High overload response nonlinear inversion test method and device

Abstract

The invention discloses a high overload response nonlinear inversion testing method and device, and relates to the field of high overload response nonlinear inversion testing, and the testing method comprises the steps: building a nonlinear autoregressive network model through data training based on a high-precision numerical simulation technology and equivalent overload experiment data through employing the idea of leveraging in all directions; the mapping relation is used for representing front and back overload response of the buffer device; the testing device comprises a buffer device designed based on polyurethane foam and honeycomb aluminum, a small-range acceleration sensor, electronic equipment used for data processing and storage and a storage medium. According to the invention, the technical bottleneck of high-overload response test is broken through, the dependence on a foreign ultra-high-range acceleration sensor is eliminated, the effective and accurate test of high-overload response is realized, and the test cost is greatly reduced.

Description

Technical Field

[0001] The present application relates to the technical field of high overload response nonlinear inversion testing, and in particular to a high overload response nonlinear inversion testing method and device. Background Art

[0002] During the firing of ammunition, a great impact force is generated. In order to obtain the impact excitation curve, a large-range accelerometer needs to be installed to measure the impact force on the projectile. At the same time, measures need to be taken to reduce the adverse effects of this extreme working environment on the accuracy of the large-range accelerometer. Therefore, the cost of using a large-range accelerometer is relatively high. Summary of the invention

[0003] In view of this, the purpose of this application is to provide a high overload response nonlinear inversion test method and device to solve the above-mentioned problem. During the ammunition launching process, a small-range acceleration sensor is used to measure the impact force exerted on the projectile body, thereby reducing the use cost.

[0004] In a first aspect, an embodiment of the present application provides a high overload response nonlinear inversion test method, wherein a buffer device is provided in a test body, and a first acceleration sensor is provided in the buffer device; the method comprises: The processor obtains a first acceleration detected by the first acceleration sensor; wherein the first acceleration is used to characterize the acceleration detected by the first acceleration sensor after buffering and reducing the excitation load on the test body, and the range of the first acceleration sensor is less than or equal to a first preset range; The processor inputs the first acceleration into a trained nonlinear autoregressive network model so that the nonlinear autoregressive network model outputs a second acceleration; wherein the second acceleration is used to characterize the acceleration to which the test body is subjected; The nonlinear autoregressive network model is obtained according to the following method: Acquire an acceleration sample set; wherein the acceleration sample set includes a plurality of first acceleration sample values ​​and a plurality of second acceleration sample values; The acceleration sample set is input into the original nonlinear autoregressive network model for training to obtain the nonlinear autoregressive network model.

[0005] Preferably, the method further comprises: Acquire an original acceleration sample set; wherein the original acceleration sample set is a set consisting of a plurality of first acceleration sample values ​​with noise and a plurality of second acceleration sample values ​​with noise; The original acceleration sample set is input to a Butterworth low-pass filter, so that the Butterworth low-pass filter outputs the acceleration sample set.

[0006] Preferably, the buffer device is provided with a buffer material; wherein the buffer material comprises polyurethane foam and honeycomb aluminum.

[0007] Preferably, the method further comprises: The processor calculates the impact force exerted on the test body according to the second acceleration.

[0008] Preferably, the buffer device is further provided with a second acceleration sensor; and the method further comprises: The processor acquires a third acceleration detected by the second acceleration sensor; wherein the third acceleration is used to characterize the acceleration detected by the second acceleration sensor to which the test body is subjected, and the range of the second acceleration sensor is greater than the first preset range and less than or equal to the second preset range; The processor calculates and obtains a square residual coefficient of the second acceleration and the third acceleration; The processor calculates a root mean square error between the second acceleration and the third acceleration; If the square residual coefficient is greater than the square residual coefficient threshold and the root mean square error is less than the root mean square error threshold, the processor indicates that there is no need to retrain the nonlinear autoregressive network model.

[0009] Preferably, the method further comprises: If the square residual coefficient is less than or equal to the square residual coefficient threshold or the root mean square error is greater than or equal to the root mean square error threshold, the processor prompts to retrain the nonlinear autoregressive network model.

