Multi-modal information fusion shipborne engine foundation impact simulation test system and method

Abstract

The present invention relates to the field of impact testing technology, and specifically to a multi-modal information fusion shipborne engine basic impact simulation test system, including a mechanical structure module, a multi-modal measurement system, a storage module, a control module, a data transmission module, a post-processing system, a visualization module, a judgment module, and an accessory system. The mechanical structure module includes a simulated nacelle device, an automatic lifting device, a test piece to be tested, a mounting frame, a mounting section, a landing platform, and a platform vibrator. The post-processing system includes a general method module, a machine learning module, a deep learning module, and a data dimension reduction module. The present invention also discloses a multi-modal information fusion shipborne engine basic impact simulation test method, which can realize the state information collection and analysis of multiple target analysis structures under different shipborne engines and different impact working conditions simulation tests, and can provide analysis means and data sources for the performance evaluation of shipborne engines and related structures under basic impact.

Description

Technical Field

[0001] The present invention relates to the technical field of impact testing, and specifically provides a multi-modal information fusion shipborne engine basic impact simulation test system and method. Background Art

[0002] With the rapid development of my country's aircraft carrier technology, aircraft landing technology has become one of the keys to whether carrier-based aircraft can serve in aircraft carrier formations and successfully perform early warning, transportation and combat tasks. In order to ensure that carrier-based aircraft can take off in time after landing failure, high-speed landing is usually adopted. The strong impact generated by high-speed landing is coupled with the vibration caused by the wave flow of ships such as aircraft carriers, and is transmitted to the aircraft engine through the landing gear, fuselage and other structures of the carrier-based aircraft, resulting in a vertical impact 3-4 times that of the landing of land-based aircraft, which seriously endangers the structural health of the aircraft engine and the flight safety of the carrier-based aircraft.

[0003] In order to ensure the performance and safety of aircraft engines, the engines of new aircraft models must undergo impact tests before they are put into service to evaluate the vibration isolation and impact resistance of the engines and their supporting structures. At present, my country's carrier-based aircraft technology is in a rapid development stage, and there are many test tasks for new aircraft models and their engines. The existing aircraft engine impact simulation system has some limitations, such as the relatively single type of aircraft engines, the difficulty in simulating the vibration of the aircraft carrier during the landing process of carrier-based aircraft, and the single data collection of the supporting measurement system. Therefore, it is difficult to achieve a system that can be used for multiple aircraft engines, multiple working conditions, and multiple test tasks. Measurement and analysis.

[0004] In summary, it is necessary to propose a basic shock simulation test system for shipborne engines with multi-modal information fusion to meet the shock test requirements of different shipborne engines under multiple working conditions and multiple analysis tasks. Summary of the invention

[0005] In order to solve the above problems, the present invention provides a multi-modal information fusion shipborne engine basic impact simulation test system and method, which can meet the impact test requirements of various types of aircraft engines landing under different working conditions, and can ensure the performance and safety of various types of aircraft engines.

[0006] The multi-modal information fusion shipborne engine basic impact simulation test system provided by the present invention specifically includes:

[0007] Mechanical structure module, used to realize the test task;

[0008] Multimodal measurement system for data acquisition and preprocessing;

[0009] A storage module, used for storing measurement data collected by the multimodal measurement system;

[0010] A control module, used to control the mechanical structure module to achieve movement;

[0011] A data transmission module, used for transmitting data in the storage module;

[0012] A post-processing system, used to realize the fusion of multiple modal data information transmitted by the data transmission module;

[0013] Visualization module, used to visualize the data and models fused by the post-processing system;

[0014] A judgment module is used to judge the operation strategy;

[0015] Accessory system for running the test piece to be tested.

[0016] Furthermore, the mechanical structure module includes:

[0017] A simulated nacelle device is used to simulate the nacelle of an aircraft where an aircraft engine is located;

[0018] An automatic lifting device, located above the simulated nacelle device, used for sucking, lifting or releasing the simulated nacelle device;

[0019] The test piece to be tested is located in the simulated nacelle device and is the aircraft engine used for testing or an equivalent with the test function;

[0020] A mounting frame, located in the simulated nacelle device and connected to the simulated nacelle device, and used for mounting and fixing the test piece to be tested;

[0021] The installation section is located in the simulated nacelle device and is used to connect the installation frame and the test piece to be tested;

[0022] A landing platform, located below the simulated nacelle device, and used for carrying the simulated nacelle device for landing;

[0023] The platform exciter is located under the landing platform. When simulating the basic impact load of landing, it is in a static state. When simulating the basic impact load with time-varying characteristic load, it enters the running state.

[0024] Furthermore, the multimodal measurement system comprises:

[0025] The measurement hardware system is used to realize the data acquisition of the multi-modal measurement system, including: an acceleration sensor arranged on the mounting frame, a strain sensor arranged on the mounting node, a machine vision measurement unit arranged on the landing platform, a laser rangefinder 1 arranged below the mounting frame and located on the base of the simulated nacelle device, a bipolar plate capacitance sensor arranged at the test piece to be tested, and a laser rangefinder 2 arranged at the bottom of the simulated nacelle device;

[0026] Positioning mode system, used in conjunction with laser tester 2, to measure the distance between the simulated nacelle device and the landing platform, and assist the control module in making motion decisions for the mechanical structure module;

[0027] The measurement and testing system is used to complete multimodal data measurement tasks.

[0028] Furthermore, the post-processing system includes:

[0029] A general method module for performing general method analysis on data transmitted to the post-processing system;

[0030] The machine learning module is used to classify, train, predict, analyze, and calculate the collected data using machine learning methods;

[0031] Deep learning module, used to classify, train, predict, analyze and calculate the collected data using deep learning methods;

[0032] The data dimension reduction module is used to reduce the dimension of the data transmitted to the post-processing system according to the analysis requirements.

[0033] Furthermore, the control module includes: an automatic lifting device motion control module, which is used to control the automatic lifting device to reach a specified position and control the retraction and extension of the simulated nacelle device; a platform exciter control module, which is used to control the landing platform to simulate a specified motion according to an input load spectrum;

[0034] The machine vision measurement unit includes a machine vision camera and adhesive marking points, and non-contact measurement of the flexible part of the installation section is achieved through the adhesive marking points.

