Collision test parameter recommendation method and device and storage medium

Abstract

The invention discloses a collision test parameter recommendation method and device and a storage medium, and relates to the technical field of automobile collision test. The scheme comprises the following steps: acquiring a target collision motion parameter expected to be realized by a collision test trolley in a deceleration sliding table test; inputting the target collision motion parameters into a trained target prediction model, and obtaining recommended test parameters including damper target parameters; wherein the target prediction model constructs a training sample based on a simulation result of a deceleration sliding table test, and establishes a mapping relation from a motion parameter to a test parameter through a machine learning method. According to the scheme, the recommendation parameters highly matched with the target can be quickly output, the accuracy and efficiency of parameter configuration are remarkably improved, and the debugging process originally needing multiple iterations is greatly simplified.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automobile crash test, in particular to a crash test parameter recommendation method, device and storage medium. BACKGROUND

[0002] Real vehicle crash test is a key link to verify the safety performance of vehicles and occupant protection systems, but it is costly, time-consuming and not well repeatable. Therefore, the industry generally uses slide table crash test to simulate the real vehicle crash process, among which, the deceleration slide table simulates the whole vehicle waveform and motion posture by making the trolley hit a set of preset damping materials after acceleration, because it can better reproduce the complex motion of vehicle in the crash process such as tail swing and pitch, it has become an important simulation test method.

[0003] In the deceleration slide table test, in order to accurately reproduce the target vehicle acceleration waveform and vehicle motion characteristics, the damping materials in the crash test device must be reasonably configured before the test, including key parameters such as the length arrangement of the damping tube, the shear force value setting of the aluminum rod and the installation position of the damper. At present, the determination of these parameters is seriously dependent on the personal experience of engineers and repeated mathematical derivation, usually needs to be adjusted through trial and error for multiple rounds combined with simulation analysis, the whole process takes a long time of several rounds or even dozens of rounds, which is low in efficiency and the consistency of the results is difficult to guarantee. SUMMARY

[0004] Therefore, the present application is committed to providing a crash test parameter recommendation method, device and storage medium, which can quickly and accurately determine the parameter configuration of the damping materials in the deceleration slide table test.

[0005] According to a first aspect of the present application, a collision test parameter recommendation method is provided, comprising: obtaining a target collision motion parameter of a collision test trolley during a deceleration sled test; the target collision motion parameter is a motion parameter expected to be achieved by the collision test trolley in the deceleration sled test; the deceleration sled test is performed by making the collision test trolley impact a damper at a collision test device to achieve the target collision motion parameter; inputting the target collision motion parameter into a target prediction model to obtain a recommended test parameter of the deceleration sled test; the training sample of the target prediction model is constructed according to the simulation result of the deceleration sled test; and the recommended test parameter includes a target parameter of the damper at the collision test device. By constructing a training sample based on the simulation result of the deceleration sled test, training a special target prediction model, and automatically outputting a recommended test parameter containing a target parameter of the damper after receiving a target collision motion parameter, the target prediction model establishes a parameter mapping relationship through machine learning, so that the output recommended test parameter can highly match the target collision motion parameter, greatly improving the accuracy of parameter configuration, reducing the trial-and-error process based on experience and mathematical derivation in the early stage, greatly reducing the number of parameter iteration rounds, significantly shortening the preparation time and test cycle of the deceleration sled test, greatly improving the test efficiency, and reducing the material consumption and time cost caused by repeated debugging, thereby indirectly reducing the comprehensive cost of vehicle development.

[0006] Optionally, the method further comprises: generating an experimental design matrix in the parameter space of the target parameter using an experimental design method; the experimental design matrix is used to represent a simulation input parameter combination of the damper each time the deceleration sled test is simulated; simulating the deceleration sled test based on the experimental design matrix to obtain a simulation result corresponding to each simulation input parameter combination; constructing a training sample for the simulation input parameter combination and the corresponding simulation result; and training an initial prediction model based on the training sample to obtain the target prediction model. By generating an experimental design matrix in the parameter space of the target parameter using an experimental design method, the number of simulation input parameter combinations is reduced while ensuring that all parameter value ranges are covered, the deceleration sled simulation is carried out based on the matrix, and the training sample is constructed, and then the target prediction model is trained. Through the systematic sampling advantage of the experimental design method, blind traversal of the parameter space is avoided, the required training sample size and the corresponding simulation times are greatly reduced, not only the computing resource consumption and time cost in the simulation process are reduced, but also the data processing workload is reduced; at the same time, the uniform and representative training sample can make the model more comprehensively learn the correlation between the parameters and the collision effect, further guaranteeing the accuracy and reliability of the recommended test parameter output by the target prediction model, and ultimately reducing the research and development investment while improving the overall efficiency of the deceleration sled test.

[0007] Optionally, the target collision motion parameter comprises a test type of the deceleration sled test and an acceleration waveform of the crash test sled in the front-rear direction; and the recommended test parameter comprises a number of dampers at the crash test device, a target length of each damper, and a damping force of each damper. By learning the accurate mapping relationship from the test type and the acceleration waveform to the number of dampers, the length of each damper, and the damping force, the dimension of the parameter space is significantly reduced, and the size of the experimental design matrix is reduced. Under the premise of ensuring coverage, the number of simulation samples required is greatly reduced, not only saving computing resources, but also making the data distribution of the training samples more concentrated, effectively reducing the learning difficulty of the target prediction model, and improving the convergence speed of the model training.

[0008] Optionally, the target collision motion parameter comprises a test type of the deceleration sled test, an acceleration waveform of the crash test sled in the front-rear direction, and tail swing data and rotation data of the crash test sled; and the recommended test parameter comprises a number of dampers at the crash test device, a target length of each damper, a damping force of each damper, and a position of each damper. By taking the test type, the acceleration waveform, the tail swing data, and the rotation data as the target collision motion parameter, and by taking the number of dampers, the target length, the damping force, and the position of each damper as the recommended test parameter, a complete parameter mapping relationship is established. This method can accurately reproduce the acceleration waveform and the three-dimensional motion posture of the vehicle during the crash process, and significantly improves the overall reproduction accuracy of the deceleration sled test.

[0009] Optionally, the target prediction model comprises a first prediction model and a second prediction model; and the inputting of the target collision motion parameter into the target prediction model to obtain the recommended test parameter of the deceleration sled test comprises: inputting the test type of the deceleration sled test and the acceleration waveform into the first prediction model to obtain the number of dampers, the target length of each damper, and the shear force value of each damper; and inputting the tail swing data and the rotation data of the crash test sled into the second prediction model to obtain the position of each damper. By adopting the training scheme of the first prediction model and the second prediction model separately, the first prediction model is used to learn the mapping relationship between the test type and the acceleration waveform and the number of dampers, the length of each damper, and the shear force value, and the second prediction model is used to learn the mapping relationship between the tail swing data, the rotation data, and the position of each damer, so that the parameter space of each model is exponentially reduced, the size of the experimental design matrix is significantly reduced, and the number of simulation samples required is greatly reduced. This not only effectively saves computing resources, but also makes the data distribution of the training samples of each model more concentrated, greatly reduces the learning difficulty of the model, significantly improves the convergence speed of the model training, and maintains the accuracy of the prediction of each special parameter.