[0010] Preferably, the step of the processor calculating the square residual coefficient of the second acceleration and the third acceleration comprises: The calculation formula of the square residual coefficient of the second acceleration and the third acceleration is:

[0011] in, is the square residual coefficient of the second acceleration and the third acceleration, is the ith third acceleration, is the ith second acceleration, is the average value of the third acceleration, I is the number of the second acceleration or the third acceleration; The step of calculating, by the processor, a root mean square error between the second acceleration and the third acceleration comprises: The calculation formula of the root mean square error between the second acceleration and the third acceleration is:

[0012] in, is the root mean square error between the second acceleration and the third acceleration.

[0013] The high overload response nonlinear inversion test method provided in this application brings the following beneficial effects: The present application provides a nonlinear inversion test method for high overload response, in which the processor inputs a first acceleration into a trained nonlinear autoregressive network model so that the nonlinear autoregressive network model outputs a second acceleration. Wherein, the first acceleration is the acceleration detected by the first acceleration sensor after the excitation load on the test body is buffered and unloaded, and the second acceleration is used to characterize the acceleration on the test body. The method adopts the concept of "four ounces to move a thousand pounds", based on high-precision numerical simulation technology and equivalent overload experimental data, and establishes a nonlinear autoregressive network model through data training, which is used to characterize the mapping relationship between the overload response before and after the buffer device; wherein, the test device includes a buffer device designed based on polyurethane foam and honeycomb aluminum, a small-range acceleration sensor, and an electronic device and storage medium for data processing and storage. This method breaks through the technical bottleneck of high overload response testing, gets rid of the dependence on foreign ultra-high-range acceleration sensors, realizes effective and accurate testing of high overload response, and greatly reduces the test cost. In this method, the detection results of a small-range sensor (a sensor with a range less than or equal to the first preset range) are as accurate as the detection results of a large-range acceleration sensor (a sensor with a range greater than the first preset range). This provides a very important reference for load measurement in high overload extreme environments and the design of load reduction measures.

[0014] In a second aspect, the present application further provides a high overload response nonlinear inversion test device, wherein a buffer device is provided in a test body, and a first acceleration sensor is provided in the buffer device; the high overload response nonlinear inversion test device comprises: an acquisition module and a data processing module; The acquisition module is used to acquire a first acceleration detected by the first acceleration sensor; wherein the first acceleration is used to characterize the acceleration detected by the first acceleration sensor after buffering and reducing the excitation load received by the test body, and the range of the first acceleration sensor is less than or equal to a first preset range; The data processing module is used to input the first acceleration into a trained nonlinear autoregressive network model so that the nonlinear autoregressive network model outputs a second acceleration; wherein the second acceleration is used to characterize the acceleration to which the test body is subjected.

[0015] The high overload response nonlinear inversion test device provided in the embodiment of the present application has the same technical features as the high overload response nonlinear inversion test method provided in the above embodiment, so it can also solve the same technical problems and achieve the same technical effects.

[0016] In a third aspect, the present application also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the beneficial effects of any of the methods described above are achieved, which will not be repeated here.

[0017] In a fourth aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the beneficial effects of any of the methods described above are achieved, which will not be repeated here.

[0018] Other features and advantages of the present application will be described in the following description, and partly become apparent from the description, or understood by practicing the present application. The purpose and other advantages of the present application are realized and obtained by the structures particularly pointed out in the description and the drawings.

[0019] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0021] Figure 1 A schematic diagram of a high overload response nonlinear inversion test method provided in an embodiment of the present application; Figure 2 A schematic diagram of the test body structure provided in the embodiment of the present application; Figure 3 A schematic diagram of the structure of a drop device provided in an embodiment of the present application; Figure 4 A schematic flow chart of another high overload response nonlinear inversion test method provided in an embodiment of the present application; Figure 5 A schematic diagram of the structure of a high overload response nonlinear inversion test device provided in an embodiment of the present application; Figure 6A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution of the present application will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0023] To facilitate understanding of this embodiment, the embodiments of the present application are described in detail below.