[0035] The multi-modal information fusion shipborne engine basic impact simulation test method adopts the multi-modal information fusion shipborne engine basic impact simulation test system for testing, and includes the following steps:

[0036] S1: Determine the test requirements, preliminarily determine the test objects, and preliminarily determine the test working conditions according to the test tasks;

[0037] S2: Determine the module operation strategy of the multimodal measurement system and post-processing system according to the test requirements;

[0038] S3: Finalize the test object and the mounting frame and mounting section that match the test object, and assemble the mechanical structure module;

[0039] S4: Finalize the working condition requirements of the simulation test, and conduct and complete the test according to the working condition requirements.

[0040] Furthermore, S4 specifically includes the following sub-steps:

[0041] S401: According to the simulation test working condition requirements initially determined in S1, combined with the mechanical structure module, the simulation test working condition requirements are finally determined;

[0042] S402: Input the simulated velocity value v of the aircraft engine drop impact, and input the simulated load spectrum of the ship vibration;

[0043] S403: Outputting the target free fall height h of the simulated nacelle device;

[0044] S404: measuring the distance x between the simulated nacelle device and the landing platform by means of a second laser rangefinder, and feeding back the distance x to the automatic lifting module;

[0045] S405: Control the lifting position of the simulated nacelle device through the automatic lifting module so that x=h;

[0046] S406: running the module type and strategy according to the module operation strategy of the multimodal measurement system determined in S2, at which time the multimodal measurement system enters a specified state to start data collection, and at the same time, the platform exciter control module starts to install a simulated load spectrum to perform a specified motion simulation;

[0047] S407: The aircraft engine enters the specified working state, releases the simulated nacelle device, and starts free fall;

[0048] S408: The multi-modal measurement system continues to collect data from each measuring point of the aircraft engine until a certain time after the vibration ends;

[0049] S409: The aircraft engine exits the specified state;

[0050] S410: Each component module of the multimodal measurement system pre-processes the collected data and transmits it to the post-processing system;

[0051] S411: operating the post-processing system according to the module operation strategy of the post-processing system determined in S2, and performing data analysis and calculation;

[0052] S412: The post-processing system transmits the calculation results to the visualization module;

[0053] S413: The test is completed and the system returns to the initial state.

[0054] Furthermore, when solving the mapping relationship between the surface signal characteristics of the carrier-based engine casing and the engine rotor-casing clearance under the landing condition, the following steps are included:

[0055] S1: The test requirement is to solve the mapping relationship between the signal characteristics of the carrier engine casing surface and the engine rotor-casing clearance under the landing condition. The test object is preliminarily determined to be a specific type of carrier engine, and the test condition is preliminarily determined to be the landing impact condition;

[0056] S2: Determine the operation strategy of the multimodal measurement system module, and operate the acceleration sensor, laser rangefinder 1, capacitive sensor and laser rangefinder 2;

[0057] Determine the module operation strategy of the post-processing system, using the general method module and the machine learning module;

[0058] S3: Determine that the test object is a specific shipborne engine; determine the structural parameters of the shipborne mounting frame and the mounting node; obtain the mounting frame and the mounting node structure and complete the assembly of the mechanical structure module;

[0059] S401: Based on the landing impact working condition requirements initially determined in S1 and combined with the mechanical structure module, the working condition requirements of the simulation test are finally determined;

[0060] S402: Inputting a ship load spectrum to control the platform exciter to simulate the vibration load of the ship at sea;

[0061] S403-S405: different pulse widths of acceleration of the engine casing are simulated by the landing platform, and different peak values ​​of acceleration of the engine casing are simulated by adjusting the landing height by the laser rangefinder 2;

[0062] S406-S409: Conduct engine landing simulation tests under different working conditions and collect data through a multi-modal measurement system, specifically: collect engine casing acceleration signals through an acceleration sensor, collect engine casing displacement signals through a laser rangefinder, and collect engine rotor-casing clearance signals through a capacitive sensor;

[0063] S410: After the acquisition is completed, the multimodal measurement system performs data processing and feature extraction, specifically including: performing data preprocessing on the acquired signals, extracting the peak values ​​Ajp of the acceleration signal and the displacement signal respectively in 1 , Djp in 2 、Pulse width Ajw in 1 、Djw in 2 , extract the minimum engine rotor-casing clearance value Imin of the clearance signal in 3 , where i is the data result of the i-th test, n is the n-th measurement point obtained by the multimodal measurement system module, and the upper right corner mark of n is used to distinguish different measurement modules;

[0064] Then the data is transmitted to the post-processing system, and different landing condition cycles S403-S409 are adjusted; the data features obtained through i-cycle tests constitute the data set including: {Ajp} n 1 、{Djp}n 2 、{Ajw} n 1 、{Djw} n 2 and {Imin} n 3 ;

[0065] S411: The post-processing system uses the general method module and the machine learning module to train the shipboard engine casing surface signal characteristics and the engine rotor-casing clearance mapping model for all data sets, and performs error analysis;

[0066] S412, selecting a model with the smallest error through error analysis, outputting it to a visualization module, and completing the construction of a mapping matrix;

[0067] S413: The test is completed and the system returns to the initial state.

[0068] Furthermore, when used to establish a carrier-borne engine-mounting node-mounting frame state prediction model under a landing condition, the following steps are included:

[0069] S1: Determine the test requirement to establish a shipborne engine-mounting section-mounting frame state prediction model under the landing condition, preliminarily determine the test object as a specific shipborne engine-mounting section-mounting frame, and preliminarily determine the test condition as the landing impact condition;

[0070] S2: Determine the operation strategy of the multimodal measurement system module, and operate the acceleration sensor, strain sensor, machine vision measurement unit, laser rangefinder 1, capacitive sensor and laser rangefinder 2;

[0071] Determine the module operation strategy of the post-processing system, the machine learning module and the deep learning module;

[0072] S3: Determine that the test object is a specific shipborne engine, determine the structural parameters of the shipborne mounting frame and the mounting node, and the mounting node is a specified vibration isolator; obtain the mounting frame and the mounting node structure and complete the assembly of the mechanical structure module; wherein, the pasting position of the shipborne engine casing acceleration sensor should be consistent with the effective acceleration sensor position of the specific shipborne engine during service;

[0073] S401: Based on the landing impact working condition requirements initially determined in S1 and combined with the mechanical structure module, the working condition requirements of the simulation test are finally determined;

[0074] S402: Inputting a ship load spectrum to control the platform exciter to simulate the vibration load of the ship at sea;

[0075] S403-S405: different pulse widths of acceleration of the engine casing are simulated by using the landing platform, and different peak values ​​of acceleration of the engine casing are simulated by adjusting the different heights of the landing platform by using the laser rangefinder 2;

