A vertical takeoff and landing aircraft vibration reliability test mechanism and test method

Abstract

The present invention relates to the technical field of vibration reliability testing for vertical takeoff and landing aircraft, and specifically to a vibration reliability testing mechanism and testing method for vertical takeoff and landing aircraft. First, the present invention converts different one-dimensional data into grayscale matrices to better represent the spatial characteristics of the data. Subsequently, multiple grayscale matrices with the same time scale are fused into a multi-channel fusion graph, which not only improves the data analysis efficiency but also reduces the risk of missing important information. The multi-modal information fusion strategy based on causal strength can effectively fuse different types of vibration signals while maintaining the spatial characteristics between the data. The deep learning model can obtain a more comprehensive and efficient feature representation, thus better adapting to the changes under different flight environments and improving the testing accuracy of the vibration reliability of the aircraft.

Description

Technical Field

[0001] The present invention relates to the technical field of vibration reliability testing for vertical takeoff and landing aircraft, and specifically to a vibration reliability testing mechanism and method for vertical takeoff and landing aircraft. Background Art

[0002] An electric vertical takeoff and landing aircraft is an electric aircraft capable of taking off and landing vertically. This electric aircraft combines an electric drive system and the ability of vertical takeoff and landing, so it has the advantage of not requiring a traditional runway and can be used in urban environments, providing a more efficient and environmentally friendly way of air travel. The design of an electric vertical takeoff and landing aircraft usually requires it to be able to operate in complex urban environments, which means that the aircraft needs to be able to cope with challenges of different flight environment variables (such as wind speed, wind direction, etc.) and needs to have high reliability. Therefore, testing the vibration reliability of the aircraft when working in urban environments is particularly important.

[0003] Existing vibration reliability tests mainly simulate the aircraft or test the aircraft in an actual environment. However, pure simulation is difficult to truly restore the actual environment, so the test accuracy is poor. When testing in an actual environment, due to the similarity of the environment within the same day or the same time period, the test environment is relatively single. If full-environment testing is to be achieved, a long test cycle is required. Thus, the existing test devices are difficult to meet the requirements of full-environment testing.

[0004] The vibration reliability test of an electric vertical takeoff and landing aircraft during flight is mainly to evaluate the response ability of the aircraft under various airflow conditions, ensuring that it can still maintain stable flight when encountering wind speed changes, turbulence or other airflow disturbances. When an electric vertical takeoff and landing aircraft encounters airflow disturbances, vibrations or dynamic loads will be generated. Therefore, the reliability of the electric vertical takeoff and landing aircraft can be judged through vibration signals.

[0005] At present, deep learning is usually used for reliability testing and prediction. Deep learning is good at dealing with complex non-linear relationships, can extract useful features from a large amount of experimental data, and perform modeling and prediction through a deep learning model.

[0006] However, most traditional deep learning models are modeled through the correlation between data. The models of this modeling method will be affected by potential confounding variables, namely flight environment variables. Under the influence of flight environment variables, the aircraft will generate a variety of dynamically changing vibration signals, making the reliability test more complex, and then resulting in a decrease in the test accuracy of traditional deep learning models, which is difficult to meet the requirements of the current high test accuracy. Therefore, it is urgently needed to be solved. Summary of the Invention

[0007] To avoid and overcome the technical problems existing in the prior art, the present invention provides a vertical takeoff and landing aircraft vibration reliability test mechanism and a test method. The test mechanism in the present invention can simulate various flight environments to improve the accuracy of aircraft testing. The test method of the present invention can consider the changes of various vibration signals to improve the accuracy of testing.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] A vertical takeoff and landing aircraft vibration reliability test mechanism includes a hexahedron-shaped frame. A plurality of fans arranged in a rectangular array are installed on each of the six faces of the frame, and each fan can operate independently. The fans on the six faces enclose a test space for testing the vibration reliability of the vertical takeoff and landing aircraft.

[0010] A vertical takeoff and landing aircraft vibration reliability test method, which applies the above-mentioned vertical takeoff and landing aircraft vibration reliability test mechanism, includes the following test steps:

[0011] S1. Obtain the timing data of the aircraft during flight tests. The timing data includes vibration signals, flight environment variables, and the corresponding vibration reliability.

[0012] S2. Convert each piece of timing data into the same type of data, and organize the converted data into corresponding grayscale matrices.

