provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-described rack train tooth switching control optimization methods.

[0017] The present invention provides a method for optimizing the control of gear switching of a rack railway train. The method regularizes the experimental smooth control parameters of the rack railway train during gear switching, assigns corresponding initial weights and inputs them into an initial gear switching control model. The initial gear switching control model is iterated with the maximum train comfort value as the evaluation target, so that an iterative evaluation result can be fitted for the smooth control parameter to be optimized. The iterative evaluation result is used to optimize the smooth control parameter to be optimized, so that the rack railway train achieves the optimal comfort during gear switching and improves the smoothness of the train gear switching. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced one by one below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 It is a flow chart of the rack train gear switching control optimization method provided by the present invention.

[0020] Figure 2 It is a schematic diagram of the principle of the rack train gear switching control optimization method provided by the present invention.

[0021] Figure 3 It is a structural schematic diagram of the rack train gear switching control optimization device provided by the present invention.

[0022] Figure 4 It is a structural schematic diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0023] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0024] The following describes the rack train gear switching control optimization method of the present invention in conjunction with the accompanying drawings. The method is applied to the control system, control server or terminal of the rack train gear switching. When the control system, control server or terminal of the rack train gear switching runs the gear switching process, the control parameters of the train gear switching in the gear switching process are optimized in real time by the method of the present invention to improve the smoothness of the train gear switching and ensure the safety of the train gear switching.

[0025] Figure 1 FIG. 1 is a flow chart of the rack train gear switching control optimization method provided by the present invention, as shown in FIG. Figure 1 As shown, the method includes the following steps 101 to 104, which are described in detail below.

[0026] Step 101 : Regularize the experimental smooth control parameters of the rack railway train during the gear switching to obtain corresponding normalized quantized values, wherein the experimental smooth control parameters include the train comfort during the gear switching.

[0027] In a rack train's gear switching scenario, various factors influence train comfort. These factors are determined by smoothing control parameters, which also factor in train comfort. This is because the goal of a gear switching operation is to complete the operation within a certain level of train comfort (maximum lateral (y-direction) and vertical (z-direction) vibration during travel). This ensures smooth switching. Higher train comfort leads to higher smoothness.

[0028] In the embodiments of the present invention, the optimization of the rack train's gear switching primarily focuses on optimizing the parameters of the gear switching, specifically the smoothing control parameters. Therefore, an initial gear switching control model must be pre-determined to process the smoothing control parameters. Here, the initial gear switching control model is optimized by obtaining experimental smoothing control parameters during the gear switching of the rack train.

[0029] However, since the data types and data units of the smoothing control parameters are different, they cannot be uniformly fitted into a model for calculation. Therefore, the embodiment of the present invention regularizes the experimental smoothing control parameters during the gear switching of the rack train to obtain corresponding normalized quantized values.

[0030] like Figure 2 As shown, the experimental smoothing control parameters include train comfort during gear switching. The remaining experimental parameters, in addition to train comfort, include at least one of the following: the gear train's gear switching speed, the impact force experienced by the gear switching device, and the traction load of the gear switching train. These smoothing control parameters have the greatest impact on train comfort. In actual implementation, other factors with less significant impact may exist, but they are not considered in this embodiment of the present invention.

[0031] During the regularization process, for the test smoothing control parameters when the rack railway train switches into gear, except for the train comfort, the actual upper and lower limits of the other test parameters when the rack railway train switches into gear are determined, and the actual upper and lower limits are normalized and quantized to obtain the corresponding normalized quantized values.

[0032] For example, when a rack train is running, the impact force on the gear entry device will have a critical value, that is, the actual upper and lower limits of the value. Then, the actual upper and lower limits of the value are normalized and quantified, for example, quantified to a parameter with a value range of [0, 1].

[0033] Regarding the train comfort during gear switching, the train comfort is normalized and quantized to obtain a corresponding normalized quantized value. Train comfort is an ideal value; the larger the better. For ease of calculation, the train comfort is also normalized and quantized to obtain a corresponding normalized quantized value, for example, quantized to a value of [0, Comf]. The parameter Comf can be set according to actual needs.

[0034] In an embodiment of the present invention, when a rack train switches gears, in order to ensure train comfort, various smoothing control parameters that affect train comfort are comprehensively considered, and the smoothing control parameters are uniformly regularized to guide gear switching control.

[0035] Step 102: assign a corresponding initial weight to each normalized quantized value, and input the normalized quantized value carrying the initial weight into the initial tooth entry control model.