[0024] The embodiment of the present application provides a high overload response nonlinear inversion test method, wherein a buffer device is provided in the test body, and a first acceleration sensor is provided in the buffer device; Figure 1 As shown, Figure 1 A flow chart of a high overload response nonlinear inversion test method provided in an embodiment of the present application. The method comprises the following steps: S101: The processor obtains a first acceleration detected by a first acceleration sensor.

[0025] The first acceleration is used to characterize the acceleration detected by the first acceleration sensor after buffering and reducing the excitation load on the test body, and the range of the first acceleration sensor is less than or equal to the first preset range.

[0026] Specifically, Figure 2 As shown, Figure 2 A schematic diagram of the structure of the test body 200 provided in the embodiment of the present application. The test body 200 is used to simulate a real artillery firing scenario in a high-efficiency wide-pulse strong shock environment simulation test. The test body 200 is coaxially provided with a buffer device 201 and a first acceleration sensor mounting base 202; wherein, the buffer device 201 is used to weaken the impact force exerted on the first acceleration sensor 203, the first acceleration sensor 203 is connected to the upper surface of the first acceleration sensor mounting base 202 by M5*25 bolts, and a first acceleration sensor protection assembly 205 is provided on the upper surface of the first acceleration sensor 203. The first acceleration sensor 203 is connected to the filter through a wire, and the wire is connected to the oscilloscope through the opening of the outer shell of the test body 200; the test body 200 is a cylinder, and the protective cover 204 of the test body 200 has a diameter of 70mm and a thickness of 5mm. The volume of the first acceleration sensor 203 is 125mm 3 , the distance from the opening of the shell of the test body 200 is 15 mm, and the width of the opening of the shell of the test body 200 is 5 mm.

[0027] In addition, the above-mentioned impact force is generated by the drop device 300. Figure 3 As shown, Figure 3 The schematic diagram of the structure of the drop device 300 provided in the embodiment of the present application. The drop device 300 includes: a drop body 301, a guide rod 302, a drive rod 303 and a base 304; during the test process, the drive rod 303 controls the drop body 301 to move downward along the guide rod 302, and the above-mentioned test body 200 is arranged on the upper surface of the drop body 301. When the lower surface of the drop body 301 moves to the upper surface of the base 304 and collides with the upper surface of the base 304, the lower surface of the drop body 301 is subjected to the reaction force (impact force) generated by the upper surface of the base 304 due to the collision. The drop body 301 transmits the impact force to the lower surface of the shell of the test body 200, and the impact force received by the lower surface of the shell of the test body 200 is transmitted to the first acceleration sensor mounting base 202 and the buffer device 201 in turn, and the buffer device 201 undergoes plastic deformation, thereby reducing the impact force received by the first acceleration sensor mounting base 202. When the impact force on the first acceleration sensor mounting base 202 is transmitted to the oscilloscope, the operator can obtain the amplitude and pulse width of the impact force on the first acceleration sensor mounting base 202 by operating the oscilloscope, and filter and analyze the amplitude and pulse width of the impact force on the first acceleration sensor mounting base 202 to obtain the first acceleration. After each test, the buffer device 201 in the test body 200 is replaced. It can be seen that the first acceleration is the acceleration of the test body 200 detected by the acceleration sensor after the buffer device 201 weakens the impact force.

[0028] S102: The processor inputs the first acceleration into a trained nonlinear autoregressive network model so that the nonlinear autoregressive network model outputs a second acceleration.

[0029] The second acceleration is used to characterize the acceleration to which the test body 200 is subjected.

[0030] Specifically, the nonlinear autoregressive network model is a nonlinear autoregressive model for time series modeling, which can handle the time series prediction problems of nonlinearity and feedback mechanism. Since the first acceleration sensor 203 is a small-range sensor (a sensor with a range less than or equal to the first preset range), the first acceleration detected by the first acceleration sensor 203 is not the acceleration generated by the impact force actually received by the test body 200. At this time, the input first acceleration needs to be processed by the pre-trained nonlinear autoregressive network model to predict the second acceleration.