[0013] S3. Calculate the weights of each piece of timing data according to the mutual information between the flight environment variables and each vibration signal. Weight the grayscale matrix with this weight to obtain a weighted matrix, and fuse each weighted matrix in sequence to obtain a multi-channel fusion graph.

[0014] S4. Use the multi-channel fusion graph to train a deep learning model to generate a vibration reliability prediction model with vibration signals as features and vibration reliability as labels.

[0015] S5. Obtain the real-time timing data of the aircraft and input it into the vibration reliability prediction model to predict the real-time vibration reliability of the aircraft.

[0016] As a further scheme of the present invention: The specific content of step S3 is as follows:

[0017] S31. Calculate the mutual information between the flight environment variables and each vibration signal.

[0018] S32. Calculate the weights of each vibration signal based on the mutual information.

[0019] S33. Weight the grayscale matrix of each vibration signal with the weight to obtain the corresponding weighted matrix.

[0020] S34. Take each weighted matrix as a channel and perform fusion in sequence to obtain a multi-channel fusion graph with the corresponding number of channels.

[0021] As a further solution of the present invention, the calculation formula of mutual information is as follows:

[0022] ;

[0023] In the formula, represents the mutual information between the flight environment variable and the vibration signal ; represents the total number of data points selected from the time series data; represents the density ratio of the joint probability to the marginal probability; represents the value of the th data point of the vibration signal represents the value of the th data point of the flight environment variable represents the flight environment variable takes the value of and the vibration signal takes the value of when the joint probability; represents the marginal probability when the vibration signal takes the value of ; represents the marginal probability when the flight environment variable takes the value of ; represents the sign of the density ratio; represents the logarithmic function.

[0024] As a further solution of the present invention, the calculation formula of the density ratio is as follows:

[0025] ;

[0026] In the formula, represents the digamma function; represents the th times when the vibration signal represents the flight environment variable takes the value of times.

[0027] As a further solution of the present invention, the calculation formula of the weight is as follows:

[0028] ;

[0029] In the formula, represents the weight of the th vibration signal; represents the total number of vibration signals; represents the flight environment variable and the th vibration signal mutual information between them; represents the natural constant.

[0030] As a further solution of the present invention: the normalization operation is expressed as follows:

[0031] ;

[0032] In the formula, represents the value after normalization; represents the minimum value among all the values of the current vibration signal ; represents the maximum value among all the values of the current vibration signal ;

[0033] As a further solution of the present invention: the multi-channel fusion graph is expressed as follows:

[0034] ;

[0035] In the formula, represents the multi-channel fusion graph with the channel number of ; represents the multi-channel fusion operation; represents the first vibration signal ; represents the th vibration signal ; represents the th vibration signal ; represents the weight matrix composed of each weight.

[0036] As a further solution of the present invention: the grayscale matrix is expressed as follows:

[0037] ;

[0038] In the formula, represents the grayscale matrix of the current vibration signal ; represents the number of rows of the grayscale matrix; represents the number of columns of the grayscale matrix; Indicates the operation of generating a grayscale matrix;

[0039] As a further solution of the present invention: multiplying the weight of the vibration signal by each element in the grayscale matrix of the vibration signal can convert the grayscale matrix into a corresponding weighted matrix.

[0040] Compared with the prior art, the beneficial effects of the present invention are:

[0041] 1. The present invention first converts different one-dimensional data into grayscale matrices, which can better represent the spatial characteristics of the data. Subsequently, multiple grayscale image matrices of the same time scale are fused into a multi-channel fusion graph. This fusion method not only improves the data analysis efficiency but also reduces the risk of missing important information. The multi-modal information fusion strategy based on causal strength can effectively fuse different types of vibration signals while maintaining the spatial characteristics between the data. The deep learning model can obtain a more comprehensive and efficient feature representation, thus better adapting to the changes in different flight environments and improving the diagnosis and optimization effect of the aircraft performance.

[0042] 2. The method of the present invention quantifies the non-linear correlation between the vibration signal and the flight environment variable through mutual information, dynamically calculates the weight of each vibration signal and performs exponential transformation to achieve directional enhancement and noise reduction in the feature space; the element-by-element fusion of the weighted matrix and the grayscale matrix constructs a multi-channel fusion graph, breaking through the limitations of traditional linear superposition, highlighting the key vibration modes while retaining the time-frequency characteristics of the data; the real-time weight update mechanism enables the system to have an adaptive ability, can track the changes in the aircraft state, significantly improve the prediction accuracy of vibration reliability, and enhance the model interpretability through explicit weight distribution, providing an efficient solution for aircraft structural health monitoring.

