Acceleration curve prediction method, device, equipment and readable storage medium

Abstract

The present invention provides an acceleration curve prediction method, device, equipment and readable storage medium. The acceleration curve prediction method includes: obtaining a training data set, the training data set includes multiple sets of training data pairs, each set of training data pairs consists of hood structural parameters and acceleration curves corresponding to the hood structural parameters; respectively combining the acceleration value at each moment with the corresponding hood structural parameters; respectively inputting multiple sets of training data pairs at each moment into a machine model for training; respectively inputting new hood structural parameters into the trained machine model at each moment to obtain the predicted acceleration value at each moment; and constructing a predicted acceleration curve. Through the present invention, the prediction of a two-dimensional parameter such as an acceleration curve is realized, which solves the problem that the existing machine learning algorithm can only learn to predict a single value. Compared with the simulation collision test and response surface method in the prior art, the prediction efficiency and the prediction accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an acceleration curve prediction method, device, equipment and readable storage medium. Background Art

[0002] As more and more attention is paid to the protection of pedestrians' heads in traffic collision accidents, engineers need to design different hood (i.e., vehicle front hood) structure models to reduce the damage to pedestrians' heads caused by the collision between the hood and pedestrians' heads. Among them, the acceleration curve generated by the collision between the vehicle hood and the pedestrian's head is an important reference indicator. It takes a lot of time and computing resources to obtain the acceleration curve through simulation of pedestrian head collision experiments. If the hood structure needs to be changed and optimized, the simulation calculation needs to be resubmitted, which is time-consuming and laborious. Later, the introduction of parametric models greatly improved the efficiency of hood structure optimization design. Engineers can set relevant important hood structure parameters through experience, and then simulate by changing the parameters to obtain a large number of output results. Then, based on these results, the response surface method is completed to establish a continuous variable surface model. The purpose is to predict the output results without simulation. However, the problem with this method is that it still requires manual selection of the method for constructing the response surface.

[0003] Support Vector Machine (SVM) is a generalized linear classification algorithm that performs binary classification of data in a supervised learning manner. It has been widely used in pattern recognition problems such as portrait recognition and text classification. However, current machine learning methods predict single values, and it is impossible to predict two-dimensional parameters such as the acceleration curve obtained from the head collision simulation in pedestrian protection. Summary of the invention

[0004] The main purpose of the present invention is to provide an acceleration curve prediction method, device, equipment and readable storage medium, aiming to solve the technical problem in the prior art that a two-dimensional parameter such as an acceleration curve cannot be predicted.

[0005] In a first aspect, the present invention provides an acceleration curve prediction method, the acceleration curve prediction method comprising:

[0006] Acquire a training data set, wherein the training data set includes a plurality of training data pairs, each training data pair consisting of a hood structure parameter and an acceleration curve corresponding to the hood structure parameter;

[0007] According to a preset period, extract multiple moments and the acceleration value corresponding to each moment from the acceleration curve of each set of training data pairs;

[0008] The acceleration value at each moment is combined with the corresponding hood structure parameter to obtain multiple sets of training data pairs at each moment;

[0009] The multiple sets of training data at each moment are respectively used to train the input machine model to obtain the trained machine model at each moment;

[0010] The new hood structure parameters are input into the trained machine model at each moment to obtain the predicted acceleration value at each moment;

[0011] Based on the predicted acceleration value at each moment, a predicted acceleration curve is constructed.

[0012] Optionally, before obtaining the training data set, the following steps are included:

[0013] By setting the attributes of the hood structure parameters and the values ​​corresponding to the attributes, a plurality of groups of hood structure parameters are obtained, wherein each group of hood structure parameters constitutes a hood structure model;

[0014] Through the simulation software, multiple hood structure models are subjected to simulated collision tests respectively, and the acceleration curve corresponding to each hood structure model after the collision is obtained;

[0015] The training data pairs are formed by combining each set of hood structural parameters and the acceleration curve corresponding to each set of hood structural parameters;

[0016] Construct multiple sets of training data pairs as training data sets.