[0076] S406-S409: Conduct engine landing simulation tests under different working conditions, and collect data through a multi-modal measurement system, specifically: measure the acceleration signals of the engine casing and the mounting frame through an acceleration sensor, collect the force conditions of the rigid part of the mounting section through a strain sensor, collect the displacement images of the measuring points of the flexible part of the mounting section through a machine vision measurement unit, measure the displacement signal of the engine casing through a laser rangefinder, and measure the engine rotor-casing clearance signal through a capacitive sensor;

[0077] S410: After the acquisition is completed, the multimodal measurement system performs data processing and feature extraction, specifically including: performing data preprocessing on the acquired signals, and extracting acceleration signals of the engine casing and the mounting bracket [Aj] respectively. in 1 and [Aa] in 2 , the peak value of the acceleration signal of the mounting frame Aap in 2 ; Calculate the maximum stress value σp through the strain information of the rigid part of the installation node in 3 ; The displacement signal of the measuring point of the flexible part of the installation node is calculated by the algorithm based on the collected displacement image [Dg] in 4 , peak value Dgp of displacement curve in 4 ; Extract the peak value Djp of the casing displacement signal in 5 ; Extract the minimum engine rotor-casing clearance value Imin for the clearance signal in 6 , where i is the data result of the i-th test, n is the n-th measurement point obtained by the multimodal measurement system module, and the upper right corner mark of n is used to distinguish different measurement modules;

[0078] The data is transmitted to the post-processing system, and the different landing conditions are adjusted in the cycle steps S403-S409; the data set composed of the data features obtained through the i-cycle test includes: {[Aj]} n 1 、{Aap} n 2 , {σp} n 3 、{Dgp} n 4 、{Djp} n 5 and {Imin}n 6 ;

[0079] The safety factor S of the peak displacement of the engine casing, the peak acceleration of the mounting frame, the maximum stress value of the rigid part of the vibration isolator, the peak displacement of the flexible part of the mounting node and the minimum engine rotor-casing clearance value are constructed respectively. Djp , S Aap , S σp , S Dgp , S Imin ; Set the data set {Aap} n 2 , {σp} n 3 、{Dgp} n 4 、{Djp} n 5 and {Imin} n 6 Each data L im Compare them with the corresponding safety factors respectively, and output new data sets {Aap*} according to the rule that if the data is less than the safety factor, it is 0, and if it is greater than the safety factor, it is 1. n 2 , {σp*} n 3 、{Dgp*} n 4 、{Djp*} n 5 and {Imin*} n 6 ; Among them, data L im Represents the i-th eigenvalue of the m-th data set, m = n 2 +n 3 +n 4 +n 5 +n 6 ;

[0080] S411: Run the post-processing system according to the module operation strategy of the post-processing system determined in S2 to perform data analysis and calculation; specifically, convert the new data set {Aap*} n 2 , {σp*} n 3 、{Dgp*} n 4 、{Djp*} n 5 and {Imin*} n 6 As the output of the prediction model, the dataset {[Aj]} n 1As the input of the prediction model, the output data set is combined with the input data set respectively, and the machine learning module and deep learning module are run through the post-processing system to perform training and error analysis of the prediction model; the model with the smallest error is selected through error analysis, that is, a prediction model of the engine and its mounting section and mounting bracket based on the engine case acceleration signal is constructed.

[0081] S412: The post-processing system transmits the calculation results to the visualization module;

[0082] S413: The test is completed and the system returns to the initial state.

[0083] Furthermore, the acceleration information collected by the acceleration sensor during the impact, the image information collected by the machine vision measurement unit, the strain information collected by the strain sensor, and the distance information collected by the laser rangefinder 1, the laser tester 2 and the capacitive sensor are fused at the feature level according to the needs of the test task; the feature level fusion method includes: splicing of data information, conversion, extraction and combination of data information features; and decision-level fusion and model-level fusion of signals transmitted by the multimodal measurement system.

[0084] Compared with the prior art, the present invention can achieve the following beneficial effects: the multi-modal information fusion carrier engine basic shock simulation test system and method in the present invention can realize the state information collection and analysis of various target analysis structures under different carrier engines and different shock working condition simulation tests, and can provide analysis means and data sources for the performance evaluation of carrier engines and related structures under basic shock. The scheme of the present invention realizes one system for multiple uses, that is, one system can solve the test and analysis tasks of different carrier engines, different shock working conditions and different target analysis objects, which can save the cost of conducting basic shock tests of aircraft engines, shorten the construction period of different engine shock test benches, and improve the research and development speed of my country's new aircraft engines. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] Figure 1 The schematic diagram of the shipborne engine foundation impact simulation test system according to the multi-modal information fusion provided in the first embodiment of the present invention is Figure 1 ;

[0086] Figure 2 The schematic diagram of the shipborne engine foundation impact simulation test system according to the multi-modal information fusion provided in the first embodiment of the present invention is Figure 2 ;

[0087] Figure 3 1 is a schematic diagram of the overall structure of a shipborne engine basic impact simulation test system with multi-modal information fusion provided according to the first embodiment of the present invention;

[0088] Figure 4It is a partial structural schematic diagram of a shipborne engine basic impact simulation test system with multi-modal information fusion provided according to the first embodiment of the present invention;

[0089] Figure 5 The process of the shipborne engine foundation impact simulation test method provided by the multi-modal information fusion provided in the first embodiment of the present invention is as follows: Figure 1 ;

[0090] Figure 6 The process of the shipborne engine foundation impact simulation test method provided by the multi-modal information fusion provided in the first embodiment of the present invention is as follows: Figure 2 ;

[0091] Figure 7 It is a flow chart of a carrier-based engine foundation impact simulation test method with multi-modal information fusion provided according to the second embodiment of the present invention.

[0092] The reference numerals therein include: automatic lifting device 101, simulated nacelle device 102, mounting frame 103, test piece casing part 104-1, test piece rotor system 104-2, mounting section 105, landing platform 106, platform vibrator 107, acceleration sensor 201, strain sensor 202, machine vision camera 203-1, pasting mark point 203-2, laser rangefinder one 204, capacitive sensor 205, laser rangefinder two 206. DETAILED DESCRIPTION

[0093] In the following, reference will be made to the Figure 1-7 Embodiments of the present invention are described. In the following description, the same modules are represented by the same reference numerals. In the case of the same reference numerals, their names and functions are also the same. Therefore, the detailed description thereof will not be repeated.

[0094] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the following Figure 1-7 It should be understood that the specific embodiments described herein are only used to explain the present invention and do not constitute a limitation of the present invention.