[0031] With such a setting, a small-range sensor (the first acceleration sensor 203) can be set only on the test body 200 to detect the acceleration (the first acceleration) generated after the impact force received by the test body 200 is weakened by the buffer device 201, and the first acceleration is input into a pre-trained non-linear autoregressive network model to predict and obtain the second acceleration (the acceleration generated by the impact force actually received by the test body 200), reducing the detection cost.

[0032] More specifically, the non-linear autoregressive network model is obtained in the following manner: Obtain an acceleration sample set; wherein, the acceleration sample set includes a plurality of first acceleration sample values and a plurality of second acceleration sample values; input the acceleration sample set into the original non-linear autoregressive network model for training to obtain the non-linear autoregressive network model.

[0033] The process of obtaining the acceleration sample set described above is as follows: (1) Set the range of the pulse width t corresponding to the second acceleration sample value: 0.1ms < t < 1ms, and set the range of the acceleration overload peak value a corresponding to the second acceleration sample value: 8000gm / s 2 < a < 60000gm / s 2 , and set the number of samples to 30.

[0034] Table 1 is a partial sample set table of the acceleration sample set: Table 1:

[0035] (2) Refer to parameters such as the pulse width and amplitude of different impact overloads generated in the high-energy efficiency wide-pulse strong impact environment simulation test to set boundary conditions, and obtain the second acceleration sample value through the Latin hypercube sampling method.

[0036] (3) Use the finite element simulation method to simulate the second acceleration sample value to obtain the first acceleration sample value.

[0037] The process of training to obtain the non-linear autoregressive network model is as follows: (1) Divide the acceleration sample set into a training set (the first acceleration sample value) and a validation set (the second acceleration sample value) and import them into 2 cell arrays containing 30 matrices to obtain a cell sample array.

[0038] (2) Create a feedforward neural network (the original non-linear autoregressive network model) through the feedforwardnet function, set the number of hidden neurons to 50, set the maximum number of iterations to 1000, and the learning rate to 0.0006.

[0039] (3) Based on the train function, the cell sample array is input into the feedforward neural network (the original nonlinear autoregressive network model) to obtain the nonlinear autoregressive network model.

[0040] The present application provides a nonlinear inversion test method for high overload response, in which a processor inputs a first acceleration into a trained nonlinear autoregressive network model so that the nonlinear autoregressive network model outputs a second acceleration. Wherein, the first acceleration is the acceleration after buffering detected by the first acceleration sensor, and the second acceleration is used to characterize the acceleration to which the test body is subjected. Compared with the prior art, which uses a large-range acceleration sensor to measure the overload to the projectile, this method adopts the concept of "four ounces to move a thousand pounds", based on high-precision numerical simulation technology and equivalent overload experimental data, and establishes a nonlinear autoregressive network model through data training, which is used to characterize the mapping relationship between the overload response before and after the buffer device; wherein, the test device includes a buffer device designed based on polyurethane foam and honeycomb aluminum, a small-range acceleration sensor, and an electronic device and storage medium for data processing and storage. This method breaks through the technical bottleneck of high overload response testing, gets rid of the dependence on foreign ultra-high-range acceleration sensors, realizes effective and accurate testing of high overload response, and greatly reduces the testing cost. This method uses a small-range sensor (a sensor with a range less than or equal to the first preset range) to replace a large-range acceleration sensor (a sensor with a range greater than the first preset range) to detect the magnitude of the acceleration generated by the impact force on the test body (projectile). No measures need to be taken to reduce the adverse effects of the extreme working environment (huge impact force) on the test results, which greatly reduces the cost of use. The test results of the small-range sensor (a sensor with a range less than or equal to the first preset range) are as accurate as those of the large-range acceleration sensor (a sensor with a range greater than the first preset range). This method provides a very important reference for load measurement in high overload extreme environments and the design of load reduction measures.