[0017] Optionally, the simulation software is used to perform a simulation collision test on multiple hood structure models respectively, and the acceleration curve corresponding to each hood structure model after the collision is obtained, including:

[0018] Through simulation software, multiple hood structure models are subjected to simulated collision tests at multiple preset collision points to obtain the acceleration curve corresponding to each hood structure model at each collision point after the collision.

[0019] Optionally, the acceleration curve prediction method further includes:

[0020] A training data set is constructed for each collision point, and the machine model training is performed to construct the predicted acceleration curve of the new hood parameters at each collision point.

[0021] Optionally, the step of training the machine model by inputting multiple sets of training data at each moment includes:

[0022] The support vector machine algorithm in the sklearn machine learning library is called through the Python programming language to train multiple sets of training data pairs at each moment.

[0023] In a second aspect, the present invention further provides an acceleration curve prediction device, the acceleration curve prediction device comprising:

[0024] An acquisition module, used for acquiring a training data set, wherein the training data set includes a plurality of training data pairs, each training data pair consisting of a hood structure parameter and an acceleration curve corresponding to the hood structure parameter;

[0025] An extraction module, used to extract multiple moments and the acceleration value corresponding to each moment from the acceleration curve of each set of training data pairs according to a preset period;

[0026] A combination module, used to respectively combine the acceleration value at each moment with the corresponding hood structure parameter to obtain multiple sets of training data pairs at each moment;

[0027] The training module is used to train the machine model with multiple sets of training data at each moment, so as to obtain a trained machine model at each moment;

[0028] A prediction module is used to input new hood structure parameters into the trained machine model at each moment to obtain the predicted acceleration value at each moment;

[0029] The building module is used to build a predicted acceleration curve based on the predicted acceleration value at each moment.

[0030] Optionally, the acceleration curve prediction device further includes a training data set construction module, which is used to:

[0031] By setting the attributes of the hood structure parameters and the values ​​corresponding to the attributes, a plurality of groups of hood structure parameters are obtained, wherein each group of hood structure parameters constitutes a hood structure model;

[0032] Through the simulation software, multiple hood structure models are subjected to simulated collision tests respectively, and the acceleration curve corresponding to each hood structure model after the collision is obtained;

[0033] The training data pairs are formed by combining each set of hood structural parameters and the acceleration curve corresponding to each set of hood structural parameters;

[0034] Construct multiple sets of training data pairs as training data sets.

[0035] Optionally, the training module is used to:

[0036] The support vector machine algorithm in the sklearn machine learning library is called through the Python programming language to train multiple sets of training data pairs at each moment.

[0037] In a third aspect, the present invention also provides an acceleration curve prediction device, comprising a processor, a memory, and an acceleration curve prediction program stored in the memory and executable by the processor, wherein when the acceleration curve prediction program is executed by the processor, the steps of the acceleration curve prediction method described above are implemented.

[0038] In a fourth aspect, the present invention further provides a readable storage medium, on which an acceleration curve prediction program is stored, wherein when the acceleration curve prediction program is executed by a processor, the steps of the acceleration curve prediction method as described above are implemented.