[0095] Embodiment 1

[0096] Multi-modal information fusion shipborne engine basic impact simulation test system, such as Figure 1 , Figure 2 As shown, it includes a mechanical structure module, a multimodal measurement system, a storage module, a control module, a post-processing system, a data transmission module, a visualization module, a judgment module and an accessory system.

[0097] The mechanical structure module is used to implement test tasks such as Figure 3 , Figure 4As shown, the mechanical structure module includes a simulated nacelle device 102, an automatic lifting device 101, a test piece to be tested, a mounting frame 103, a mounting section 105, a landing platform 106 and a platform exciter 107. The details are as follows:

[0098] The simulated nacelle device 102 is used to simulate the nacelle of the aircraft where the aircraft engine is located, and its structural strength and capacity correspond to the tested aircraft model;

[0099] An automatic lifting device 101, located above the simulated nacelle device 102, is used to suck up, lift or release the simulated nacelle device 102;

[0100] The test piece to be tested is located in the simulated nacelle device 102 and is an aircraft engine for testing or an equivalent body with the same test function, and can be divided into two parts: a test piece casing part 104-1 and a test piece rotor system 104-2;

[0101] A mounting frame 103, located in the simulated nacelle device 102 and connected to the simulated nacelle device 102, and used for mounting and fixing the test piece to be tested;

[0102] The mounting node 105 is located in the simulated nacelle device 102 and is used to connect the mounting frame 103 and the test piece to be tested. In some types of aircraft engines, it also has the function of blocking the vibration transmission between the aircraft engine and the aircraft. The number of the mounting node 105 is determined according to the installation requirements of the test piece to be tested. The mounting node 105 includes a flexible part and a rigid part. The rigid part of the mounting node 105 includes a metal frame of the mounting node 105, a connecting screw, etc.;

[0103] The landing platform 106 is located below the simulated nacelle device 102 and is used to carry the landing simulated nacelle device 102 and form a required basic impact load therewith;

[0104] The platform exciter 107 is located below the landing platform 106. When simulating the basic impact load of landing, it is in a static state. When simulating the basic impact load with time-varying characteristic load, it enters the running state and simulates the time-varying load through the corresponding vibration spectrum.

[0105] The multimodal measurement system is used for data acquisition and preprocessing. The multimodal measurement system includes a measurement hardware system, a positioning mode system, and a measurement and test system. The details are as follows:

[0106] Measurement hardware system, used to realize multi-modal measurement system data acquisition, including:

[0107] The acceleration sensor 201 disposed on the engine casing and the mounting frame 103 is used to measure the acceleration information transmitted to the measuring point during the basic impact process;

[0108] The strain sensor 202 provided on the mounting joint 105 is used to measure the stress on the rigid part of the engine mounting joint;

[0109] The machine vision measurement unit disposed on the landing platform 106 is used to measure the displacement information of the measuring points of the flexible part of the installation section; the machine vision measurement unit includes a machine vision camera 203-1 and a pasting mark point 203-2, and the non-contact measurement of the flexible part of the installation section 105 is realized by pasting the mark point 203-2;

[0110] A laser rangefinder 204 disposed below the mounting frame 103 and located on the base of the simulated nacelle device 102 is used to monitor the vertical motion trajectory of the aircraft engine during the foundation impact process in real time;

[0111] A bipolar plate capacitance sensor 205 is arranged at a specific position of the engine rotor blade tip and the casing at the test piece to be tested, and is used to measure the clearance between the engine rotor and the casing during the basic impact process;

[0112] The second laser rangefinder 206 disposed at the bottom of the simulated nacelle device 102 is used to measure the distance X between the simulated nacelle device and the landing platform.

[0113] The positioning mode system is used in conjunction with the laser rangefinder 206 to measure the distance X between the simulated nacelle device 102 and the landing platform 106 and to assist the control module in realizing the motion decision of the mechanical structure module.

[0114] The measurement and testing system can be used in conjunction with all measurement hardware systems to complete multimodal data measurement tasks.

[0115] Each component module of the multimodal fusion measurement system operates independently, that is, the specified measurement module can be run according to the specific test task. At the same time, each measurement module has a signal processing function and can perform signal processing operations such as filtering and smoothing at a specified frequency.

[0116] The storage module is used to store the measurement data collected by the multimodal measurement system.

[0117] The control module is used to control the mechanical structure module to realize movement. The control module includes an automatic lifting device motion control module and a platform exciter control module. The automatic lifting device motion control module is used to control the automatic lifting device 101 to reach the specified position and control the retraction and extension of the simulated nacelle device 102, and the platform exciter control module is used to control the landing platform 106 to simulate the specified movement according to the input load spectrum.

[0118] The data transmission module is used to transmit the data in the storage module.

[0119] The post-processing system is used to realize the fusion of multiple modal data information transmitted by the data transmission module. The post-processing system includes a general method module, a machine learning module, a deep learning module and a data dimension reduction module. The details are as follows:

[0120] The general method module is used to perform general method analyses such as descriptive analysis, difference analysis, and correlation analysis on the data transmitted to the post-processing system;

[0121] The machine learning module is used to classify, train, predict, analyze, and calculate the collected data using machine learning methods;

[0122] Deep learning module, used to classify, train, predict, analyze and calculate the collected data using deep learning methods;

[0123] The data dimension reduction module is used to reduce the dimension of the data transmitted to the post-processing system according to the analysis requirements, and also has the function of adding new tags.

[0124] The visualization module is used to visualize the data and models fused by the post-processing system.

[0125] The judgment module is used to judge the operation strategy.

[0126] Accessory system for running the test piece to be tested.

[0127] Figure 2 The direction of the arrows in the figure represents the logical relationship of information transmission among the modules of the system.

[0128] The multi-modal information fusion shipborne engine basic impact simulation test method adopts the multi-modal information fusion shipborne engine basic impact simulation test system for testing, such as Figure 5 As shown, there are four stages in total, including the following steps:

[0129] Phase 1:

[0130] S1: Determine the test requirements, preliminarily determine the test objects, and preliminarily determine the test working conditions based on the test tasks.

[0131] Phase 2:

[0132] S2: Determine the module operation strategy of the multimodal measurement system and post-processing system according to the test requirements.

[0133] Phase 3:

[0134] S3: Finally determine the test object and the mounting frame 103 and mounting section 105 that match the test object, and assemble the mechanical structure module. Specifically, it includes the following sub-steps:

[0135] S301: According to the test object initially determined in S1 and combined with the analysis requirements in S2, the test object is finally determined;

[0136] S302: Determine the structural parameters of the matching mounting frame 103 and mounting node 105;

[0137] S303: Obtain matching mounting frame 103 and mounting node 105 structures;

[0138] S304: Assemble the mechanical structure modules required to complete the test task.