[0010] Optionally, the experimental design method is used to generate an experimental design matrix in the parameter space of the target parameter, including: using an optimal Latin hypercube algorithm to generate an experimental design matrix in the parameter space of the target parameter, which can obtain more representative sample coverage of a high-dimensional design space composed of key parameters such as damper number, length, damping strength and position with fewer simulation times, thereby effectively reducing the simulation calculation resources required for constructing the target prediction model under the premise of ensuring the training sample coverage and model accuracy, and improving the model training efficiency.

[0011] Optionally, the method further comprises: performing a deceleration sled test based on the recommended test parameters to obtain actual collision motion parameters; and correcting the target prediction model based on the actual collision motion parameters. By applying the model recommended parameters to the physical test and using the measured data for feedback correction, a closed-loop model optimization process is constructed. This process enables the target prediction model to continuously learn the response characteristics of the real physical system and gradually correct the deviation between simulation and measurement, thereby continuously improving the prediction accuracy of the model under actual working conditions.

[0012] Optionally, the target collision motion parameters are input into the target prediction model to obtain the recommended test parameters of the deceleration sled test, including: calling the target prediction model based on a target intelligent agent to obtain the recommended test parameters of the deceleration sled test. By introducing a target intelligent agent, automatic calling of the prediction model is realized. The user only needs to input the target collision motion parameters, and the target intelligent agent can automatically call the target prediction model according to the preset process to obtain complete recommended test parameters. This process does not require manual intervention in model selection and data processing, significantly reducing the system use threshold. This enables engineers to conveniently obtain professional-level parameter recommendation results without deep understanding of the internal structure and calling logic of the model, effectively improving the ease of use in actual engineering applications.

[0013] According to a second aspect of the present application, a collision test parameter recommendation device is provided, comprising: An acquisition module is configured to acquire target collision motion parameters of a collision test vehicle during a deceleration sled test; the target collision motion parameters are motion parameters expected to be achieved by the collision test vehicle in the deceleration sled test; the deceleration sled test is performed by making the collision test vehicle collide with a damper at a collision test device to achieve the target collision motion parameters; A prediction module is configured to input the target collision motion parameters into a target prediction model to obtain recommended test parameters of the deceleration sled test; training samples of the target prediction model are constructed based on simulation results of the deceleration sled test; and the recommended test parameters include target parameters of the damper at the collision test device.

[0014] According to a third aspect of the present application, a computer readable storage medium is provided, the storage medium storing a computer program for executing the method according to any one of the above embodiments.

[0015] According to a fourth aspect of the present application, an electronic device is provided, comprising: a processor; a memory for storing instructions executable by the processor; and the processor is configured to execute the method according to any one of the above embodiments.

[0016] The present application provides a collision test parameter recommendation method, device and storage medium, which comprises the following steps: obtaining target collision motion parameters expected to be achieved by a collision test trolley in a deceleration sliding platform test; inputting the target collision motion parameters into a trained target prediction model to obtain recommended test parameters containing damper target parameters; wherein the target prediction model is constructed based on simulation results of the deceleration sliding platform test, and a mapping relationship from motion parameters to test parameters is established by a machine learning method. The scheme can quickly output recommended parameters that are highly matched with the target, significantly improving the accuracy and efficiency of parameter configuration, and greatly simplifying the debugging process that originally requires multiple iterations. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 The figure is a schematic diagram of the overall structure of the collision test trolley according to an embodiment of the present application.

[0018] Figure 2 The figure is a schematic diagram of the structure of the base provided with a full-frontal collision buffer unit according to an embodiment of the present application.

[0019] Figure 3 The figure is a schematic diagram of the structure of the base provided with a full-frontal collision buffer unit according to an embodiment of the present application. Figure 2 The figure is a front view of the collision test device.

[0020] Figure 4 The figure is a front view of the collision test device. Figure 2 The figure is a partial enlarged view of the part B in the figure.

[0021] Figure 5 The figure is a schematic diagram of the structure of the base provided with a full-frontal collision buffer unit according to an embodiment of the present application.

[0022] Figure 6 The figure is a front view of the collision test device. Figure 5 The figure is a front view of the collision test device.

[0023] Figure 7 The figure is a front view of the collision test device. Figure 5 The figure is a side view of the collision test device.

[0024] Figure 8 The figure is a flowchart of the collision test parameter recommendation method provided by an embodiment of the present application.

[0025] Figure 9 Fig. 1 shows a block diagram of a collision test parameter recommendation device according to an embodiment of the present application.

[0026] Figure 10 Fig. 2 shows a structural block diagram of an electronic device according to an embodiment of the present application.

[0027] Legend of reference signs: 1, suspension body; 2, middle frame; 3, front frame; 4, rear frame; 10, wheel; 50, fixed bottom plate; 51, support frame; 52, mounting base plate; 520, assembly hole; 60, first fixing seat; 600, fixed mounting plate; 601, first extension connecting plate; 602, first connecting hole; 611, first damping mechanism; 612, second damping mechanism; 62, first shearing structural member; 7, frontal offset collision buffering unit; 70, second fixing seat; 701, first damper fixing frame; 702, second damper fixing frame; 71, first damper; 72, second damper; 73, second extension connecting plate; 730, second connecting hole; 74, guiding mechanism; 8, damping tube; 80, damper base. DETAILED DESCRIPTION

[0028] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0029] Summary The deceleration sled is a core collision test equipment in the field of automobile research and development. The core role is to replace the high-cost real vehicle collision test. The acceleration waveform and motion posture during vehicle collision are simulated to provide data support for constraint system optimization and occupant protection development, and the deceleration sled is widely used in vehicle design verification and production verification stage.

[0030] The working mode of the deceleration sled is to accelerate the sled with a body-in-white and interior to a target speed, and then make it impact a preset damping material array in front. These damping materials (such as damping tubes) will deform by crushing, shearing, etc. when impacted, and the kinetic energy of the sled is absorbed to simulate the deformation and energy absorption process during vehicle collision, thereby generating an acceleration waveform similar to real vehicle collision on the sled.

[0031] However, in the early stage of the test, the simulation analysis is relied on for parameter research, and the vehicle collision involves the dynamic interaction of hundreds of parts, the coupling effect of material nonlinear deformation and energy transmission, and the existing simulation model of the deceleration slide table is difficult to completely reproduce the real physical scene. There is always a certain deviation between the simulation results and the actual test. The limitation of simulation accuracy makes it difficult to independently and reliably guide the parameter setting of the test, and the subsequent complex physical debugging process cannot be avoided.

[0032] In order to reproduce the actual vehicle collision waveform, engineers must pre-determine a series of key parameters, such as the length and number of damping tubes, the shear force value of aluminum rods, and the installation position of dampers. If the deceleration slide table is required to reproduce the complex three-dimensional motion of the vehicle in X direction (forward and backward), Y direction (tail swing), and Z direction (tail swing), it is more complex.