[0039] In the present invention, a training data set is obtained, and the training data set includes multiple groups of training data pairs, each group of training data pairs consists of hood structural parameters and acceleration curves corresponding to the hood structural parameters; according to a preset period, multiple moments and acceleration values ​​corresponding to each moment are extracted from the acceleration curve of each group of training data pairs; the acceleration value at each moment is combined with the corresponding hood structural parameters to obtain multiple groups of training data pairs at each moment; the multiple groups of training data pairs at each moment are input into a machine model for training to obtain a trained machine model at each moment; new hood structural parameters are input into the trained machine model at each moment to obtain a predicted acceleration value at each moment; based on the predicted acceleration value at each moment, a predicted acceleration curve is constructed. The present invention first obtains a training data set, wherein the training data set includes multiple sets of training data pairs, each set of training data pairs is composed of a hood structure parameter and an acceleration curve corresponding to the hood structure parameter, that is, an acceleration curve generated after a collision between a hood structure model composed of different hood structure parameters and a pedestrian head, and then decomposes and extracts multiple moments and acceleration values ​​corresponding to each moment from the acceleration curve according to a preset period, and then combines multiple sets of one-to-one corresponding acceleration values ​​in each moment with the corresponding hood structure parameters to obtain multiple sets of training data pairs at each moment, and inputs multiple sets of training data pairs at each moment into a machine model for training according to the extracted moments to obtain a trained machine model at each moment, and inputs new hood structure parameters into the trained machine model at each moment to obtain a predicted acceleration value at each moment, and finally constructs a predicted acceleration curve based on the predicted acceleration value at each moment and the moment value extracted according to the preset period. Through the present invention, the prediction of a two-dimensional parameter such as an acceleration curve is realized, and the problem that the existing machine learning algorithm can only learn to predict a single value is solved. Compared with the simulation collision test and the response surface method in the prior art, the prediction efficiency and the prediction accuracy are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1A schematic diagram of the hardware structure of an acceleration curve prediction device according to an embodiment of the present invention;

[0041] Figure 2 It is a schematic diagram of a flow chart of an embodiment of an acceleration curve prediction method of the present invention;

[0042] Figure 3 A schematic diagram of a training data set construction process for an acceleration curve prediction method according to an embodiment of the present invention;

[0043] Figure 4 It is a schematic diagram of functional modules of an acceleration curve prediction device according to an embodiment of the present invention;

[0044] Figure 5 Schematic diagram of a training data set construction module of an acceleration curve prediction device according to an embodiment of the present invention.

[0045] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0046] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0047] In a first aspect, an embodiment of the present invention provides an acceleration curve prediction device.

[0048] Reference Figure 1 , Figure 1 The hardware structure diagram of an embodiment of the acceleration curve prediction device of the present invention. In the embodiment of the present invention, the acceleration curve prediction device may include a processor 1001 (e.g., a central processing unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components; the user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard); the network interface 1004 may optionally include a standard wired interface, a wireless interface (such as a wireless fidelity WIreless-FIdelity, WI-FI interface); the memory 1005 may be a high-speed random access memory (random access memory, RAM), or a stable memory (non-volatile memory), such as a disk storage, and the memory 1005 may optionally be a storage device independent of the aforementioned processor 1001. Those skilled in the art will understand that Figure 1 The hardware structure shown in the figure does not constitute a limitation of the present invention, and may include more or less components than those shown in the figure, or combine certain components, or arrange the components differently.

[0049] Continue to refer to Figure 1 , Figure 1 The memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an acceleration curve prediction program. The processor 1001 may call the acceleration curve prediction program stored in the memory 1005 and execute the acceleration curve prediction method provided by the embodiment of the present invention.

[0050] In a second aspect, an embodiment of the present invention provides an acceleration curve prediction method.

[0051] In order to more clearly demonstrate the acceleration curve prediction method provided in the embodiment of the present application, the application scenario of the acceleration curve prediction method provided in the embodiment of the present application is first introduced.

[0052] The acceleration curve prediction method provided in the embodiment of the present application is applied in the existing collision protection of pedestrian heads. It is necessary to obtain the acceleration curve formed after the collision between the pedestrian head and the vehicle hood, so as to evaluate the different degrees of damage caused to the pedestrian head by different hood structures, so as to better improve and design the hood structure.

[0053] In one embodiment, referring to Figure 2 , Figure 2 FIG. 1 is a flow chart of an embodiment of an acceleration curve prediction method of the present invention. Figure 2 As shown, the acceleration curve prediction method includes:

[0054] Step S10, obtaining a training data set, wherein the training data set includes a plurality of training data pairs, each training data pair consisting of a hood structure parameter and an acceleration curve corresponding to the hood structure parameter.