[0036] like Figure 2 As shown in the figure, the quality of the tooth entry control model is closely related to the weight distribution of each smoothing control parameter. Each smoothing control parameter is assigned a corresponding weight. The weight distribution represents the influence and contribution of the tooth entry control parameter to the tooth entry smoothing control. The sum of the weights of all smoothing control parameters is 1.

[0037] Therefore, in the embodiment of the present invention, a corresponding initial weight is assigned to each normalized quantized value. The weight assignment is performed on each parameter based on prior knowledge. The weight is greatly affected by objective factors. Therefore, at the beginning of model establishment, it is necessary to carefully assign corresponding weights to each parameter based on prior knowledge.

[0038] Then the normalized quantized value carrying the initial weight is input into the initial tooth entry control model, such as Figure 2 fx1, fx2, fx3, ..., fxn in the figure. In this embodiment of the present invention, the initial tooth entry control model includes a maximum likelihood function based on maximum train comfort. This function can be set as a multivariate function that accepts multiple smoothing control parameters. In essence, it is a maximum likelihood function based on maximum train comfort. This function is used to select the parameter value that maximizes the likelihood function value within the possible range of the experimental smoothing control parameter, that is, to determine the optimal estimated value of the experimental smoothing control parameter when the maximum train comfort is achieved.

[0039] Here, the normalized quantized value carrying the initial weight is input into the initial tooth entry control model, that is, input into the maximum likelihood function, for subsequent maximum likelihood calculation during simulation iterations.

[0040] Step 103: Iterate the initial gear-entry control model according to a preset control strategy, and stop the iteration when the train comfort reaches a maximum value, to obtain a universal gear-entry control model.

[0041] During model iteration, a maximum likelihood calculation is performed based on the input normalized quantized values ​​with initial weights. This embodiment of the present invention iterates the initial tooth entry control model based on a preset control strategy. The control strategy determines the direction of the model iteration and specifically includes dynamic weight iteration and fixed weight iteration. In actual implementation, one or both control strategies can be used based on the iterative evaluation results.

[0042] The following describes the process of iterating the initial tooth entry control model according to the preset control strategy.

[0043] When the control strategy is dynamic weight iteration, this means that the initial weights are dynamically variable during the model iteration process. After the normalized quantized values ​​carrying the initial weights are input into the initial tooth entry control model, an iterative calculation is performed based on the maximum likelihood function of the input normalized quantized values, and the target weights of each normalized quantized value are determined when the train comfort reaches its maximum value. This target weight can accurately reflect the impact of different smoothing control parameters on train comfort. Because the initial weights are dynamically variable during model iteration, the normalized quantized values ​​and initial weights are continuously updated during the maximum likelihood function calculation, ultimately obtaining a set of target weights when the train comfort reaches its maximum value.

[0044] When the control strategy uses fixed-weight iteration, the initial weights are fixed during the model iteration process. After the normalized quantized values ​​carrying the initial weights are input into the initial gear entry control model, an iterative calculation is performed using the maximum likelihood function of the input normalized quantized values ​​to determine the target normalized quantized value at which the train comfort reaches its maximum value. Because the initial weights remain unchanged during model iteration, the normalized quantized values ​​are continuously updated only during the maximum likelihood function calculation, ultimately resulting in a set of target normalized quantized values ​​at the point where the train comfort reaches its maximum value.

[0045] In some embodiments, as Figure 2As shown, during the iteration of the initial tooth entry control model according to the control strategy, after the initial tooth entry control model obtains a set of target weights or a set of target normalized quantized values ​​in a certain iteration, the initial tooth entry control model can be further verified using the experimental smoothing control parameters during the tooth entry switching of the rack train in the next round of iteration to determine the control effect of the initial tooth entry control model, and then the strategy or model correction is performed based on the control effect. Strategy correction is to change the corresponding control strategy, and model correction is to continuously iteratively update the experimental smoothing control parameters during the tooth entry switching of the rack train to obtain new target normalized quantized values ​​or new target weights.

[0046] During the model iteration process, iteration stops when the train comfort reaches its maximum value, or after a preset number of iterations. After the initial gear-entry control model stops iterating, a universal gear-entry control model is generated. This universal gear-entry control model can be applied to actual gear-entry switching scenarios for rack trains, enabling real-time optimization of the desired smoothing control parameters.