[0033] Currently, the determination of these parameters is highly dependent on the mathematical derivation and rich practical experience of engineers, and usually requires multiple rounds of adjustment to achieve a relatively ideal result, which is time-consuming and laborious. Therefore, how to quickly and accurately determine the parameter configuration in the deceleration slide table test has become a key challenge to improve the efficiency of the deceleration slide table test.

[0034] To solve the above problems, the collision test parameter recommendation method provided by the embodiments of the present application converts the traditional parameter configuration process relying on artificial experience and repeated trial and error into an intelligent process of automatically calculating and outputting recommended parameters by a model, by constructing and training a special target prediction model.

[0035] Specifically, the present scheme first constructs a training sample set covering multiple collision conditions based on a large amount of historical deceleration slide table test data and corresponding simulation results. The prediction model is trained using these samples to learn the mapping relationship from the target collision motion parameters to the test parameters. The user only needs to input the desired target collision motion parameters, and the trained prediction model can be called to automatically and quickly output the recommended test parameters matching the target collision motion parameters.

[0036] This method provides engineers with high-precision initial parameter settings, greatly reduces the trial-and-error process based on experience and mathematical derivation in the early stage, significantly reduces the number of parameter iteration rounds, effectively avoids the inefficiency of traditional manual debugging, significantly improves the efficiency of the deceleration slide table test, and ultimately shortens the development cycle of the vehicle restraint system, accelerating the vehicle development process.

[0037] After introducing the basic principles of the present application, various non-limiting embodiments of the present application will be specifically introduced with reference to the accompanying drawings.

[0038] Exemplary System The vehicle collision test system includes a collision test platform vehicle and a collision test device for bearing the impact of the collision test platform vehicle.

[0039] Figure 1 This is a schematic diagram of the overall structure of the crash test trolley according to an embodiment of the present invention. The crash test trolley includes a front frame 3, a middle frame 2, and a rear frame 4, which are constructed and assembled sequentially. The middle frame 2 has a chassis for supporting the body-in-white of the vehicle to be crash tested. Suspension assemblies and wheels 10 mounted on the suspension assemblies are provided on both the left and right sides of the front frame 3 and the rear frame 4. The suspension assembly includes a suspension body 1 and a wheel mounting bracket mounted on the suspension body 1 for mounting the wheels 10.

[0040] The wheelbase, track width, and center of gravity of the crash test trolley can be flexibly adjusted, allowing it to replace different vehicle models in crash tests and realistically reproduce the complete motion postures of a real vehicle during a collision, such as Y-axis tail-swing and Z-axis tail-lift. For example, it can replace actual vehicles in 50 km / h frontal wheel-mounted plate 100% overlap rigid barrier (FRB) crash tests and 50% overlap frontal moving progressive deformable barrier (MPDB) crash tests; it also includes other low-to-medium speed test conditions for frontal and side impact tests, such as 25 km / h FRB crash tests, 35 km / h frontal pole impact tests, and component subsystem tests.

[0041] The crash test apparatus includes a base and alternating full-frontal collision buffer units or front-offset collision buffer units 7 on the base. Each full-frontal collision buffer unit includes multiple first damping mechanisms spaced apart on the base. Each front-offset collision buffer unit 7 includes a second damping mechanism located on the left or right half of the base. Upon impact by the crash test trolley, both the first and second damping mechanisms can collapse and deform to achieve the test effect of a vehicle collision.

[0042] Figure 2 This is a schematic diagram of the structure of the base as described in an embodiment of the present invention when it is equipped with a full frontal collision buffer unit; Figure 3 for Figure 2 The side view of the collision test apparatus shown. Figure 4 for Figure 2 A magnified view of the area shown in B. (See image below.) Figures 2-4As shown, the base comprises a fixed bottom plate 50, a mounting base plate 52, and a support frame 51 connected and supported between the fixed bottom plate 50 and the mounting base plate 52; wherein the fixed bottom plate 50 can be directly fixed on the fixed facilities such as walls in the test site, and the mounting base plate 52 is located on the side that bears the vehicle impact, and is used to arrange the full-frontal impact buffer unit or the offset frontal impact buffer unit 7; the mounting base plate 52 is arranged with a plurality of assembly holes 520 at intervals, and the first damping mechanism and the second damping mechanism are both fixed on the mounting base plate 52 through part of the assembly holes 520; when the assembly holes 520 used for fixing are changed to install the full-frontal impact buffer unit or the offset frontal impact buffer unit 7, the arrangement positions of the first damping mechanism and the second damping mechanism on the mounting base plate 52 can be changed.

[0043] Each first damping mechanism in the full-frontal impact buffer unit is fixed on the mounting base plate 52 through a first fixing seat 60. Specifically, the first fixing seat 60 is provided with a fixed mounting plate 600, which is arranged to extend towards the direction of the test vehicle, and the first damping mechanism is mounted on the fixed mounting plate 600. A plurality of groups of first connecting holes 602 for connecting the first fixing seat 60 are arranged at intervals on the fixed mounting plate 600, and different groups of the first connecting holes 602 are selected to connect the first fixing seat 60, so as to change the connection position of the fixed mounting plate 600 on the first fixing seat 60.

[0044] Each first damping mechanism in the full-frontal impact buffer unit can adopt a unified protruding length specification, or different protruding lengths to form a multi-level damping echelon. Specifically, the plurality of first damping mechanisms comprise a first-level damping mechanism 611 arranged in the middle of the mounting base plate 52, and second-level damping mechanisms 612 arranged on the left half and the right half of the mounting base plate 52 respectively; wherein the first-level damping mechanism 611 is arranged to protrude towards the side of the test vehicle relative to the second-level damping mechanisms 612.

[0045] The primary damping mechanism 611 is connected to the fixed mounting plate 600 via the first extension connecting plate 601, while each secondary damping mechanism 612 is directly fixed to the end of the fixed mounting plate 600. This makes the primary damping mechanism 611 more prominent than the secondary damping mechanisms 612, enabling it to withstand the impact of the vehicle first. Simultaneously, the first extension connecting plate 601 and the fixed mounting plate 600 are connected by a first shear structure 62 that can break under a set impact force. After the primary damping mechanism 611 is crushed by the impact and travels a certain distance, when the impact force reaches the shear force value of the first shear structure 62, the first shear structure 62 will be sheared, causing the first extension connecting plate 601 and the fixed mounting plate 600 to disconnect. During this process, the second-level waveform will be reproduced due to the force cutting of the first shear structure 62. After that, the secondary damping mechanism 612 will be impacted and crushed, thus reproducing the third-level waveform of the maximum peak curve. The synergistic buffering effect of the primary damping mechanism 611, the first shear structure 62 and the secondary damping mechanism 612 will realize the reproduction of the third-level waveform of the vehicle in the full frontal collision process, which is conducive to better presenting the attitude changes and waveform situation in the actual vehicle collision process.