[0055] In this embodiment, a training data set is first obtained through a simulation test scenario or other channels. The specific method of obtaining the training data set is not limited in this embodiment. The training data set includes multiple groups of training data pairs, each group of training data pairs consists of hood structure parameters and acceleration curves corresponding to the hood structure parameters. Different hood structure parameters constitute different hood structure models, which correspond one-to-one to the acceleration curves generated after a collision with a pedestrian's head. The hood is the front engine cover of a car. Usually, different hood structure models can be obtained by adjusting and setting different hood structure parameters, such as changing the height and width of the reinforcement ribs in the hood and the length and width of the hollow shape.

[0056] Step S20: extracting multiple moments and the acceleration value corresponding to each moment from the acceleration curve of each set of training data pairs according to a preset period.

[0057] In this embodiment, the acceleration curve is a two-dimensional variable, which is composed of each moment and the acceleration value corresponding to each moment. According to a preset period, multiple moments and the acceleration values ​​corresponding to each moment are extracted from the acceleration curve, such as 1 second after the collision and the acceleration value corresponding to 1 second after the collision, 2 seconds after the collision and the acceleration value corresponding to 2 seconds after the collision, 3 seconds after the collision and the acceleration value corresponding to 3 seconds after the collision, etc.

[0058] Step S30, respectively combining the acceleration value at each moment with the corresponding hood structure parameter to obtain multiple sets of training data pairs at each moment.

[0059] In this embodiment, the acceleration value at each moment and its corresponding hood structure parameter at the same moment are combined. Since there are multiple sets of hood structure parameters and the acceleration curves corresponding to each hood structure parameter, multiple sets of training data pairs at each moment can be obtained after the combination. For example, the moment corresponding to 1 second after the collision includes hood structure parameter 1 and corresponding acceleration value 1, hood structure parameter 2 and corresponding acceleration value 2, hood structure parameter n and corresponding acceleration value n, and then the moment corresponding to 2 seconds after the collision also includes the above-mentioned hood structure parameters and corresponding acceleration values.

[0060] Step S40, respectively train the input machine model with multiple sets of training data at each moment to obtain a trained machine model at each moment.

[0061] In this embodiment, multiple sets of training data at each moment are respectively input into the machine model for training. When the number of training times reaches a preset number of times or the prediction accuracy of the machine model obtained through statistics reaches a preset accuracy, the training is stopped to obtain a trained machine model at each moment.

[0062] Step S50, inputting the new hood structure parameters into the trained machine model at each moment respectively to obtain the predicted acceleration value at each moment.

[0063] In this embodiment, new hood structure parameters are obtained by adjusting the hood structure parameters, and the machine model trained at each moment is used for prediction, so that the predicted acceleration value of the new hood structure parameters at each moment can be obtained.

[0064] Step S60: construct a predicted acceleration curve based on the predicted acceleration value at each moment.

[0065] In this embodiment, based on the predicted acceleration value at each moment combined with the moment value extracted in step S20, curve fitting is performed to construct an acceleration curve. In the Python programming language (Python is an interpreted, object-oriented, and high-level programming language with dynamic semantics), the predicted acceleration curve can be output with the help of the plot drawing function.