[0041] In one embodiment, the processor obtains an original acceleration sample set, wherein the original acceleration sample set is a set consisting of a plurality of first acceleration sample values ​​with noise and a plurality of second acceleration sample values ​​with noise; and the original acceleration sample set is input into a Butterworth low-pass filter so that the Butterworth low-pass filter outputs an acceleration sample set.

[0042] Specifically, the Butterworth low-pass filter is an analog filter with many advantages such as high precision, stability, flexibility, and no impedance matching. The acceleration sample set obtained above is obtained by filtering the original acceleration sample set. Before the filtering process starts, it is necessary to set the sampling interval and sampling frequency and normalize the cutoff frequency, and set the order of the Butterworth low-pass filter to 5th order. After the setting is completed, the original acceleration sample set is input into the Butterworth low-pass filter to obtain the acceleration sample set.

[0043] This method can filter the noise in the original acceleration set, and the data of the obtained acceleration sample set is noise-free, thereby making the nonlinear autoregressive network model trained by the acceleration sample set have stronger reliability.

[0044] In one embodiment, the buffer device is provided with a buffer material; wherein the buffer material comprises polyurethane foam and honeycomb aluminum.

[0045] Specifically, polyurethane foam and honeycomb aluminum have high damping performance and can absorb a large amount of energy generated by impact. The buffer device weakens the impact force received by the load device through the buffer material, so that the first acceleration sensor detects an acceleration (first acceleration) that meets the range requirement of the first acceleration sensor.

[0046] In this manner, due to the characteristics of polyurethane foam and honeycomb aluminum, the buffer device is able to weaken the impact force received by the load device, thereby ensuring that the first acceleration detected by the first acceleration sensor does not exceed the measuring range of the first acceleration sensor, thereby improving the reliability of the first acceleration detected by the first acceleration sensor.

[0047] In one implementation, the processor calculates the impact force on the test body according to the second acceleration.

[0048] Specifically, since the second acceleration is predicted by a nonlinear autoregressive network model and has a high accuracy, the impact force on the test body calculated by the second acceleration also has a high accuracy.

[0049] In this method, the impact force of the test body obtained by calculating the second acceleration has a high accuracy. The staff can adjust the impact force of the next test body (projectile) when launching according to the impact force, thereby improving the safety of the next test body (projectile) when launching.

[0050] In one embodiment, a second acceleration sensor is further provided in the buffer device 201, and the processor obtains a third acceleration detected by the second acceleration sensor; wherein the third acceleration is used to characterize the acceleration detected by the second acceleration sensor to which the test body is subjected, and the range of the second acceleration sensor is greater than the first preset range and less than or equal to the second preset range; the square residual coefficients of the second acceleration and the third acceleration are calculated; the root mean square error of the second acceleration and the third acceleration is calculated; if the square residual coefficient is greater than the square residual coefficient threshold and the root mean square error is less than the root mean square error threshold, it is prompted that the nonlinear autoregressive network model does not need to be retrained; if the square residual coefficient is less than or equal to the square residual coefficient threshold or the root mean square error is greater than or equal to the root mean square error threshold, it is prompted to retrain the nonlinear autoregressive network model.

[0051] Specifically, the calculation formula of the square residual coefficient of the second acceleration and the third acceleration is:

[0052] in, is the square residual coefficient of the second acceleration and the third acceleration, is the ith third acceleration, is the ith second acceleration, is the average value of the third acceleration, I is the number of the second acceleration or the third acceleration; The calculation formula of the root mean square error of the second acceleration and the third acceleration is:

[0053] in, is the root mean square error between the second acceleration and the third acceleration, is the ith third acceleration, is the i-th second acceleration, and I is the number of second accelerations or third accelerations.

[0054] The square residual coefficient and the root mean square error of the second acceleration and the third acceleration are used to characterize the fit between the first acceleration and the second acceleration, that is, the degree of approximation. The closer the square residual coefficient is to 1, the closer the second acceleration is to the third acceleration, and the higher the accuracy of the nonlinear autoregressive network model. The smaller the root mean square error is, the closer the second acceleration is to the third acceleration, and the higher the accuracy of the nonlinear autoregressive network model. The range of the square residual coefficient of the second acceleration and the third acceleration is 0.75≤ ≤0.9, the range of the root mean square error between the second acceleration and the third acceleration is 45≤ ≤50.