[0046] Figure 5 This is a schematic diagram of the structure of the base described in an embodiment of the present invention when it is provided with a front-biased collision buffer unit. Figure 6 for Figure 5 The front view of the collision test apparatus shown. Figure 7 for Figure 5 A side view of the crash test apparatus shown. Figures 5-7 As shown, the frontal offset collision buffer unit 7 includes a second damping mechanism arranged on the left or right half of the base, so that the frontal offset collision part of the vehicle's front end impacts the frontal offset collision buffer unit 7.

[0047] A large number of mounting holes 520 are spaced apart on the mounting base 52 as mounting positions for the first or second damping mechanism. A primary damper mounting bracket 701 and multiple secondary damper mounting brackets 702 arranged around the primary damper mounting bracket 701 are arranged on the second fixed base 70. A primary damper 71 with a longer protrusion is provided on the primary damper mounting bracket 701, and a secondary damper 72 with a shorter protrusion is provided on each secondary damper mounting bracket 702, thus achieving an arrangement effect similar to the primary damping mechanism 611 and the secondary damping mechanism 612. A guide mechanism 74 is provided between the primary damper 71 and the primary damper mounting bracket 701. The guide mechanism 74 can effectively guide the primary damper 71 to collapse and move along the direction of vehicle impact, so as to effectively apply the impact force to the second shear structure. The synergistic buffering effect of the primary damper 71 and the secondary damper 72 can effectively reproduce the primary and secondary waveforms during a frontal offset collision of the vehicle.

[0048] Specifically, the second fixed seat 70 is provided with a primary damper fixing frame 701 and a plurality of secondary damper fixing frames 702 arranged around the primary damper fixing frame 701; correspondingly, the second damping mechanism includes a primary damper 71 provided on the primary damper fixing frame 701 and a plurality of secondary dampers 72 respectively provided on each secondary damper fixing frame 702; and the primary damper 71 is arranged to protrude towards the side of the test vehicle relative to the secondary damper 72. The second damping mechanism is installed on the mounting base plate 52 by using the second fixed seat 70, which can better realize the assembly between the front offset collision buffer unit 7 and the base; arranging one primary damper fixing frame 701 and a plurality of secondary damper fixing frames 702 on the second fixed seat 70, arranging a primary damper 71 with a greater protruding length on the primary damper fixing frame 701, and arranging a secondary damper 72 with a shorter protruding length on each secondary damper fixing frame 702, can achieve an arrangement effect similar to that of the primary damping mechanism 611 and the secondary damping mechanism 612; the coordinated buffering action of the primary damper 71 and the secondary damper 72 can well reproduce the primary and secondary waveforms during the front offset collision of the vehicle.

[0049] A second shear structure can be used between the second extension connecting plate 73 and the primary damper fixing frame 701; after the primary damper 71 is impacted and crushed to a certain distance, when the impact force reaches the shear force value of the second shear structure, the second shear structure will be sheared off, causing the second extension connecting plate 73 and the primary damper fixing frame 701 to be disconnected; in this process, the force shearing of the second shear structure will reproduce the secondary waveform; only then will the secondary damper 72 be impacted and crushed, thereby reproducing the tertiary waveform of the maximum peak curve.

[0050] Each first damping mechanism, the primary damper 71 and the secondary damper 72 can all be in the form of a damper with a damper base 80; specifically, the first damping mechanism, the primary damper 71 and the secondary damper 72 all include a damper base 80 and a plurality of damping tubes 8 arranged on the damper base 80; wherein the damper base 80 is used to mount the entire damper, and in the front offset collision buffer unit 7, the damper base 80 is fixed on the secondary damper fixing frame 702 or the second extension connecting plate 73, and in the full front collision buffer unit, the damper base 80 is fixed on the fixed mounting plate 600 or the first extension connecting plate 601, and each damping tube 8 on the damper base 80 is arranged to extend towards the oncoming vehicle direction along the impact direction of the vehicle to withstand the impact of the vehicle.

[0051] A detection unit is arranged on the crash test trolley, so that the crash test trolley collides with the crash test device at a set speed, and the acceleration waveform or the impact force-displacement curve of the crash test trolley in the front-rear direction and the acceleration waveforms in the left-right and up-down directions during the collision are obtained by the detection unit; When it is required to reproduce the full frontal collision waveform of the real vehicle, the number, position, length or damping strength of the first damping mechanism arranged on the base is adjusted, so that the acceleration waveforms in the front-rear direction, the left-right direction and the up-down direction obtained are consistent with the real vehicle collision. The crash test trolley can perfectly replace the real vehicle of different vehicle types to perform full frontal collision experiments, and completely reproduce the acceleration changes of the real vehicle and the passengers in the vehicle in the X, Y and Z directions, especially the complete motion posture of the real vehicle in the Y direction during the collision process, such as tail swing in the Y direction and tail swing in the Z direction.

[0052] Similarly, when it is required to reproduce the full frontal collision waveform of the real vehicle, the number, position, length or damping strength of the first damping mechanism arranged on the base is adjusted, so that the acceleration waveforms in the front-rear direction, the left-right direction and the up-down direction obtained are consistent with the real vehicle collision.

[0053] Exemplary Method Figure 8 is a flowchart of a collision test parameter recommendation method provided by an embodiment of the present application. Figure 8 The method is executed by a computing device (for example, a server), but the embodiments of the present application are not limited thereto. The server can be a server, or composed of several servers, or a virtualization platform, or a cloud computing service center, and the embodiments of the present application are not limited thereto. As shown in the figure, the method includes the following contents: Figure 8 The method includes the following contents: Step S810: obtaining a target collision motion parameter of the crash test trolley during a deceleration sliding platform test; the target collision motion parameter is a motion parameter expected to be achieved by the crash test trolley in the deceleration sliding platform test; and the deceleration sliding platform test is achieved by making the crash test trolley impact a damper at the crash test device to achieve the target collision motion parameter.

[0054] In the embodiments of the present application, the deceleration sliding platform test is used to simulate the real vehicle collision process, and the target acceleration waveform or even the vehicle motion posture is reproduced by making the crash test trolley impact the damper at the crash test device, so as to provide a test environment for vehicle restraint system development. The deceleration sliding platform test can include various collision tests under different overlap rates and different speeds, for example, 50km / h and 56km / h full frontal collision tests, 50km / h and 56km / h MPDB collision tests, and 64km / h 40% frontal offset collision tests.

[0055] In the embodiment of the present application, the collision test platform can include a vehicle body-in-white and interior components, and is accelerated to a preset speed by an external driving system and then collides with a collision test device to reproduce the motion state of a real vehicle during the test.

[0056] In the embodiment of the present application, the collision test device includes a base and a plurality of dampers arranged thereon, and the kinetic energy of the collision test platform is absorbed by the crushing deformation of the dampers to simulate the energy absorption characteristics during the real vehicle collision process.

[0057] In the embodiment of the present application, the damper can adopt a crushable damper tube structure, such as a shear aluminum rod. The damper can be installed on the base and absorb collision energy through the sequential crushing deformation of a plurality of damper tubes. The damper can be configured as a multi-stage damper according to the target waveform requirements, for example, the damper is configured as a one-stage damper or a two-stage damper, wherein the one-stage damper is arranged outwardly relative to the two-stage damper and is connected to the fixed frame through an extension connecting plate, and a shear structure between the extension connecting plate and the fixed frame can be broken under a set impact force to achieve the reproduction of a multi-stage waveform.