[0066] In this embodiment, a training data set is first obtained through a simulation test scene or other channels, and the training data set includes multiple groups of training data pairs, each group of training data pairs consists of hood structure parameters and acceleration curves corresponding to the hood structure parameters, different hood structure parameters constitute different hood structure models, and correspond one to one with the acceleration curves generated after a collision with a pedestrian's head. The hood is the front engine cover of a car, and different hood structure models can usually be obtained by adjusting and setting different hood structure parameters, such as changing the height and width of the reinforcement ribs in the hood and the length and width of the hollow shape. The original machine model can only learn to predict a single value. The principle of the present invention is to split the acceleration curve into moment values ​​and acceleration values, and to separate different hood structure parameters and each hood structure parameter at the same moment. The corresponding acceleration values ​​are put together and input into the machine model for training, so as to obtain the trained machine model at each moment, and then the new hood structure parameters are used to predict the predicted acceleration values ​​of the new hood structure parameters at each moment through the trained machine model at each moment, and then the predicted acceleration values ​​at each moment are combined with the extracted moment values, and the predicted acceleration curve is output with the help of the plot drawing function in the Python programming language, thereby realizing the prediction of two-dimensional parameters such as the acceleration curve, solving the problem that the existing machine learning algorithms can only learn to predict a single value, and through machine learning, a large number of high-accuracy acceleration curve predictions can be quickly and easily performed, thereby improving the design efficiency of the hood structure model for protecting pedestrian head collisions.

[0067] Further, in one embodiment, referring to Figure 3 , Figure 3 FIG. 1 is a flow chart of constructing a training data set for an acceleration curve prediction method according to an embodiment of the present invention. Figure 3 As shown, before step S10, it includes:

[0068] Step S01, obtaining multiple groups of hood structural parameters by setting attributes of hood structural parameters and values ​​corresponding to the attributes, wherein each group of hood structural parameters constitutes a hood structural model;

[0069] Step S02, using simulation software, respectively performing simulation collision tests on multiple hood structure models to obtain acceleration curves corresponding to each hood structure model after collision;

[0070] Step S03, forming a training data pair with each set of hood structural parameters and the acceleration curve corresponding to each set of hood structural parameters;

[0071] Step S04: construct multiple sets of training data pairs into a training data set.

[0072] In this embodiment, a collision experiment is carried out on the pedestrian's head and different hood structure models through simulation software to obtain the acceleration curve corresponding to each hood structure model after the collision, wherein each set of hood structure parameters constitutes a hood structure model, and each set of hood structure parameters and the acceleration curve corresponding to each set of hood structure parameters constitute a training data pair, and multiple sets of training data pairs are constructed into a training data set.

[0073] Furthermore, in one embodiment, step S02 includes:

[0074] Through simulation software, multiple hood structure models are subjected to simulated collision tests at multiple preset collision points to obtain the acceleration curve corresponding to each hood structure model at each collision point after the collision.

[0075] In this embodiment, since there may be multiple position points when a pedestrian's head collides with the hood, it is necessary to perform simulated collision tests at multiple preset collision points for each hood structure model to obtain the acceleration curve corresponding to each collision point of each hood structure model after the collision.

[0076] Furthermore, in one embodiment, the acceleration curve prediction method further includes:

[0077] A training data set is constructed for each collision point, and the machine model training is performed to construct the predicted acceleration curve of the new hood parameters at each collision point.

[0078] In this embodiment, for multiple hood structural parameters and for each collision point, training data sets are constructed respectively, and machine model training is performed respectively. Ultimately, the predicted acceleration curve of the new hood parameters at each collision point can be predicted. By analyzing the acceleration curves formed by the new hood structural parameters at multiple collision points, the hood structural parameters can be further optimized and improved.

[0079] Furthermore, in one embodiment, step S40 includes:

[0080] The support vector machine algorithm in the sklearn machine learning library is called through the Python programming language to train multiple sets of training data pairs at each moment.

[0081] In this embodiment, sklearn, also known as scikit-learn, is an open source machine learning toolkit based on the Python programming language, which covers almost all mainstream machine learning algorithms. The support vector machine algorithm in the sklearn machine learning library is called by the Python programming language to train multiple sets of training data pairs at each moment.

[0082] In a third aspect, an embodiment of the present invention further provides an acceleration curve prediction device.

[0083] Reference Figure 4 , Figure 4 Schematic diagram of functional modules of an acceleration curve prediction device according to an embodiment of the present invention.