[0047] Step 104 , substitute the smooth control parameters to be optimized during the gear entry switching of the rack railway train into the universal gear entry control model to obtain an iterative evaluation result under the control strategy, and the iterative evaluation result is used to optimize the smooth control parameters to be optimized.

[0048] After the universal gear entry control model is determined in step 103, in an actual rack train gear entry scenario, the smoothing control parameters to be optimized during the rack train gear entry switching are substituted into the universal gear entry control model to obtain an iterative evaluation result under the control strategy.

[0049] Here, due to different control strategies, the iterative evaluation results under each control strategy are also different. Corresponding to the control strategy, the iterative evaluation results include weight allocation evaluation results and parameter deviation evaluation results. The weight allocation evaluation results are obtained under dynamic weight iteration, while the parameter deviation evaluation results are obtained under fixed weight iteration.

[0050] First, the true normalized quantized value corresponding to the smoothing control parameter to be optimized during the gear entry switching of the rack railway train is determined through the same regularization process as step 101, and then the true normalized quantized value is input into the universal gear entry control model.

[0051] In some embodiments, when the control strategy is dynamic weight iteration, in the universal tooth-entry control model, the target weight corresponding to each true normalized quantized value is determined, and the target weight is determined as the weight distribution evaluation result.

[0052] If the control strategy used is dynamic weight iteration, the universal in-gear control model will generate a set of target weights after iteration. These target weights are determined when the train comfort reaches its maximum value during the model iteration and accurately reflect the impact of different smoothing control parameters on train comfort. Therefore, in the universal in-gear control model, the corresponding target weights can be found according to the true normalized quantized values. This directly determines the target weight corresponding to each true normalized quantized value and uses it as the weight distribution evaluation result.

[0053] In some embodiments, when the control strategy is fixed weight iteration, in the universal tooth entry control model, the parameter difference value between each true normalized quantized value and the target normalized quantized value is determined respectively, and the parameter difference value is used as the parameter offset evaluation result.

[0054] Here, if the control strategy used is fixed-weight iteration, the universal in-gear control model will generate a set of target normalized quantized values ​​after iteration. These target normalized quantized values ​​maximize train comfort under fixed weights and can be used as a criterion for setting smoothing control parameters. Therefore, in the universal in-gear control model, the parameter difference between each true normalized quantized value and the corresponding target normalized quantized value is determined. This parameter difference represents the degree of deviation from the true normalized quantized value and can be used as a parameter offset evaluation result.

[0055] Therefore, in the universal tooth-entry control model, different iterative evaluation results can be obtained according to different control strategies. This iterative evaluation result can be used to directly optimize the smoothing control parameters to be optimized in the tooth-entry switching process of the actual rack train tooth-entry switching scenario.

[0056] Optionally, when a control strategy of either dynamic weight iteration or fixed weight iteration is used, the smoothing control parameter to be optimized is optimized according to an iterative evaluation result corresponding to the control strategy.

[0057] Specifically, when the iterative evaluation result is a weight distribution evaluation result, the target weight included in the weight distribution evaluation result is used to update the distribution weight of the smoothing control parameter to be optimized, and is used for the rack train gear switching.

[0058] When the iterative evaluation result is a parameter offset evaluation result, the parameter difference value included in the parameter offset evaluation result is used to update the smoothing control parameter to be optimized, thereby achieving parameter optimization or weight optimization of the smoothing control parameter to be optimized, so that the optimal train comfort value can be obtained when the rack train switches into gear.

[0059] Optionally, when the dynamic weight iteration and the fixed weight iteration control strategies are used simultaneously, the smoothing control parameters to be optimized are optimized simultaneously according to the iterative evaluation results corresponding to the two control strategies.

[0060] Specifically, the distribution weights of the smooth control parameters to be optimized are updated by using the target weights included in the weight distribution evaluation results in the iterative evaluation results, and the smooth control parameters to be optimized are updated by using the parameter difference values ​​included in the parameter offset evaluation results in the iterative evaluation results. This not only achieves the optimization of the smooth control parameters to be optimized, but also achieves the update of the distribution weights of the smooth control parameters to be optimized, so that the optimal train comfort value can be achieved when the rack train switches into gear.

[0061] In this embodiment of the present invention, different control strategies are set to perform bidirectional model iterations, all with the goal of maximizing train comfort. After the model iterations, the gear-entry control model determines the iterative evaluation results corresponding to the different control strategies, achieving parameter optimization and weight optimization of the smoothing control parameters to be optimized. This not only effectively improves the smoothness of the rack train's gear-entry switching, but also effectively identifies potential parameter conflicts arising from the optimized smoothing control parameters, ensuring the safety of the train's gear-entry switching.