[0058] In the embodiment of the present application, the target collision motion parameters are used to describe the motion state expected to be achieved by the collision test platform during the deceleration slide test. The target collision motion parameters at least include an acceleration waveform, and in some cases can also include a vehicle tail angle curve and a vehicle rotation angle curve.

[0059] Step S820: input the target collision motion parameters into a target prediction model to obtain recommended test parameters of the deceleration slide test; the training sample of the target prediction model is constructed according to the simulation results of the deceleration slide test; and the recommended test parameters include target parameters of the dampers at the collision test device.

[0060] In the embodiment of the present application, the target prediction model is a parameter mapping model constructed based on a machine learning algorithm. Through the learning of a large amount of historical test data or simulation results, the target prediction model can establish the correlation between the target collision motion parameters and the recommended test parameters, and after receiving the input target collision motion parameters, the target prediction model can quickly calculate and output the corresponding recommended test parameters.

[0061] In the embodiment of the present application, the training sample is formed by the simulation results of the deceleration slide test, wherein the sample is the damper parameter configuration in the simulation scene, and the label is the collision motion parameters in the simulation results.

[0062] In the embodiments of the present application, the recommended test parameters are parameter configuration suggestions output by the target prediction model based on the target collision motion parameters, which are used to directly guide the damper settings of the crash test device. The recommended test parameters can include target parameters of the damper, such as the length, number and damping force of the damper tube, the installation position of the damper, etc.

[0063] In the embodiments of the present application, the simulation results refer to data output obtained after simulating and calculating the deceleration sled test process by computer simulation software, which includes the simulation results of the collision process under specific test parameter settings, such as acceleration waveform, motion posture, etc.

[0064] The crash test parameter recommendation method of the embodiments of the present application constructs a training sample based on the simulation results of the deceleration sled test, trains a special target prediction model, and automatically outputs recommended test parameters containing target parameters of the damper after receiving target collision motion parameters. The target prediction model establishes a parameter mapping relationship through machine learning, so that the output recommended test parameters can highly match the target collision motion parameters, greatly improving the accuracy of parameter configuration, reducing the trial-and-error process based on experience and mathematical derivation in the early stage, greatly reducing the parameter iteration rounds, significantly shortening the preparation time and test cycle of the deceleration sled test, greatly improving the test efficiency, and reducing the material consumption and time cost caused by repeated debugging, indirectly reducing the comprehensive cost of vehicle development.

[0065] Based on the method in the present application, Figure 8 The embodiments of the present application also provide some specific implementation solutions of the method, which are described below.

[0066] Optionally, the method further comprises: generating an experimental design matrix in the parameter space of the target parameters by using an experimental design method; the experimental design matrix is used to represent the simulation input parameter combination of the damper each time the deceleration sled test is simulated; simulating the deceleration sled test based on the experimental design matrix to obtain the simulation results corresponding to each group of simulation input parameter combinations; constructing a training sample for the simulation input parameter combinations and the simulation results corresponding thereto; training an initial prediction model based on the training sample to obtain the target prediction model.

[0067] In the embodiments of the present application, the parameter space of the target parameters refers to a multi-dimensional set composed of all possible values of all to-be-determined target parameters of the damper, which defines the upper and lower boundaries of each parameter and their combination relationship, and covers all possible damper configuration situations in the deceleration sled test.

[0068] In the embodiments of the present application, the Design of Experiments Matrix (DOE Matrix) is a parameter combination table generated by an experimental design method, and each row in the matrix represents a complete simulation input parameter combination.

[0069] In the embodiments of the present application, the simulation input parameter combination refers to a specific parameter setting determined based on the experimental design matrix for a single deceleration sled simulation test, which explicitly defines the specific parameter values of each damper in this simulation and is the direct input condition for driving a single simulation run and generating corresponding results.

[0070] In the embodiments of the present application, the experimental design method refers to a systematic parameter sampling strategy based on statistical principles, such as Latin hypercube sampling or orthogonal experimental design, which aims to efficiently explore the entire possible value range of damper key parameters with limited simulation times and ensure that the sampling points are uniformly distributed and representative in the parameter space. The experimental design method is used to reduce the number of simulation input parameter combinations while ensuring that the experimental design matrix covers the parameter space.

[0071] In the embodiments of the present application, the simulation results are output data obtained through deceleration sled simulation tests, which can include vehicle acceleration waveforms, vehicle tail swing angle curves, and vehicle rotation angle curves calculated under specific simulation input parameter combinations. The simulation results are used to reflect the motion response characteristics of the crash test sled under simulation conditions.

[0072] In the embodiments of the present application, the training samples are data pairs composed of simulation input parameter combinations and their corresponding simulation results, and each sample completely records the input conditions and output responses of a simulation test. The training samples are used to reflect the internal mapping relationship between damper parameter configurations and collision motion responses.

[0073] In the embodiments of the present application, the initial prediction model can be a parameter search model based on optimization algorithms, such as simulated annealing algorithm or genetic algorithm, etc. The initial prediction model can automatically find the optimal solution in the parameter space to obtain the best test parameter configuration corresponding to the target collision motion parameters.

[0074] In the embodiments of the present application, an experimental design method is adopted to generate an experimental design matrix in the parameter space of the target parameters, to reduce the number of simulation input parameter combinations under the premise of ensuring coverage of the entire parameter value range, to carry out deceleration slide table simulation and construct training samples based on the matrix, and to train the target prediction model. Through the systematic sampling advantage of the experimental design method, blind traversal of the parameter space is avoided, and the required training sample size and the corresponding simulation times are greatly reduced, not only reducing the calculation resource consumption and time cost in the simulation process, but also reducing the data processing workload; at the same time, the uniform and representative training samples can enable the model to more comprehensively learn the correlation between the parameters and the collision effect, further ensuring the accuracy and reliability of the target prediction model output recommended test parameters, and ultimately reducing the research and development investment while improving the overall efficiency of the deceleration slide table test.

[0075] Optionally, the target collision motion parameters include a test type of the deceleration slide table test, and an acceleration waveform of the collision test trolley in the front-rear direction; and the recommended test parameters include a number of dampers at the collision test device, a target length of each damper, and a damping strength.

[0076] In the embodiments of the present application, the test type refers to a specific collision working condition simulated by the deceleration slide table test, which can include 50km / h frontal full overlap rate collision, 56km / h MPDB (Mobile Progressive Deformable Barrier) collision, and 64km / h, 40% offset collision, and different test conditions.

[0077] In the embodiments of the present application, the acceleration waveform refers to a curve of the front-rear direction (X direction) acceleration of the collision test trolley changing with time in the collision process, which is the core target parameter for reproducing the real vehicle collision effect.

[0078] In the embodiments of the present application, the target length refers to a preset length specification set for each damper in the recommended test parameters.