[0084] In this embodiment, the acceleration curve prediction device includes:

[0085] An acquisition module 10 is used to acquire a training data set, wherein the training data set includes a plurality of training data pairs, each training data pair consisting of a hood structure parameter and an acceleration curve corresponding to the hood structure parameter;

[0086] An extraction module 20, for extracting a plurality of moments and the acceleration value corresponding to each moment from the acceleration curve of each set of training data pairs according to a preset period;

[0087] A combination module 30, for respectively combining the acceleration value at each moment with the corresponding hood structure parameter to obtain multiple sets of training data pairs at each moment;

[0088] A training module 40 is used to train the input machine model using multiple sets of training data at each moment to obtain a trained machine model at each moment;

[0089] A prediction module 50, for inputting new hood structural parameters into the trained machine model at each moment, to obtain a predicted acceleration value at each moment;

[0090] The construction module 60 is used to construct a predicted acceleration curve based on the predicted acceleration value at each moment.

[0091] Further, in one embodiment, referring to Figure 5 , Figure 5 FIG. 1 is a schematic diagram of a training data set construction module of an acceleration curve prediction device according to an embodiment of the present invention. Figure 5 As shown, the acceleration curve prediction device also includes a training data set construction module, including:

[0092] A setting unit 01 is used to obtain multiple groups of hood structure parameters by setting the attributes of the hood structure parameters and the values ​​corresponding to the attributes, wherein each group of hood structure parameters constitutes a hood structure model;

[0093] The test unit 02 is used to perform a simulation collision test on a plurality of hood structure models respectively through simulation software to obtain an acceleration curve corresponding to each hood structure model after the collision;

[0094] A composition unit 03 is used to form a training data pair by combining each set of hood structure parameters and an acceleration curve corresponding to each set of hood structure parameters;

[0095] The construction unit 04 is used to construct a plurality of training data pairs into a training data set.

[0096] Furthermore, in one embodiment, the testing unit is used to:

[0097] Through simulation software, multiple hood structure models are subjected to simulated collision tests at multiple preset collision points to obtain the acceleration curve corresponding to each hood structure model at each collision point after the collision.

[0098] Furthermore, in one embodiment, the acceleration curve prediction device further includes an acceleration curve construction module, which is used to:

[0099] A training data set is constructed for each collision point, and the machine model training is performed to construct the predicted acceleration curve of the new hood parameters at each collision point.

[0100] Furthermore, in one embodiment, the training module 40 is used to:

[0101] The support vector machine algorithm in the sklearn machine learning library is called through the Python programming language to train multiple sets of training data pairs at each moment.

[0102] Among them, the functional implementation of each module in the above-mentioned acceleration curve prediction device corresponds to each step in the above-mentioned acceleration curve prediction method embodiment, and its functions and implementation processes are no longer repeated here.

[0103] In a fourth aspect, an embodiment of the present invention further provides a readable storage medium.

[0104] The readable storage medium of the present invention stores an acceleration curve prediction program, wherein when the acceleration curve prediction program is executed by a processor, the steps of the acceleration curve prediction method described above are implemented.

[0105] The method implemented when the acceleration curve prediction program is executed can refer to the various embodiments of the acceleration curve prediction method of the present invention, and will not be described in detail here.

[0106] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0107] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0108] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device to execute the methods described in each embodiment of the present invention.