[0139] Phase 4:

[0140] S4: Finally determine the working condition requirements of the simulation test, and conduct and complete the test according to the working condition requirements. S4 specifically includes the following sub-steps:

[0141] S401: According to the simulation test working condition requirements initially determined in S1, combined with the mechanical structure module, the simulation test working condition requirements are finally determined;

[0142] S402: Input the simulated velocity value v of the aircraft engine drop impact, and input the simulated load spectrum of the ship vibration;

[0143] S403: Outputting the target free fall height h of the simulated nacelle device 102;

[0144] S404: measuring the distance x between the simulated nacelle device 102 and the landing platform 106 by means of the second laser rangefinder 206, and feeding back to the automatic lifting module;

[0145] S405: controlling the lifting position of the simulated nacelle device 102 through the automatic lifting module so that x=h;

[0146] S406: running the module type and strategy according to the module operation strategy of the multimodal measurement system determined in S2, at which time the multimodal measurement system enters a specified state to start data collection, and at the same time, the platform exciter control module starts to install a simulated load spectrum to perform a specified motion simulation;

[0147] S407: the aircraft engine (or its equivalent device) enters a specified working state, releases the simulated nacelle device 102, and starts free fall;

[0148] S408: The multi-modal measurement system continues to collect data from each measuring point of the aircraft engine until a certain time after the vibration ends;

[0149] S409: The aircraft engine exits the specified state;

[0150] S410: Each component module of the multimodal measurement system pre-processes the collected data and transmits it to the post-processing system;

[0151] S411: operating the post-processing system according to the module operation strategy of the post-processing system determined in S2, and performing data analysis and calculation;

[0152] S412: The post-processing system transmits the calculation results to the visualization module for the test personnel to receive and analyze the target structure performance evaluation results;

[0153] S413: The test is completed and the system returns to the initial state.

[0154] This embodiment is specifically applied to the following specific scenarios: when solving the mapping relationship between the surface signal characteristics of the carrier-based engine casing and the engine rotor-casing clearance under the landing condition, such as Figure 6 As shown, the following steps are included:

[0155] S1: The test requirement is to solve the mapping relationship between the signal characteristics of the carrier engine casing surface and the engine rotor-casing clearance under the landing condition. The test object is preliminarily determined to be a specific type of carrier engine, and the test condition is preliminarily determined to be the landing impact condition;

[0156] S2: Determine the operation strategy of the multi-modal measurement system module, and operate the acceleration sensor 201, the laser rangefinder 1 204, the capacitive sensor 205, and the laser rangefinder 2 206;

[0157] Determine the module operation strategy of the post-processing system, using the general method module and the machine learning module;

[0158] S3: Determine that the test object is a specific shipborne engine; determine the structural parameters of the shipborne mounting frame 103 and the mounting node 105; obtain the structure of the mounting frame 103 and the mounting node 105 and complete the assembly of the mechanical structure module;

[0159] S401: Based on the landing impact working condition requirements initially determined in S1 and combined with the mechanical structure module, the working condition requirements of the simulation test are finally determined;

[0160] S402: inputting a ship load spectrum to control the platform exciter 107 to simulate the vibration load of the ship at sea;

[0161] S403-S405: simulating different pulse widths of acceleration of the engine casing by adjusting the cushion of the landing platform 106, and simulating different peak values ​​of acceleration of the engine casing by adjusting the landing height by the laser rangefinder 206;

[0162] S406-S409: Perform engine landing simulation tests under different working conditions, and collect data through a multi-modal measurement system, specifically: collect engine casing acceleration signals through an acceleration sensor 201, collect engine casing displacement signals through a laser rangefinder 204, and collect engine rotor-casing clearance signals through a capacitive sensor 205;

[0163] S410: After the acquisition is completed, the multimodal measurement system performs data processing and feature extraction, specifically including: performing data preprocessing such as denoising and filtering on the acquired signals, and extracting the peak values ​​Ajp of the acceleration signal and the displacement signal respectively. in 1 , Djp in 2 、Pulse width Ajw in 1 、Djw in 2 , extract the minimum engine rotor-casing clearance value Imin of the clearance signal in 3 , where i is the data result of the i-th test, n is the n-th measurement point obtained by the multimodal measurement system module, and the upper right corner mark of n is used to distinguish different measurement modules;

[0164] Then the data is transmitted to the post-processing system, and different landing condition cycles S403-S409 are adjusted; the data features obtained through i-cycle tests constitute the data set including: {Ajp} n 1 、{Djp} n 2 、{Ajw} n 1 、{Djw} n 2 and {Imin} n 3 ;

[0165] S411: The post-processing system uses the general method module and the machine learning module to train the shipborne engine casing surface signal characteristics and the engine rotor-casing clearance mapping model for all data sets, including curve fitting, different machine learning model training, etc., and performs error analysis;

[0166] S412, selecting a model with the smallest error through error analysis, outputting it to a visualization module, and completing the construction of a mapping matrix;

[0167] S413: The test is completed and the system returns to the initial state.

[0168] The mapping relationship matrix obtained by this method can be used by researchers to quantitatively analyze the intrinsic relationship between the state of the carrier-based aircraft engine casing under landing conditions and the minimum clearance between the engine rotor and casing, and can provide an experimental basis and theoretical basis for the impact resistance design of carrier-based engines for landing conditions.

[0169] At the same time, the acceleration information collected by the acceleration sensor 201 during the impact process, the image information collected by the machine vision measurement unit, the strain information collected by the strain sensor 202, and the distance information collected by the laser rangefinder 1 204, the laser rangefinder 2 206 and the capacitive sensor 205 can be fused at the feature level according to the needs of the test task; the feature level fusion method includes: splicing of data information, conversion, extraction and combination of data information features; decision-level fusion and model-level fusion of signals transmitted by the multimodal measurement system.