[0055] In this method, the square residual coefficient and the root mean square error of the second acceleration and the third acceleration are used to judge the degree of fit of the second acceleration and the third acceleration, and then the accuracy of the nonlinear autoregressive network model is judged. This method enables the staff to accurately judge whether the currently used nonlinear autoregressive network model needs to be retrained.

[0056] Based on the above method embodiment, the present application embodiment provides another high overload response nonlinear inversion test method. Figure 4 As shown, Figure 4 A flow chart of another high overload response nonlinear inversion test method provided in an embodiment of the present application. The method comprises the following steps: S401, obtaining a second acceleration sample value by using a Latin hypercube sampling method.

[0057] For details, please refer to the aforementioned embodiments, and the embodiments of the present invention will not be described in detail here.

[0058] S402: Simulate the second acceleration sample value by using a finite element simulation method to obtain a first acceleration sample value.

[0059] For details, please refer to the aforementioned embodiments, and the embodiments of the present invention will not be described in detail here.

[0060] S403: Input the first acceleration sample value and the second acceleration sample value into the original nonlinear autoregressive network model for training to obtain a nonlinear autoregressive network model.

[0061] For details, please refer to the aforementioned embodiments, and the embodiments of the present invention will not be described in detail here.

[0062] S404: Acquire a first acceleration detected by a first acceleration sensor.

[0063] The first acceleration is used to characterize the acceleration detected by the first acceleration sensor after buffering and reducing the excitation load on the test body, and the range of the first acceleration sensor is less than or equal to the first preset range.

[0064] For details, please refer to the aforementioned embodiments, and the embodiments of the present invention will not be described in detail here.

[0065] S405: Input the first acceleration to the trained nonlinear autoregressive network model, so that the nonlinear autoregressive network model outputs a second acceleration.

[0066] The second acceleration is used to characterize the acceleration to which the test body is subjected.

[0067] For details, please refer to the aforementioned embodiments, and the embodiments of the present invention will not be described in detail here.

[0068] The above step S404 can be performed simultaneously with step S401, step S402 and step S403.

[0069] Based on the above method embodiment, the present application embodiment further provides a high overload response nonlinear inversion test device. A buffer device is provided in the test body, and a first acceleration sensor is provided in the buffer device; Figure 5 As shown, Figure 5 A schematic diagram of the structure of a high overload response nonlinear inversion test device provided in an embodiment of the present application. The device includes an acquisition module 51 and a data processing module 52; wherein the functions of each module are as follows: An acquisition module 51 is used to acquire a first acceleration detected by a first acceleration sensor; wherein the first acceleration is used to characterize the acceleration detected by the first acceleration sensor after buffering and reducing the excitation load on the test body, and the range of the first acceleration sensor is less than or equal to the first preset range; The data processing module 52 is used to input the first acceleration into the trained nonlinear autoregressive network model so that the nonlinear autoregressive network model outputs a second acceleration; wherein the second acceleration is used to characterize the acceleration to which the test body is subjected; The nonlinear autoregressive network model is obtained as follows: Acquire an acceleration sample set; wherein the acceleration sample set includes a plurality of first acceleration sample values ​​and a plurality of second acceleration sample values; The acceleration sample set is input into the original nonlinear autoregressive network model for training to obtain a nonlinear autoregressive network model.

[0070] Preferably, the acquisition module 51 is also used to acquire an original acceleration sample set; wherein the original acceleration sample set is a set consisting of a plurality of first acceleration sample values ​​with noise and a plurality of second acceleration sample values ​​with noise; the original acceleration sample set is input into a Butterworth low-pass filter so that the Butterworth low-pass filter outputs an acceleration sample set.

[0071] Preferably, the buffer device is provided with a buffer material; wherein the buffer material comprises polyurethane foam and honeycomb aluminum.