[0043] 3. The mutual information weight calculation method realizes differential weighting of multi-source data by quantifying the correlation between the vibration signal and the flight environment variable. This mechanism can dynamically identify key vibration parameters, avoid interference from redundant information, and improve the data fusion efficiency. The weight allocation strategy based on mutual information provides a scientific basis for the subsequent construction of the multi-channel fusion graph, ensuring the effectiveness of model training.

[0044] 4. The mutual information calculation formula This formula uses a non-parametric density ratio estimation method, avoiding the model bias caused by traditional Gaussian assumptions. By introducing a sign function to process logarithmic operations, it can effectively suppress noise interference and improve the stability of mutual information calculation.

[0045] 5. Based on the density ratio estimation of the double gamma function, it can accurately estimate the ratio of the joint probability to the marginal probability under finite sample conditions, avoiding the parameter selection problem of traditional kernel density estimation and significantly reducing the computational complexity. It is especially suitable for the non-stationary characteristic analysis of aircraft vibration signals, improving the timeliness of mutual information calculation.

[0046] 6. The weight calculation formula performs an exponential transformation on the mutual information through the natural constant, which can effectively amplify the weight differences of key vibration signals. Combined with the normalization process, the weight distribution not only maintains the relative proportional relationship but also has absolute quantization significance.

[0047] 7. The method based on extreme value normalization can unify vibration signals with different dimensions into the interval [0, 255], eliminate the influence of dimensional differences on the analysis results, maintain the relative distribution characteristics of the data, provide a standardized input for the subsequent construction of the grayscale matrix, and ensure the consistency of multi-source data fusion.

[0048] 8. By multiplying the weight matrix and the vibration signal matrix element by element, the differential enhancement of data in different channels is realized. The multi-channel fusion operation can effectively integrate multi-dimensional vibration characteristics, form a composite feature map containing spatial distribution and weight information, provide richer input features for the deep learning model, and significantly improve the feature expression ability of the model.

[0049] 9. Convert the time-series vibration data into a two-dimensional grayscale matrix to realize the time-frequency domain visualization of vibration signals. By optimizing the settings of the matrix row and column parameters, the time-varying characteristics and frequency components of the vibration signals can be effectively retained. The construction of the grayscale matrix provides an intuitive data form for subsequent image processing and deep learning model training, improving the analysis efficiency.

[0050] 10. The element-by-element weighting strategy ensures that the importance of each vibration signal is accurately mapped to each element of the matrix. Through the linear combination of the weight matrix and the grayscale matrix, while maintaining the original data structure, the key vibration characteristics can be highlighted, providing an efficient data preprocessing method for the construction of the multi-channel fusion map and ensuring the effectiveness of model training. Description of the Drawings

[0051] Figure 1 It is a schematic structural diagram of the test mechanism in the present invention.

[0052] Figure 2 It is a flowchart of the test method in the present invention.

[0053] Figure 3 It is a distribution diagram of pixel values after the vibration acceleration is normalized in the present invention.

[0054] Figure 4 It is a distribution diagram of pixel values after the vibration displacement is normalized in the present invention.

[0055] Figure 5 It is a distribution diagram of pixel values after the vibration pressure is normalized in the present invention.

[0056] Figure 6This is the pixel value distribution diagram of the horizontal wind speed in the wind field environment after normalization in the present invention.

[0057] Figure 7 This is the gray matrix diagram of vibration acceleration, vibration displacement and vibration pressure corresponding to high vibration reliability in the present invention.

[0058] Figure 8 This is the gray matrix diagram of vibration acceleration, vibration displacement and vibration pressure corresponding to medium vibration reliability in the present invention.

[0059] Figure 9 This is the gray matrix diagram of vibration acceleration, vibration displacement and vibration pressure corresponding to low vibration reliability in the present invention.

[0060] Figure 10 This is the multi-channel fusion diagram corresponding to high vibration reliability, medium vibration reliability and low vibration reliability in the present invention.