[0170] Embodiment 2

[0171] The difference between this embodiment and the embodiment is that this embodiment is used for the following specific scenario, when used to establish a state prediction model of the carrier engine-installation section 105-installation frame 103 under the landing condition, such as Figure 7 As shown, the following steps are included:

[0172] S1: Determine the test requirement to establish a state prediction model of the carrier engine-installation section 105-installation frame 103 under the landing condition, preliminarily determine the test object to be a specific carrier engine-installation section 105-installation frame 103, and preliminarily determine the test condition to be a landing impact condition;

[0173] S2: Determine the operation strategy of the multimodal measurement system module, and operate the acceleration sensor 201, the strain sensor 202, the machine vision measurement unit, the laser rangefinder 1 204, the capacitive sensor 205 and the laser rangefinder 2 206;

[0174] Determine the module operation strategy of the post-processing system, the machine learning module and the deep learning module;

[0175] S3: Determine that the test object is a specific shipborne engine, determine the structural parameters of the shipborne mounting frame 103 and the mounting node 105, and the mounting node 105 is a specified vibration isolator; obtain the structure of the mounting frame 103 and the mounting node 105 and complete the assembly of the mechanical structure module; wherein, the pasting position of the shipborne engine casing acceleration sensor 201 should be consistent with the effective acceleration sensor 201 position of the specific shipborne engine during service;

[0176] S401: Based on the landing impact working condition requirements initially determined in S1 and combined with the mechanical structure module, the working condition requirements of the simulation test are finally determined;

[0177] S402: inputting a ship load spectrum to control the platform exciter 107 to simulate the vibration load of the ship at sea;

[0178] S403-S405: simulating different pulse widths of acceleration of the engine casing by adjusting the cushion of the landing platform 106, and simulating different peak values ​​of acceleration of the engine casing by adjusting different heights of the landing platform 106 by the laser rangefinder 206;

[0179] S406-S409: Conduct engine landing simulation tests under different working conditions, and collect data through a multimodal measurement system, specifically: measure the acceleration signals of the engine casing and the mounting frame 103 through the acceleration sensor 201, collect the force conditions of the rigid part of the mounting joint 105 through the strain sensor 202, collect the displacement image of the measuring point of the flexible part of the mounting joint 105 through the machine vision measurement unit, measure the displacement signal of the engine casing through the laser rangefinder 204, and measure the engine rotor-casing clearance signal through the capacitive sensor 205;

[0180] S410: After the acquisition is completed, the multimodal measurement system performs data processing and feature extraction, specifically including: performing data preprocessing such as denoising and filtering on the acquired signals, and extracting acceleration signals [Aj] of the engine casing and the mounting frame 103 respectively. in 1 and [Aa] in 2 , the peak value Aap of the acceleration signal of the mounting frame 103 in 2 ; Calculate the maximum stress value σp through the strain information of the rigid part of the installation node 105 in 3 ; The displacement signal [Dg] of the measuring point of the flexible part of the installation section 105 is calculated by binocular vision and image processing algorithm for the collected displacement image in 4 , peak value Dgp of displacement curve in 4 ; Extract the peak value Djp of the casing displacement signal in 5 ; Extract the minimum engine rotor-casing clearance value Imin for the clearance signal in 6 , where i is the data result of the i-th test, n is the n-th measurement point obtained by the multimodal measurement system module, and the upper right corner mark of n is used to distinguish different measurement modules;

[0181] The data is transmitted to the post-processing system, and the different landing conditions are adjusted in the cycle steps S403-S409; the data set composed of the data features obtained through the i-cycle test includes: {[Aj]} n 1 、{Aap}n 2 , {σp} n 3 、{Dgp} n 4 、{Djp} n 5 and {Imin} n 6 ;

[0182] The safety factor S of the peak displacement of the engine casing, the peak acceleration of the mounting frame 103, the maximum stress value of the rigid part of the vibration isolator, the peak displacement of the flexible part of the mounting node 105 and the minimum engine rotor-casing clearance value are constructed respectively. Djp , S Aap , S σp , S Dgp , S Imin ; Set the data set {Aap} n 2 , {σp} n 3 、{Dgp} n 4 、{Djp} n 5 and {Imin} n 6 Each data L im Compare them with the corresponding safety factors respectively, and output new data sets {Aap*} according to the rule that if the data is less than the safety factor, it is 0, and if it is greater than the safety factor, it is 1. n 2 , {σp*} n 3 、{Dgp*} n 4 、{Djp*} n 5 and {Imin*} n 6 ; Among them, data L im Represents the i-th eigenvalue of the m-th data set, m = n 2 +n 3 +n 4 +n 5 +n 6 ;

[0183] S411: Run the post-processing system according to the module operation strategy of the post-processing system determined in S2 to perform data analysis and calculation; specifically, convert the new data set {Aap*} n 2 , {σp*} n 3 、{Dgp*} n4 、{Djp*} n 5 and {Imin*} n 6 As the output of the prediction model, the dataset {[Aj]} n 1 As the input of the prediction model, the output data set is combined with the input data set respectively, and the machine learning module and the deep learning module are run through the post-processing system to perform training and error analysis of the prediction model; the model with the smallest error is selected through error analysis, that is, a prediction model of the engine and its mounting node 105 and mounting bracket 103 based on the engine casing acceleration signal is constructed.

[0184] S412: The post-processing system transmits the calculation results to the visualization module;

[0185] S413: The test is completed and the system returns to the initial state.

[0186] The prediction model obtained by this method is transferred to the in-service shipborne engines and can be used for the status assessment of the engine-vibration isolation system mounting frame during the carrier-based aircraft landing process.

[0187] Although the embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and cannot be understood as limiting the present invention. Those skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