[0062] The following describes the rack train gear entry switching control optimization device provided by the present invention. The rack train gear entry switching control optimization device described below and the rack train gear entry switching control optimization method described above can be referenced to each other.

[0063] like Figure 3 As shown, the rack train gear switching control optimization device includes: a regularization module 301, an allocation module 302, an iteration module 303, and an optimization module 304. Specifically, the regularization module 301 is used to regularize the experimental smooth control parameters of the rack train during gear switching to obtain corresponding normalized quantized values, wherein the experimental smooth control parameters include the train comfort during gear switching; the allocation module 302 is used to assign a corresponding initial weight to each of the normalized quantized values, and input the normalized quantized values ​​carrying the initial weights into the initial gear switching control model; the iteration module 303 is used to iterate the initial gear switching control model according to a preset control strategy, and stop the iteration when the train comfort reaches the maximum value to obtain a universal gear switching control model; the optimization module 304 is used to substitute the smooth control parameters to be optimized during gear switching of the rack train into the universal gear switching control model to obtain an iterative evaluation result under the control strategy, and the iterative evaluation result is used to optimize the smooth control parameters to be optimized.

[0064] It should be noted that the beneficial effects of the rack railway train gear entry switching control optimization device here and the rack railway train gear entry switching control optimization method mentioned above can correspond to each other, so the beneficial effects of the rack railway train gear entry switching control optimization device will not be repeated here.

[0065] Figure 4An example of a physical structure diagram of an electronic device is shown below. Figure 4 As shown, the electronic device may include: a processor (processor) 410 , a communication interface (Communications Interface) 420 , a memory (memory) 430 and a communication bus 440 , wherein the processor 410 , the communication interface 420 , and the memory 430 communicate with each other via the communication bus 440 . The processor 410 can call the logic instructions in the memory 430 to execute the rack train gear switching control optimization method, which includes: regularizing the experimental smooth control parameters of the rack train during gear switching to obtain corresponding normalized quantized values, wherein the experimental smooth control parameters include the train comfort during gear switching; assigning a corresponding initial weight to each of the normalized quantized values, and inputting the normalized quantized value carrying the initial weight into the initial gear entry control model; iterating the initial gear entry control model according to a preset control strategy, and stopping the iteration when the train comfort reaches the maximum value to obtain a universal gear entry control model; substituting the smooth control parameters to be optimized during gear switching of the rack train into the universal gear entry control model to obtain an iterative evaluation result under the control strategy, and the iterative evaluation result is used to optimize the smooth control parameters to be optimized.

[0066] Furthermore, the logic instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product, stored in a storage medium, includes instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0067] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the rack train gear switching control optimization method provided by the above methods, the method including: regularizing the experimental smooth control parameters of the rack train during gear switching to obtain corresponding normalized quantized values, wherein the experimental smooth control parameters include the train comfort during gear switching; assigning a corresponding initial weight to each of the normalized quantized values, and inputting the normalized quantized value carrying the initial weight into the initial gear entry control model; iterating the initial gear entry control model according to a preset control strategy, and stopping the iteration when the train comfort reaches the maximum value to obtain a universal gear entry control model; substituting the smooth control parameters to be optimized during gear switching of the rack train into the universal gear entry control model to obtain an iterative evaluation result under the control strategy, and the iterative evaluation result is used to optimize the smooth control parameters to be optimized.

[0068] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the rack train gear switching control optimization method provided by the above-mentioned methods, the method comprising: regularizing the experimental smooth control parameters of the rack train during gear switching to obtain corresponding normalized quantized values, wherein the experimental smooth control parameters include the train comfort during gear switching; assigning a corresponding initial weight to each of the normalized quantized values, and inputting the normalized quantized value carrying the initial weight into the initial gear entering control model; iterating the initial gear entering control model according to a preset control strategy, and stopping the iteration when the train comfort reaches a maximum value to obtain a universal gear entering control model; substituting the smooth control parameters to be optimized during gear switching of the rack train into the universal gear entering control model to obtain an iterative evaluation result under the control strategy, and the iterative evaluation result is used to optimize the smooth control parameters to be optimized.