[0079] In the embodiments of the present application, the damping strength refers to the resistance provided by the damper in the collision process, which is mainly realized through the shear force value of the damper. This parameter, together with the length of the damper tube, determines the force-displacement characteristics of the damper in the collision process.

[0080] The engineers found that when the test type is determined, the vehicle acceleration waveform is mainly related to the number of dampers, the target length of each damper, and the damping force, and the specific installation position of the damer has relatively small influence on the waveform reproduction accuracy. Based on this finding, in the embodiments of the present application, by learning the accurate mapping relationship from the test type and the acceleration waveform to the number, length and damping force of the damper, the dimension of the parameter space is significantly reduced, and the size of the experimental design matrix is reduced. Under the premise of ensuring coverage, the number of simulation samples required is greatly reduced, not only saving computing resources, but also making the data distribution of the training samples more concentrated, effectively reducing the learning difficulty of the target prediction model, and improving the convergence speed of the model training.

[0081] During model training, for the deceleration sliding platform test of each test type, the number of dampers is taken as the first training variable, and the length configuration and corresponding damping force of each damper are taken as the second training variable; the optimal Latin hypercube experimental design method is used to generate an experimental design matrix in the parameter space composed of the above training variables, which significantly reduces the number of simulation samples required while ensuring the coverage of the parameter space. Based on the experimental design matrix, deceleration sliding platform simulation analysis is carried out to obtain the data relationship between the length of the damper tube, the shear force value of the damper, and the collision waveform corresponding to each simulation input parameter combination. When constructing the training sample, not only the simulation data results are included, but also the stress-strain curve of the damper tube and the acceleration characteristic data of a single damper tube are introduced, and the stress-strain curve of the aluminum bar is also included, so that the model can fully understand the material mechanics characteristics. The training sample is used to train the initial prediction model to obtain a target prediction model that can accurately recommend the number, length and shear force value of the damper tube. Finally, the model recommended scheme is compared and verified based on the actual test results, and the prediction model is iteratively corrected through feedback data to continuously improve the prediction accuracy and reliability of the model in actual application.

[0082] Optionally, the target collision motion parameter includes a test type of the deceleration sliding platform test, an acceleration waveform of the collision test platform vehicle in the front-rear direction, and tailing data and rotation data of the collision test platform vehicle; and the recommended test parameter includes a number of dampers at the collision test device, a target length of each damper, a damping force, and a position.

[0083] In the embodiments of the present application, the test type refers to the specific collision working condition simulated by the deceleration sliding platform test, including 50km / h frontal full overlap rate collision, 56km / h MPDB (Mobile Progressive Deformable Barrier) collision, and 64km / h, 40% offset collision, and different test conditions.

[0084] In the embodiment of the present application, the acceleration waveform refers to a curve of the front-rear direction (X direction) acceleration of the crash test trolley changing with time in the crash process, and is a core target parameter for reproducing the real vehicle crash effect.

[0085] In the embodiment of the present application, the target length refers to a preset length specification set for each damper in the recommended test parameters.

[0086] In the embodiment of the present application, the damping force refers to the resistance provided by the damper in the crash process, which is mainly realized by the shear force value of the damper. This parameter, together with the length of the damping tube, determines the force-displacement characteristic of the damper in the crash process.

[0087] In the embodiment of the present application, the tail-up data refers to the attitude change data of the crash test trolley in the up-down direction (Z direction) in the crash process, which can be specifically a curve of the tail-up angle changing with time or displacement. It is a key parameter for reproducing the three-dimensional motion attitude of the whole vehicle in the crash, and directly reflects the longitudinal pitch state of the trolley in the crash.

[0088] In the embodiment of the present application, the rotation data refers to the tail-swing rotation data of the crash test trolley in the left-right direction (Y direction) in the crash process, which can be specifically a curve of the rotation angle changing with time or displacement. It is used for accurately reproducing the lateral rotation motion state of the whole vehicle in the crash, and is an important part of the complete reproduction of the real vehicle crash attitude.

[0089] In the embodiment of the present application, the position of the damper refers to the specific arrangement direction of the damper on the mounting substrate of the crash test device.

[0090] In the model training process, the number of dampers, the target length, the damping force and the mounting position of each damper are taken as a complete set of training variables. An experimental design method is used to generate an experimental design matrix in the parameter space composed of all target parameters. By simultaneously considering all the above variable parameters, a complete parameter mapping relationship including the position factor is established, so that the prediction model obtained by training can accurately reflect the comprehensive influence of the damper configuration on the acceleration waveform, the tail-up data and the rotation data. This full-variable training method significantly enhances the prediction ability of the model for complex three-dimensional motion characteristics such as vehicle tail-swing and tail-up, and ensures the completeness of the vehicle motion attitude reproduction in the deceleration skid test.

[0091] In the embodiment of the present application, the test type, the acceleration waveform, the tail-up data and the rotation data are taken as the target crash motion parameters, and the number of dampers, the target length, the damping force and the position are all included in the recommended test parameter system. A complete parameter mapping relationship is established. This method can accurately reproduce the acceleration waveform and three-dimensional motion attitude of the vehicle in the crash process at the same time, and significantly improves the overall reproduction accuracy of the deceleration skid test.

[0092] Optionally, the target prediction model includes a first prediction model and a second prediction model; The step of inputting the target collision motion parameters into the target prediction model to obtain the recommended test parameters for the deceleration slide test includes: The test type and acceleration waveform of the deceleration slide test are input into the first prediction model to obtain the number of dampers, the target length of each damper, and the shear force value. The tail-end data and rotation data of the collision test trolley are input into the second prediction model to obtain the position of each damper.

[0093] In this embodiment of the application, the first prediction model is used to predict the number of dampers, the target length of each damper, and the shear force value based on the test type and acceleration waveform of the deceleration slide test; for its training process, please refer to the training process of the target prediction model that only recommends parameters based on the acceleration waveform above.

[0094] In this embodiment, the second prediction model is used to predict the position of each damper based on the tail-end data and rotation data of the crash test trolley. The permissible installation positions of the dampers are limited to multiple preset positions on the crash test device. The total thrust magnitude and its time history curve are calculated by selecting representative whole-vehicle acceleration waveforms. Based on this, the optimal Latin hypercube experimental design method is used to generate a DOE matrix containing different combinations of damper installation positions and thrust magnitudes. A deceleration slide simulation analysis is performed on each parameter combination sample in the matrix to obtain the corresponding tail-end data and rotation data results. These simulation data are used as training samples to input into the initial model for training, thereby constructing the correspondence between the damper installation position and the three-dimensional motion response. Finally, actual experimental data is used to provide feedback correction to the initially trained prediction model, continuously improving the model's prediction accuracy for complex motion postures.