[0109] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for predicting an acceleration curve, characterized in that: The acceleration curve prediction method comprises: Acquire a training data set, wherein the training data set includes a plurality of training data pairs, each training data pair consisting of a hood structure parameter and an acceleration curve corresponding to the hood structure parameter; According to a preset period, extract multiple moments and the acceleration value corresponding to each moment from the acceleration curve of each set of training data pairs; The acceleration value at each moment is combined with the corresponding hood structure parameter to obtain multiple sets of training data pairs at each moment; The multiple sets of training data at each moment are respectively used to train the input machine model to obtain the trained machine model at each moment; The new hood structure parameters are input into the trained machine model at each moment to obtain the predicted acceleration value at each moment; Based on the predicted acceleration value at each moment, a predicted acceleration curve is constructed; The training of the machine model by inputting multiple sets of training data at each moment comprises: The support vector machine algorithm in the sklearn machine learning library is called through the Python programming language to train multiple sets of training data pairs at each moment; Before obtaining the training data set, the following steps are included: By setting the attributes of the hood structure parameters and the values ​​corresponding to the attributes, a plurality of groups of hood structure parameters are obtained, wherein each group of hood structure parameters constitutes a hood structure model; Through the simulation software, multiple hood structure models are subjected to simulated collision tests respectively, and the acceleration curve corresponding to each hood structure model after the collision is obtained; The training data pairs are formed by combining each set of hood structural parameters and the acceleration curve corresponding to each set of hood structural parameters; Constructing multiple sets of training data pairs into training data sets; The simulation software is used to perform simulation collision tests on multiple hood structure models respectively, and the acceleration curve corresponding to each hood structure model after the collision is obtained, including: Through simulation software, multiple hood structure models are subjected to simulated collision tests at multiple preset collision points to obtain the acceleration curve corresponding to each hood structure model at each collision point after the collision.

2. The acceleration curve prediction method according to claim 1, characterized in that: The acceleration curve prediction method further includes: A training data set is constructed for each collision point, and the machine model training is performed to construct the predicted acceleration curve of the new hood parameters at each collision point.

3. An acceleration curve prediction device, characterized in that: The acceleration curve prediction device comprises: An acquisition module, used for acquiring a training data set, wherein the training data set includes a plurality of training data pairs, each training data pair consisting of a hood structure parameter and an acceleration curve corresponding to the hood structure parameter; An extraction module, used to extract multiple moments and the acceleration value corresponding to each moment from the acceleration curve of each set of training data pairs according to a preset period; A combination module, used to respectively combine the acceleration value at each moment with the corresponding hood structure parameter to obtain multiple sets of training data pairs at each moment; The training module is used to train the machine model with multiple sets of training data at each moment, so as to obtain a trained machine model at each moment; A prediction module is used to input new hood structure parameters into the trained machine model at each moment to obtain the predicted acceleration value at each moment; A construction module, used for constructing a predicted acceleration curve based on the predicted acceleration value at each moment; The training module is used to: The support vector machine algorithm in the sklearn machine learning library is called through the Python programming language to train multiple sets of training data pairs at each moment; The acceleration curve prediction device also includes a training data set construction module, including: A setting unit, used for obtaining a plurality of groups of hood structure parameters by setting attributes of the hood structure parameters and values ​​corresponding to the attributes, wherein each group of hood structure parameters constitutes a hood structure model; A test unit, used to perform simulated collision tests on multiple hood structure models respectively through simulation software to obtain an acceleration curve corresponding to each hood structure model after the collision; A composition unit, used for composing each set of hood structural parameters and an acceleration curve corresponding to each set of hood structural parameters into a training data pair; A construction unit, used for constructing a plurality of sets of training data pairs into a training data set; Test unit for: Through simulation software, multiple hood structure models are subjected to simulated collision tests at multiple preset collision points to obtain the acceleration curve corresponding to each hood structure model at each collision point after the collision.

4. An acceleration curve prediction device, characterized in that: The acceleration curve prediction device includes a processor, a memory, and an acceleration curve prediction program stored in the memory and executable by the processor, wherein when the acceleration curve prediction program is executed by the processor, the steps of the acceleration curve prediction method according to any one of claims 1 to 2 are implemented.

5. A readable storage medium, characterized in that: The readable storage medium stores an acceleration curve prediction program, wherein when the acceleration curve prediction program is executed by a processor, the steps of the acceleration curve prediction method according to any one of claims 1 to 2 are implemented.

Images