[0061] In the figure: 1. Frame; 2. Fan; 3. Test space. Specific embodiments

[0062] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0063] Please refer to Figure 1 , the test mechanism of the present invention includes a hexahedron-shaped frame 1 formed by sequentially enclosing three rectangular frames. The frame 1 is hollow. Nine fans 2 are respectively installed on the six inner surfaces of the frame 1, and the nine fans 2 on the same surface are arranged in a rectangular array. The center of the rectangular array arrangement coincides with the center of the surface where it is located. The thirty-six fans 2 installed on the frame 1 are all connected to the control terminal, and each fan 2 can be independently controlled to operate. The aircraft conducts vibration reliability tests in the test space 3 formed inside the frame 1. At the same time, corresponding vibration sensors are provided at each key position of the aircraft to detect the corresponding vibration signals and flight environment variables, and transmit the corresponding information to the control terminal. The control terminal determines the vibration reliability corresponding to the current vibration signal and flight environment variables according to the corresponding vibration reliability evaluation criteria.

[0064] To simulate the distribution of different wind fields in a real environment, the layout and control of the fan 2 matrix are crucial. By precisely designing the layout and startup sequence of the fan 2, different wind field distribution patterns can be simulated. The layout and control of the fan 2 not only need to consider the overall direction of the air flow but also accurately control the intensity and distribution characteristics of the air flow to accurately reproduce a complex environment. Specifically, the direction control of the wind field involves two important angles: the pitch angle and the yaw angle. To achieve precise adjustment of the wind direction in the wind field, each rotating shaft of the fan 2 needs to have an independent control function. By dynamically adjusting the pitch and yaw angles, the accuracy and stability of the wind direction control are ensured.

[0065] In addition, the control of the wind speed is another important factor. In wind field simulation, the change in wind speed directly affects the intensity and distribution of the air flow. To achieve precise control of the wind speed, the control terminal changes the intensity of the air flow by adjusting the rotation speed of the fan 2. By precisely controlling the rotation speed of the fan 2, different wind speed changes can be simulated in different regions and under different wind directions, thus providing diverse wind field simulation effects.

[0066] To achieve these complex control objectives, the PID control algorithm is adopted. The PID control algorithm continuously adjusts the output of the motor by calculating the error between the target angle and the current angle in real time, thereby achieving precise control of the angle of the fan 2. This feedback mechanism can not only ensure that the fan 2 quickly reaches the set target angle but also effectively maintain stable operation in practical applications, thus ensuring the accuracy of the simulated wind field in all directions.

[0067] In the vibration reliability test of a vertical takeoff and landing aircraft, the vibration acceleration, vibration displacement, and vibration pressure in the vibration signal are three core signals, respectively reflecting the vibration characteristics of different parts of the aircraft. The flight environment variables include wind speed, air volume, etc., respectively reflecting the specific state of the wind field environment in the test space. Table 1 shows the specific functions, processing methods, and collaborative values of the vibration signal in the test process.

[0068] Table 1 Vibration Signal Types

[0069]

[0070] Please refer to Figure 2 , the test method of the present invention includes the following test steps:

[0071] 1. Obtain time series data

[0072] Obtain the timing data of the aircraft during testing in the test space. The timing data includes vibration signals, flight environment variables, and the corresponding vibration reliability. Obtain the vibration signals, flight environment variables, and the corresponding vibration reliability that change with time under different test environments of the aircraft in the test space, that is, obtain the timing data of the aircraft during flight for subsequent analysis.

[0073] II. Obtain the grayscale matrix

[0074] Convert each piece of timing data into the same type of normalized data through normalization operations, and organize each piece of normalized data into a corresponding grayscale matrix.

[0075] The normalization operation is expressed as follows:

[0076] ;

[0077] In the formula, represents the value after normalization; represents the minimum value among all the values of the current vibration signal ; represents the maximum value among all the values of the current vibration signal .

[0078] The necessity of normalization eliminates the dimensional difference. Data from different sensors (such as vibration acceleration, vibration displacement) have different physical units and magnitudes. Direct fusion will cause signals with large values to dominate the image features. After normalization, all data are presented as gray-scale changes with the same weight in the image.

[0079] As Figures 3 to 5 shown, the pixel values of each timing point of vibration acceleration, vibration displacement, and vibration pressure obtained by normalizing the vibration signal samples with high vibration reliability are presented, and each vibration signal sample contains a total of data points. As Figure 6 shown, the pixel values of each timing point obtained by normalizing the horizontal wind speed in the wind field environment are presented, and a total of data points are included.