[0069] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0070] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0071] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A rack train gear switching control optimization method, characterized in that: The method comprises: Regularizing the test smoothing control parameters of the rack train during the gear switching to obtain corresponding normalized quantized values, wherein the test smoothing control parameters include the train comfort during the gear switching; Assigning a corresponding initial weight to each of the normalized quantized values, and inputting the normalized quantized values ​​carrying the initial weights into an initial tooth entry control model; Iterating the initial tooth-entry control model according to a preset control strategy, and stopping the iteration when the train comfort reaches a maximum value, to obtain a universal tooth-entry control model; The smooth control parameters to be optimized during the gear entry switching of the rack railway train are substituted into the general gear entry control model to obtain an iterative evaluation result under the control strategy, and the iterative evaluation result is used to optimize the smooth control parameters to be optimized.

2. The rack train gear switching control optimization method according to claim 1 is characterized in that: The test smoothing control parameters of the rack train during the gear switching are regularized to obtain the corresponding normalized quantized values, including: For the other test parameters except the train comfort in the test smoothing control parameters when the rack train switches into the gear, determine the actual upper and lower limits of the other test parameters when the rack train switches into the gear, and normalize and quantize the actual upper and lower limits to obtain corresponding normalized quantized values, wherein the other test parameters include at least one of the following: the speed of the rack train switching into the gear; the impact force borne by the gear-entering device; the traction load of the gear-entering train; With respect to the train comfort during gear switching, the train comfort is normalized and quantized to obtain a corresponding normalized and quantized value.

3. The rack train gear switching control optimization method according to claim 1, characterized in that: The control strategy includes dynamic weight iteration and fixed weight iteration, and the iterative evaluation results under the control strategy include weight distribution evaluation results and parameter deviation evaluation results.

4. The rack train gear switching control optimization method according to claim 1, characterized in that: The initial tooth entry control model includes a maximum likelihood function based on maximum train comfort, and the initial tooth entry control model is iterated according to a preset control strategy, including: When the control strategy is dynamic weight iteration, iterative calculation is performed in the maximum likelihood function of the input normalized quantized value, and the target weight of each normalized quantized value when the train comfort reaches the maximum value is determined; When the control strategy is fixed weight iteration, an iterative calculation is performed according to the maximum likelihood function of the input normalized quantized value, and the target normalized quantized value when the train comfort reaches the maximum value is determined.

5. The rack train gear switching control optimization method according to claim 1, characterized in that: Substituting the smooth control parameters to be optimized when the rack train switches into the universal rack-entry control model to obtain an iterative evaluation result under the control strategy includes: Determine the true normalized quantized value corresponding to the smooth control parameter to be optimized when the rack train switches into the gear; When the control strategy is dynamic weight iteration, in the universal gear entry control model, the target weight corresponding to each of the true normalized quantized values ​​is determined, and the target weight is determined as the weight allocation evaluation result; When the control strategy is fixed weight iteration, in the universal gear entry control model, the parameter difference value between each true normalized quantized value and the target normalized quantized value is determined respectively, and the parameter difference value is used as the parameter offset evaluation result.

6. The rack train gear switching control optimization method according to claim 1, characterized in that: The method further comprises: When the iterative evaluation result is a weight allocation evaluation result, the allocation weight of the smoothing control parameter to be optimized is updated by the target weight included in the weight allocation evaluation result. When the iterative evaluation result is a parameter offset evaluation result, the smoothing control parameter to be optimized is updated according to the parameter difference value included in the parameter offset evaluation result.

7. A rack train gear switching control optimization device, characterized in that: The device comprises: A regularization module, used for regularizing the test smoothing control parameters of the rack train during the gear switching to obtain the corresponding normalized quantized values, wherein the test smoothing control parameters include the train comfort during the gear switching; An allocation module, configured to allocate a corresponding initial weight to each of the normalized quantized values, and input the normalized quantized value carrying the initial weight into an initial tooth entry control model; An iteration module, used for iterating the initial tooth entry control model according to a preset control strategy, and stopping the iteration when the train comfort reaches a maximum value, so as to obtain a universal tooth entry control model; The optimization module is used to substitute the smooth control parameters to be optimized when the rack train switches into the universal rack control model to obtain an iterative evaluation result under the control strategy, and the iterative evaluation result is used to optimize the smooth control parameters to be optimized.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the rack train tooth switching control optimization method as described in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the rack railway train tooth switching control optimization method as described in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the rack railway train tooth switching control optimization method as described in any one of claims 1 to 6 is implemented.

Images