[0095] Engineers discovered that the test type and acceleration waveform of the deceleration slide test are highly correlated with the number of dampers, target length, and shear force value, while tail-end data and rotational data are more correlated with the position of the dampers. Therefore, this application adopts a scheme of training the first prediction model and the second prediction model separately. The first prediction model is used to learn the mapping relationship between the test type and acceleration waveform and the number, length, and shear force value of the dampers, while the second prediction model is used to learn the mapping relationship between tail-end data, rotational data, and the position of the dampers. This results in an exponential reduction in the parameter space of each model, significantly reducing the size of the experimental design matrix and greatly reducing the number of simulation samples required. This not only effectively saves computing resources but also makes the training sample data distribution of each model more concentrated, greatly reducing the learning difficulty of the model and significantly improving the convergence speed of model training, while maintaining the accuracy of the prediction of each specific parameter.

[0096] Optionally, the step of employing an experimental design method to generate an experimental design matrix within the parameter space of the target parameters includes: The optimal Latin hypercube algorithm is used to generate the experimental design matrix in the parameter space of the target parameters.

[0097] In this embodiment, the optimal Latin hypercube algorithm is an improved space-filling experimental design strategy. Its core lies in enhancing basic Latin hypercube sampling by introducing optimization criteria to generate a more uniform array of sample points within the parameter space. This method first uniformly divides the value range of each damper target parameter to be optimized into an equal number of intervals, generating an initial Latin hypercube design. This ensures that each parameter has only one sample point (i.e., the simulation input parameter combination) projected within each interval, thus achieving uniform coverage within the parameter space. Subsequently, the method applies a specific optimization algorithm and iteratively optimizes the spatial location of the sample points in the initial design according to the selected optimization criteria. The goal is to make the overall distribution of all sample points in the multidimensional parameter space as uniform as possible, avoiding clustering or blank areas. Compared to basic Latin hypercube sampling, the optimal Latin hypercube algorithm, through this optimization process, can significantly improve the representativeness of parameter combinations while maintaining the uniformity of projection in each parameter dimension.

[0098] In this embodiment, the optimal Latin hypercube algorithm is used to generate an experimental design matrix in the parameter space of the target parameters. This can obtain a more representative sample coverage of the high-dimensional design space composed of key parameters such as the number, length, damping force and position of dampers with fewer simulations. Thus, while ensuring the coverage of training samples and the accuracy of the model, the simulation computing resources required to build the target prediction model are effectively reduced, and the model training efficiency is improved.

[0099] Optionally, the method further includes: Based on the recommended test parameters, a deceleration slide test was conducted to obtain the actual collision motion parameters; The target prediction model is modified based on the actual collision motion parameters.

[0100] In this embodiment of the application, the actual collision motion parameters refer to the motion characteristic data of the collision test trolley actually collected by the measuring equipment after the deceleration slide test is conducted based on the recommended test parameters. These parameters may include the actual obtained vehicle acceleration waveform, tail-end angle curve, and rotation angle curve.

[0101] In this embodiment of the application, the actual collision motion parameters obtained from the deceleration slide test based on the recommended test parameters are compared and analyzed with the target collision motion parameters set before the test, and the deviation between the two is calculated. Then, based on the deviation, the internal parameters of the target prediction model are adjusted to reduce the deviation between the model prediction value and the actual test result.

[0102] In this embodiment of the application, the target prediction model is modified based on the actual collision motion parameters.

[0103] In this embodiment, a closed-loop model optimization process is constructed by applying the model-recommended parameters to physical experiments and using measured data for feedback correction. This process enables the target prediction model to continuously learn the response characteristics of the real physical system, gradually correcting the deviation between simulation and measurement, thereby continuously improving the model's prediction accuracy under actual working conditions.

[0104] Optionally, the step of inputting the target collision motion parameters into the target prediction model to obtain the recommended test parameters for the deceleration slide test includes: Based on the target agent calling the target prediction model, the recommended test parameters for the deceleration slide test are obtained.

[0105] In this embodiment, the target agent is used to automatically call the corresponding prediction model and generate recommended test parameters according to preset user input parameters and a preset process. For example, the target agent sequentially calls the first prediction model and the second prediction model, and generates a response message containing the number of dampers, target length, shear force value, and installation position based on the user-input test type, acceleration waveform, tail data, and rotation data. This message carries a complete recommended configuration scheme for the deceleration slide test.

[0106] In this embodiment of the application, the target intelligent agent is also used to execute the model training process, automatically generate an experimental design matrix according to a preset training strategy, drive the simulation analysis process, construct training samples and complete the training task of the prediction model, so as to realize the fully automated management of the entire process from data preparation to model deployment.

[0107] In this embodiment, a target intelligent agent is introduced to automatically invoke the prediction model. The user only needs to input the target collision motion parameters, and the target intelligent agent can automatically invoke the target prediction model according to a preset process to obtain complete recommended experimental parameters. This process eliminates the need for manual intervention in model selection and data processing, significantly lowering the barrier to entry for the system. This allows engineers to easily obtain professional-grade parameter recommendations without needing to deeply understand the model's internal structure and invocation logic, effectively improving usability in practical engineering applications.

[0108] Exemplary Apparatus The apparatus embodiments of this application can be used to execute the method embodiments of this application. For details not disclosed in the apparatus embodiments of this application, please refer to the method embodiments of this application.

[0109] Figure 9 The diagram shown is a block diagram of a collision test parameter recommendation device provided in one embodiment of this application. Figure 9 As shown, the device 900 includes: The acquisition module 910 is used to acquire the target collision motion parameters of the collision test trolley during the deceleration slide test; the target collision motion parameters are the motion parameters that the collision test trolley is expected to achieve in the deceleration slide test; the deceleration slide test is performed by causing the collision test trolley to hit the damper at the collision test device to achieve the target collision motion parameters. The prediction module 920 is used to input the target collision motion parameters into the target prediction model to obtain the recommended test parameters for the deceleration slide test; the training samples of the target prediction model are constructed based on the simulation results of the deceleration slide test; the recommended test parameters include the target parameters of the damper at the collision test device.

[0110] Optionally, the method further includes: The matrix generation module is used to generate an experimental design matrix in the parameter space of the target parameters using an experimental design method; the experimental design matrix is ​​used to represent the combination of simulation input parameters of the damper each time the deceleration slide test is simulated. The simulation module is used to simulate the deceleration slide test based on the experimental design matrix and obtain the simulation results corresponding to each set of simulation input parameter combinations. The sample module is used to construct training samples for the combination of simulation input parameters and their corresponding simulation results; The training module is used to train the initial prediction model based on the training samples to obtain the target prediction model.

[0111] Optionally, the target collision motion parameters include the test type of the deceleration slide test and the acceleration waveform of the collision test trolley in the longitudinal direction; the recommended test parameters include the number of dampers at the collision test device, the target length of each damper, and the damping force.

[0112] Optionally, the target collision motion parameters include the test type of the deceleration slide test, the acceleration waveform of the collision test trolley in the longitudinal direction, and the tail-end data and rotation data of the collision test trolley; the recommended test parameters include the number of dampers at the collision test device, the target length of each damper, the damping force and position.

[0113] Optionally, the target prediction model includes a first prediction model and a second prediction model; The prediction module 920 is used for: The test type and acceleration waveform of the deceleration slide test are input into the first prediction model to obtain the number of dampers, the target length of each damper, and the shear force value. The tail-end data and rotation data of the collision test trolley are input into the second prediction model to obtain the position of each damper.