[0188] The above specific implementations of the present invention do not constitute a limitation on the protection scope of the present invention. Any other corresponding changes and modifications made based on the technical concept of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A multi-modal information fusion shipborne engine foundation impact simulation test method, characterized in that: The steps include: S1: Determine the test requirements, preliminarily determine the test objects, and preliminarily determine the test working conditions according to the test tasks; S2: Determine the module operation strategy of the multimodal measurement system and post-processing system according to the test requirements; S3: finally determining the test object and the mounting frame (103) and mounting section (105) matched with the test object, and assembling the mechanical structure module; S4: Finally determine the working condition requirements of the simulation test, and conduct and complete the test according to the working condition requirements; S4 specifically includes the following sub-steps: S401: According to the simulation test working condition requirements initially determined in S1, combined with the mechanical structure module, the simulation test working condition requirements are finally determined; S402: Input the simulated velocity value v of the aircraft engine drop impact, and input the simulated load spectrum of the ship vibration; S403: Outputting a target free fall height h of the simulated nacelle device (102); S404: measuring the distance x between the simulated nacelle device (102) and the landing platform (106) by means of a second laser rangefinder (206), and feeding back the distance x to the automatic lifting module; S405: controlling the lifting position of the simulated nacelle device (102) through the automatic lifting module so that x=h; S406: running the module type and strategy according to the module operation strategy of the multimodal measurement system determined in S2, at which time the multimodal measurement system enters a specified state to start data collection, and at the same time, the platform exciter control module starts to install a simulated load spectrum to perform a specified motion simulation; S407: the aircraft engine enters a specified working state, releases the simulated nacelle device (102), and starts free fall; S408: The multi-modal measurement system continues to collect data from each measuring point of the aircraft engine until a certain time after the vibration ends; S409: The aircraft engine exits the specified state; S410: Each component module of the multimodal measurement system pre-processes the collected data and transmits it to the post-processing system; S411: operating the post-processing system according to the module operation strategy of the post-processing system determined in S2, and performing data analysis and calculation; S412: The post-processing system transmits the calculation results to the visualization module; S413: The test ends and the system returns to the initial state; When solving the mapping relationship between the surface signal characteristics of the carrier-based engine casing and the engine rotor-casing clearance under the landing condition, the following steps are included: S1: The test requirement is to solve the mapping relationship between the signal characteristics of the carrier engine casing surface and the engine rotor-casing clearance under the landing condition. The test object is preliminarily determined to be a specific type of carrier engine, and the test condition is preliminarily determined to be the landing impact condition; S2: Determine the operation strategy of the multi-modal measurement system module, and operate the acceleration sensor (201), the laser rangefinder 1 (204), the capacitive sensor (205), and the laser rangefinder 2 (206); Determine the module operation strategy of the post-processing system, using the general method module and the machine learning module; S3: Determine that the test object is a specific shipborne engine; determine the structural parameters of the shipborne mounting frame (103) and the mounting node (105); obtain the structure of the mounting frame (103) and the mounting node (105) and complete the assembly of the mechanical structure module; S401: Based on the landing impact working condition requirements initially determined in S1 and combined with the mechanical structure module, the working condition requirements of the simulation test are finally determined; S402: inputting a ship load spectrum to control the platform exciter (107) to simulate the vibration load of the ship at sea; S403-S405: simulating different pulse widths of acceleration of the engine casing by using the landing platform (106), and simulating different peak values ​​of acceleration of the engine casing by adjusting the landing height by using the second laser rangefinder (206); S406-S409: performing engine landing simulation tests under different working conditions, and collecting data through a multi-modal measurement system, specifically: collecting engine casing acceleration signals through an acceleration sensor (201), collecting engine casing displacement signals through a laser rangefinder (204), and collecting engine rotor-casing clearance signals through a capacitive sensor (205); S410: After the acquisition is completed, the multimodal measurement system performs data processing and feature extraction, specifically including: performing data preprocessing on the acquired signals, extracting the peak values ​​Ajp of the acceleration signal and the displacement signal respectively in 1 , Djp in 2 、Pulse width Ajw in 1 、Djw in 2 , extract the minimum engine rotor-casing clearance value Imin of the clearance signal in 3 , where i is the data result of the i-th test, n is the n-th measurement point obtained by the multimodal measurement system module, and the upper right corner mark of n is used to distinguish different measurement modules; Then the data is transmitted to the post-processing system, and different landing condition cycles S403-S409 are adjusted; the data features obtained through i-cycle tests constitute the data set including: {Ajp} n 1 、{Djp} n 2 、{Ajw} n 1 、{Djw} n 2 and {Imin} n 3 ; S411: The post-processing system uses the general method module and the machine learning module to train the shipboard engine casing surface signal characteristics and the engine rotor-casing clearance mapping model for all data sets, and performs error analysis; S412, selecting a model with the smallest error through error analysis, outputting it to a visualization module, and completing the construction of a mapping matrix; S413: The test ends and the system returns to the initial state; Or when used to establish a state prediction model of a carrier engine-installation section (105)-installation frame (103) under a landing condition, the method comprises the following steps: S1: determining the test requirement to establish a state prediction model of a shipborne engine-installation section (105)-installation frame (103) under a landing condition, preliminarily determining the test object to be a specific shipborne engine-installation section (105)-installation frame (103), and preliminarily determining the test condition to be a landing impact condition; S2: Determine the operation strategy of the multimodal measurement system module, and operate the acceleration sensor (201), the strain sensor (202), the machine vision measurement unit, the laser rangefinder 1 (204), the capacitive sensor (205) and the laser rangefinder 2 (206); Determine the module operation strategy of the post-processing system, the machine learning module and the deep learning module; S3: Determine that the test object is a specific shipborne engine, determine the structural parameters of the shipborne mounting frame (103) and the mounting node (105), and the mounting node (105) is a specified vibration isolator; obtain the structure of the mounting frame (103) and the mounting node (105) and complete the assembly of the mechanical structure module; wherein the pasting position of the shipborne engine casing acceleration sensor (201) should be consistent with the effective acceleration sensor (201) position of the specific shipborne engine during service; S401: Based on the landing impact working condition requirements initially determined in S1 and combined with the mechanical structure module, the working condition requirements of the simulation test are finally determined; S402: inputting a ship load spectrum to control the platform exciter (107) to simulate the vibration load of the ship at sea; S403-S405: A simulation of different pulse widths of acceleration of the engine casing is realized by using the landing platform (106), and a simulation of different peak values ​​of acceleration of the engine casing is realized by adjusting different heights of the landing platform (106) by using the second laser rangefinder (206); S406-S409: Performing engine landing simulation tests under different working conditions, and collecting data through a multi-modal measurement system, specifically: measuring acceleration signals of the engine casing and the mounting frame (103) through an acceleration sensor (201), collecting the stress conditions of the rigid part of the mounting joint (105) through a strain sensor (202), collecting a measurement point displacement image of the flexible part of the mounting joint (105) through a machine vision measurement unit, measuring the engine casing displacement signal through a laser rangefinder (204), and measuring the engine rotor-casing clearance signal through a capacitive sensor (205); S410: After the acquisition is completed, the multi-modal measurement system performs data processing and feature extraction, specifically including: performing data preprocessing on the acquired signals, respectively extracting acceleration signals [Aj] of the engine casing and the mounting frame (103) in 1 and [Aa] in 2 , the peak value Aap of the acceleration signal of the mounting frame (103) in 2 ; Calculate the maximum stress value σp through the strain information of the rigid part of the installation node (105) in 3 ; The displacement signal [Dg] of the measuring point of the flexible part of the installation node (105) is calculated by an algorithm based on the collected displacement image. in 4 , peak value Dgp of displacement curve in 4 ; Extract the peak value Djp of the casing displacement signal in 5 ; Extract the minimum engine rotor-casing clearance value Imin for the clearance signal in 6 , where i is the data result of the i-th test, n is the n-th measurement point obtained by the multimodal measurement system module, and the upper right corner mark of n is used to distinguish different measurement modules; The data is transmitted to the post-processing system, and the different landing conditions are adjusted in the cycle steps S403-S409; the data set composed of the data features obtained through the i-cycle test includes: {[Aj]} n 1 、{Aap} n 2 , {σp} n 3 、{Dgp} n 4 、{Djp} n 5 and {Imin} n 6 ; The safety factor S of the displacement peak value of the engine casing, the acceleration peak value of the mounting frame (103), the maximum stress value of the rigid part of the vibration isolator, the displacement peak value of the flexible part of the mounting section (105) and the minimum engine rotor-casing clearance value are constructed respectively. Djp , S Aap , S σp , S Dgp , S Imin ; Set the data set {Aap} n 2 , {σp} n 3 、{Dgp} n 4 、{Djp} n 5 and {Imin} n 6 Each data L im Compare them with the corresponding safety factors respectively, and output new data sets {Aap*} according to the rule that if the data is less than the safety factor, it is 0, and if it is greater than the safety factor, it is 1. n 2 , {σp*} n 3 、{Dgp*} n 4 、{Djp*} n 5 and {Imin*} n 6 ; Among them, data L im Represents the i-th eigenvalue of the m-th data set, m = n 2 +n 3 +n 4 +n 5 +n 6 ; S411: Run the post-processing system according to the module operation strategy of the post-processing system determined in S2 to perform data analysis and calculation; specifically, convert the new data set {Aap*} n 2 , {σp*} n 3 、{Dgp*} n 4 、{Djp*} n 5 and {Imin*} n 6 As the output of the prediction model, the dataset {[Aj]} n 1 As the input of the prediction model, the output data set is combined with the input data set respectively, and the machine learning module and the deep learning module are run through the post-processing system to perform training and error analysis of the prediction model; the model with the smallest error is selected through the error analysis, that is, the prediction model of the engine and its mounting section (105) and mounting frame (103) based on the acceleration signal of the engine casing is constructed; S412: The post-processing system transmits the calculation results to the visualization module; S413: The test is completed and the system returns to the initial state.