[0072] Preferably, the data processing module 52 is further used to calculate the impact force exerted on the test body according to the second acceleration.

[0073] Preferably, a second acceleration sensor is also provided in the buffer device; the acquisition module 51 is further used to acquire a third acceleration detected by the second acceleration sensor; wherein the third acceleration is used to characterize the acceleration of the test body detected by the second acceleration sensor, and the range of the second acceleration sensor is greater than the first preset range and less than or equal to the second preset range; The data processing module 52 is also used to calculate the square residual coefficient of the second acceleration and the third acceleration; calculate the root mean square error of the second acceleration and the third acceleration; if the square residual coefficient is greater than the square residual coefficient threshold and the root mean square error is less than the root mean square error threshold, it is prompted that there is no need to retrain the nonlinear autoregressive network model.

[0074] Preferably, the data processing module 52 is further configured to prompt the nonlinear autoregressive network model to be retrained if the square residual coefficient is less than or equal to the square residual coefficient threshold or the root mean square error is greater than or equal to the root mean square error threshold.

[0075] Preferably, the data processing module 52 is specifically configured to obtain the square residual coefficient of the second acceleration and the third acceleration according to the following formula:

[0076] in, is the square residual coefficient of the second acceleration and the third acceleration, is the ith third acceleration, is the ith second acceleration, is the average value of the third acceleration, I is the number of the second acceleration or the third acceleration; The root mean square error between the second acceleration and the third acceleration is obtained according to the following formula:

[0077] in, is the root mean square error between the second acceleration and the third acceleration, is the ith third acceleration, is the i-th second acceleration, and I is the number of second accelerations or third accelerations.

[0078] The high overload response nonlinear inversion test device provided in the embodiment of the present application has the same technical features as the high overload response nonlinear inversion test method provided in the above embodiment, so it can also solve the same technical problems and achieve the same technical effects.

[0079] An embodiment of the present application also provides an electronic device, including a processor and a memory, wherein the memory stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement the above-mentioned high overload response nonlinear inversion test method.

[0080] See also Figure 6 As shown, the electronic device includes a processor 100 and a memory 101 , wherein the memory 101 stores machine executable instructions that can be executed by the processor 100 , and the processor 100 executes the machine executable instructions to implement the above-mentioned high overload response nonlinear inversion test method.

[0081] Further, Figure 6 The electronic device shown further includes a bus 102 and a communication interface 103 , and the processor 100 , the communication interface 103 and the memory 101 are connected via the bus 102 .

[0082] Among them, the memory 101 may include a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 103 (which can be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. can be used. The bus 102 can be an ISA (Industrial Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Enhanced Industry Standard Architecture) bus, etc. The above-mentioned bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 6 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0083] The processor 100 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 100. The above processor 100 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module may be located in a storage medium mature in the art, such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 101, and the processor 100 reads the information in the memory 101 and completes the steps of the method of the above embodiment in combination with its hardware.

[0084] This embodiment also provides a machine-readable storage medium, which stores machine-executable instructions. When the machine-executable instructions are called and executed by a processor, the machine-executable instructions prompt the processor to implement the above-mentioned high overload response nonlinear inversion test method.

[0085] The computer program product of the high overload response nonlinear inversion test method, device and electronic device provided in the embodiments of the present application includes a computer-readable storage medium storing program code, and the instructions included in the program code can be used to execute the method described in the previous method embodiment. The specific implementation can refer to the method embodiment, which will not be repeated here. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the system and device described above can refer to the corresponding process in the previous method embodiment, which will not be repeated here.