[0114] Optionally, the matrix generation module is used for: The optimal Latin hypercube algorithm is used to generate the experimental design matrix in the parameter space of the target parameters.

[0115] Optionally, the method further includes: The correction module is used to conduct a deceleration slide test based on the recommended test parameters to obtain the actual collision motion parameters; and to correct the target prediction model based on the actual collision motion parameters.

[0116] Optionally, the prediction module 920 is used for: Based on the target agent calling the target prediction model, the recommended test parameters for the deceleration slide test are obtained.

[0117] Exemplary Electronic Device Below, for reference Figure 10 This describes an electronic device according to embodiments of the present application. Figure 10 A block diagram of an electronic device according to an embodiment of this application is illustrated.

[0118] like Figure 10 As shown, the electronic device 1000 includes one or more processors 1010 and memory 1020.

[0119] The processor 1010 may be another form of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 1000 to perform desired functions.

[0120] Specifically, processor 1010 can be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits used to control the execution of the program of the present invention. It can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. Processor 1010 may also include a main processor, and may also include a baseband chip, a modem, etc.

[0121] The memory 1020 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 1010 may execute the program instructions to implement the collision test parameter recommendation methods of the various embodiments of this application described above, and / or other desired functions. Various contents, such as category correspondences, may also be stored in the computer-readable storage medium.

[0122] In one example, the electronic device 1000 may also include an input device 1030 and an output device 1040, which are interconnected via a bus system and / or other forms of connection mechanism (not shown).

[0123] In addition, the input device 1030 can also be a device that receives user input data and information, such as a keyboard, mouse, camera, scanner, light pen, voice input device, touch screen, pedometer, or gravity sensor. The output device 1040 can output various information to the outside. The output device 1040 may include, for example, a display, speaker, printer, and communication network and its connected remote output devices, etc.

[0124] Of course, for the sake of simplicity, Figure 10 Only some of the components of the electronic device 1000 relevant to this application are shown in this illustration; components such as buses, input / output interfaces, etc., are omitted. In addition, the electronic device 1000 may include any other suitable components depending on the specific application.

[0125] Exemplary Computer Program Product and Computer-Readable Storage Medium In addition to the methods and apparatus described above, embodiments of this application may also be computer program products, which include computer program instructions that, when executed by a processor, cause the processor to perform the steps in the collision test parameter recommendation methods according to various embodiments of this application described in the "Exemplary Methods" section of this specification.

[0126] The computer program product can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this application. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0127] Furthermore, embodiments of this application may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the steps in the collision test parameter recommendation method according to various embodiments of this application described in the "Exemplary Methods" section above.

[0128] The computer-readable storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0129] The basic principles of this application have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this application are merely examples and not limitations, and should not be considered as essential features of each embodiment of this application. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the application to the necessity of employing the aforementioned specific details for implementation.

[0130] For the foregoing method embodiments, in order to simplify the description, they are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0131] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For apparatus embodiments, since they are basically similar to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0132] The steps in the methods of the various embodiments of this application can be adjusted, merged, or deleted in order according to actual needs, and the technical features described in each embodiment can be replaced or combined.

[0133] The block diagrams of devices, apparatuses, devices, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0134] It should also be noted that in the apparatus, equipment, and methods of this application, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions of this application.

[0135] The modules or submodules described as separate components may or may not be physically separate. The components that constitute a module or submodule may or may not be physical modules or submodules; that is, they may be located in one place or distributed across multiple network modules or submodules. Some or all of the modules or submodules can be selected to achieve the purpose of this embodiment's solution, depending on actual needs.

[0136] Furthermore, the functional modules or sub-modules in the various embodiments of this application can be integrated into one processing module, or each module or sub-module can exist physically separately, or two or more modules or sub-modules can be integrated into one module. The integrated modules or sub-modules described above can be implemented in hardware or in the form of software functional modules or sub-modules.

[0137] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0138] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software unit executed by a processor, or a combination of both. The software unit can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0139] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0140] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for recommending collision test parameters, characterized in that, include: The target collision motion parameters of the collision test rig are obtained when the rig is performing a deceleration slide test. The target collision motion parameters are the motion parameters that the rig is expected to achieve in the deceleration slide test. The deceleration slide test is performed by causing the rig to impact the damper at the collision test device to achieve the target collision motion parameters. The target collision motion parameters are input into the target prediction model to obtain the recommended test parameters for the deceleration slide test; the training samples of the target prediction model are constructed based on the simulation results of the deceleration slide test; the recommended test parameters include the target parameters of the damper at the collision test device.

2. The method according to claim 1, characterized in that, The method further includes: An experimental design matrix is ​​generated in the parameter space of the target parameters using an experimental design method; the experimental design matrix is ​​used to represent the combination of simulation input parameters of the damper each time the deceleration slide test is simulated. Based on the experimental design matrix, the deceleration slide test was simulated to obtain the simulation results corresponding to each set of simulation input parameter combinations. Training samples are constructed based on the combination of simulation input parameters and their corresponding simulation results; Based on the training samples, the initial prediction model is trained to obtain the target prediction model.

3. The method according to claim 2, characterized in that, The target collision motion parameters include the test type of the deceleration slide test and the acceleration waveform of the collision test trolley in the forward and backward directions; the recommended test parameters include the number of dampers at the collision test device, the target length of each damper, and the damping force.

4. The method according to claim 2, characterized in that, The target collision motion parameters include the test type of the deceleration slide test, the acceleration waveform of the collision test trolley in the forward and backward directions, and the tail-lift data and rotation data of the collision test trolley; the recommended test parameters include the number of dampers at the collision test device, the target length of each damper, the damping force and position.

5. The method according to claim 4, characterized in that, The target prediction model includes a first prediction model and a second prediction model; The step of inputting the target collision motion parameters into the target prediction model to obtain the recommended test parameters for the deceleration slide test includes: The test type and acceleration waveform of the deceleration slide test are input into the first prediction model to obtain the number of dampers, the target length of each damper, and the shear force value. The tail-end data and rotation data of the collision test trolley are input into the second prediction model to obtain the position of each damper.

6. The method according to claim 2, characterized in that, The step of employing experimental design methods to generate an experimental design matrix within the parameter space of the target parameters includes: The optimal Latin hypercube algorithm is used to generate the experimental design matrix in the parameter space of the target parameters.

7. The method according to claim 2, characterized in that, The method further includes: Based on the recommended test parameters, a deceleration slide test was conducted to obtain the actual collision motion parameters; The target prediction model is modified based on the actual collision motion parameters.

8. The method according to claim 1, characterized in that, The step of inputting the target collision motion parameters into the target prediction model to obtain the recommended test parameters for the deceleration slide test includes: Based on the target agent calling the target prediction model, the recommended test parameters for the deceleration slide test are obtained.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions that, when executed by a processor, cause the processor to perform the method as described in any one of claims 1 to 8.

10. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to perform the method according to any one of claims 1 to 8.

Images