2. The multi-modal information fusion shipborne engine foundation impact simulation test method according to claim 1 is characterized in that: The acceleration information collected by the acceleration sensor (201) during the impact process, the image information collected by the machine vision measurement unit, the strain information collected by the strain sensor (202), the distance information collected by the laser rangefinder 1 (204), the laser rangefinder 2 (206) and the capacitive sensor (205) are fused at multiple information feature levels according to the needs of the test task; Feature-level fusion methods include: data information splicing, data information feature conversion, extraction and combination; Perform decision-level fusion and model-level fusion on the signals transmitted by the multimodal measurement system.

3. A multi-modal information fusion shipborne engine basic impact simulation test system, which uses the simulation test method in claim 2 to perform a basic impact simulation test on a multi-modal information fusion shipborne engine, characterized in that: include: Mechanical structure module, used to realize the test task; Multimodal measurement system for data acquisition and preprocessing; A storage module, used for storing measurement data collected by the multimodal measurement system; A control module, used to control the mechanical structure module to achieve movement; A data transmission module, used for transmitting data in the storage module; A post-processing system, used to realize the fusion of multiple modal data information transmitted by the data transmission module; Visualization module, used to visualize the data and models fused by the post-processing system; A judgment module is used to judge the operation strategy; Accessory system for running the test piece to be tested.

4. The multi-modal information fusion shipborne engine foundation impact simulation test system according to claim 3 is characterized in that: The mechanical structure module comprises: A simulated nacelle device (102) is used to simulate an aircraft nacelle in which an aircraft engine is located; An automatic lifting device (101), located above the simulated nacelle device (102), used for sucking up, lifting or releasing the simulated nacelle device (102); The test piece to be tested is located in the simulated nacelle device (102) and is an aircraft engine used for testing or an equivalent body with the same test function; A mounting frame (103), located in the simulated nacelle device (102) and connected to the simulated nacelle device (102), and used for mounting and fixing the test piece to be tested; An installation section (105), located in the simulated nacelle device (102), and used to connect the installation frame (103) and the test piece to be tested; A landing platform (106), located below the simulated nacelle device (102), and used for carrying the simulated nacelle device (102) for landing; The platform exciter (107) is located below the landing platform (106). When simulating the basic impact load of landing, it is in a static state, and when simulating the basic impact load with time-varying characteristic load, it enters a running state.

5. The multi-modal information fusion shipborne engine foundation impact simulation test system according to claim 4 is characterized in that: The multimodal measurement system comprises: A measurement hardware system for realizing data acquisition of a multi-modal measurement system, comprising: an acceleration sensor (201) arranged on a mounting frame (103), a strain sensor (202) arranged on a mounting section (105), a machine vision measurement unit arranged on a landing platform (106), a laser rangefinder 1 (204) arranged below the mounting frame (103) and located on a base of a simulated nacelle device (102), a bipolar plate capacitance sensor (205) arranged at a test piece to be tested, and a laser rangefinder 2 (206) arranged at the bottom of the simulated nacelle device (102); A positioning mode system, used in conjunction with the second laser rangefinder (206), is used to measure the distance between the simulated nacelle device (102) and the landing platform (106), and to assist the control module in realizing the motion decision of the mechanical structure module; The measurement and testing system is used to complete multimodal data measurement tasks.

6. The multi-modal information fusion shipborne engine foundation impact simulation test system according to claim 5 is characterized in that: The post-processing system comprises: A general method module for performing general method analysis on data transmitted to the post-processing system; The machine learning module is used to classify, train, predict, analyze, and calculate the collected data using machine learning methods; Deep learning module, used to classify, train, predict, analyze and calculate the collected data using deep learning methods; The data dimension reduction module is used to reduce the dimension of the data transmitted to the post-processing system according to the analysis requirements.

7. The multi-modal information fusion shipborne engine foundation impact simulation test system according to claim 6 is characterized in that: The control module comprises: an automatic lifting device motion control module, used for controlling the automatic lifting device (101) to reach a specified position and controlling the retraction and extension of the simulated nacelle device (102); and a platform exciter control module, used for controlling the landing platform (106) to simulate a specified motion according to an input load spectrum; The machine vision measurement unit comprises a machine vision camera (203-1) and a sticking mark point (203-2), and non-contact measurement of the flexible part of the installation node (105) is achieved through the sticking mark point (203-2).

Images