[0086] In addition, in the description of the embodiments of the present application, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, or it can be the internal connection of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances. If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a non-volatile computer-readable storage medium that can be executed by a processor. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0087] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present application. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0088] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present application, which are used to illustrate the technical solutions of the present application, rather than to limit them. The protection scope of the present application is not limited thereto. Although the present application is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed in the present application, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A high overload response nonlinear inversion test method, characterized in that: A buffer device is provided in the test body, and a first acceleration sensor is provided in the buffer device; the method comprises: The processor obtains a first acceleration detected by the first acceleration sensor; wherein the first acceleration is used to characterize the acceleration detected by the first acceleration sensor after buffering and reducing the excitation load on the test body, and the range of the first acceleration sensor is less than or equal to a first preset range; The processor inputs the first acceleration into a trained nonlinear autoregressive network model so that the nonlinear autoregressive network model outputs a second acceleration; wherein the second acceleration is used to characterize the acceleration to which the test body is subjected; The nonlinear autoregressive network model is obtained according to the following method: Acquire an acceleration sample set; wherein the acceleration sample set includes a plurality of first acceleration sample values ​​and a plurality of second acceleration sample values; The acceleration sample set is input into the original nonlinear autoregressive network model for training to obtain the nonlinear autoregressive network model.

2. The high overload response nonlinear inversion test method according to claim 1, characterized in that: The method further comprises: Acquire an original acceleration sample set; wherein the original acceleration sample set is a set consisting of a plurality of first acceleration sample values ​​with noise and a plurality of second acceleration sample values ​​with noise; The original acceleration sample set is input to a Butterworth low-pass filter, so that the Butterworth low-pass filter outputs the acceleration sample set.

3. The high overload response nonlinear inversion test method according to claim 1, characterized in that: The buffer device is provided with a buffer material; wherein the buffer material comprises polyurethane foam and honeycomb aluminum.

4. The high overload response nonlinear inversion test method according to claim 1, characterized in that: The method further comprises: The processor calculates the impact force exerted on the test body according to the second acceleration.

5. The high overload response nonlinear inversion test method according to claim 1, characterized in that: The buffer device is also provided with a second acceleration sensor; the method further comprises: The processor acquires a third acceleration detected by the second acceleration sensor; wherein the third acceleration is used to characterize the acceleration detected by the second acceleration sensor to which the test body is subjected, and the range of the second acceleration sensor is greater than the first preset range and less than or equal to the second preset range; The processor calculates and obtains a square residual coefficient of the second acceleration and the third acceleration; The processor calculates a root mean square error between the second acceleration and the third acceleration; If the square residual coefficient is greater than the square residual coefficient threshold and the root mean square error is less than the root mean square error threshold, the processor indicates that there is no need to retrain the nonlinear autoregressive network model.

6. The high overload response nonlinear inversion test method according to claim 5, characterized in that: The method further comprises: If the square residual coefficient is less than or equal to the square residual coefficient threshold or the root mean square error is greater than or equal to the root mean square error threshold, the processor prompts to retrain the nonlinear autoregressive network model.

7. The high overload response nonlinear inversion test method according to claim 5, characterized in that: The step of calculating, by the processor, the square residual coefficient of the second acceleration and the third acceleration comprises: The calculation formula of the square residual coefficient of the second acceleration and the third acceleration is: in, is the square residual coefficient of the second acceleration and the third acceleration, is the ith third acceleration, is the ith second acceleration, is the average value of the third acceleration, I is the number of the second acceleration or the third acceleration; The step of calculating, by the processor, a root mean square error between the second acceleration and the third acceleration comprises: The calculation formula of the root mean square error between the second acceleration and the third acceleration is: in, is the root mean square error between the second acceleration and the third acceleration.

8. A high overload response nonlinear inversion test device, characterized in that: A buffer device is provided in the test body, and a first acceleration sensor is provided in the buffer device; the high overload response nonlinear inversion test device comprises: an acquisition module and a data processing module; The acquisition module is used to acquire a first acceleration detected by the first acceleration sensor; wherein the first acceleration is used to characterize the acceleration detected by the first acceleration sensor after buffering and reducing the excitation load received by the test body, and the range of the first acceleration sensor is less than or equal to a first preset range; The data processing module is used to input the first acceleration into a trained nonlinear autoregressive network model so that the nonlinear autoregressive network model outputs a second acceleration; wherein the second acceleration is used to characterize the acceleration to which the test body is subjected.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is executed.

Images