[0080] Through normalization, the original data of vibration acceleration, vibration displacement, vibration pressure, and horizontal wind speed in the wind field environment are uniformly mapped to the color intensity range of 0 - 255 for subsequent conversion into a grayscale matrix for multimodal fusion.

[0081] Perform normalization operations on all the obtained vibration signal samples, and then construct the corresponding grayscale matrices, and the dimension of each grayscale matrix is . As Figures 7 to 9As shown, there are grayscale matrices corresponding to three sets of vibration signal samples, namely the grayscale matrices of vibration acceleration, vibration displacement, and vibration pressure corresponding to high vibration reliability, medium vibration reliability, and low vibration reliability, respectively.

[0082] III. Multi-channel fusion diagram

[0083] Calculate the weights of each vibration signal, and use these weights to weight the grayscale matrix to obtain a weighted matrix. Then, fuse the weighted matrices in sequence to obtain a multi-channel fusion diagram. The specific steps are as follows:

[0084] First, calculate the mutual information between the flight environment variables and each vibration signal. The calculation formula for mutual information is as follows:

[0085] ;

[0086] ;

[0087] In the formula, represents the mutual information between the flight environment variable and the vibration signal ; represents the total number of data points selected from the time series data; represents the density ratio of the joint probability to the marginal probability; represents the value of the th data point of the vibration signal ; represents the value of the th data point of the vibration signal ; represents the joint probability when the flight environment variable takes the value and the vibration signal takes the value ; represents the marginal probability when the vibration signal takes the value ; represents the marginal probability when the flight environment variable takes the value ; represents the sign of the density ratio; represents the logarithmic function. represents the digamma function; represents the number of times the vibration signal takes the value ; represents the number of times the flight environment variable takes the value ;

[0088] Taking the vibration signal data with high vibration reliability as an example, based on the above formula, we can obtain:

[0089] ;

[0090] ;

[0091] ;

[0092] Among them, represents the vibration acceleration, represents the vibration displacement, represents the vibration pressure, represents the horizontal wind speed of the wind field environment.

[0093] Then, based on the mutual information, calculate the weights of each vibration signal. The calculation formula for the weights is as follows:

[0094] ;

[0095] In the formula, represents the weight of the th vibration signal ; represents the total number of vibration signals; represents the flight environment variable and the th vibration signal between the mutual information; represents the natural constant.

[0096] Following the calculation results of the above mutual information, the calculation results of the weights are as follows: the weight of vibration acceleration is 0.28499471; the weight of vibration displacement is 0.70891309; the weight of vibration pressure is 0.00609219.

[0097] Then multiply the weight of the vibration signal by each element in the gray matrix of the vibration signal, and the gray matrix can be transformed into the corresponding weighted matrix.

[0098] Finally, take each weighted matrix as a channel and perform fusion in sequence to fuse and obtain a multi-channel fusion graph corresponding to the number of channels. The multi-channel fusion graph is represented as follows:

[0099] ;

[0100] In the formula, represents the multi-channel fusion graph with the number of channels being ; represents the multi-channel fusion operation; represents the first vibration signal ; Indicates the th vibration signal ; Indicates the th vibration signal ; Indicates the weight matrix composed of each weight.

[0101] As Figure 10 shown, they are respectively the multi-channel fusion diagrams of high vibration reliability, medium vibration reliability, and low vibration reliability obtained.

[0102] IV. Obtaining the vibration reliability prediction model

[0103] Use the multi-channel fusion diagram to predict the deep learning model to generate a vibration reliability prediction model with vibration signals as features and vibration reliability as labels.

[0104] V. Real-time vibration reliability prediction

[0105] Obtain the real-time vibration signal of the aircraft and process it according to the content of steps S2 - S4 to predict the real-time vibration reliability of the aircraft.

[0106] By applying the method in the present invention to various models, the gap between the accuracy of the model using this method and the accuracy of the model not using this method can be obtained, as shown in Table 2 specifically.

[0107] Table 2 Comparison of prediction accuracies

[0108]

[0109] Based on the data in Table 2, it can be known that:

[0110] Autoencoder: The accuracy rate is only 19.50% without using the method, and it is increased to 34.75% after using it.

[0111] Convolutional neural network: The accuracy rate without using the method is 40.05%, and it jumps to 99.75% after using it, with a significant improvement.

[0112] AlexNet network: It is 95.75% without using the method, and it is further increased to 98.50% after using it.

[0113] Other models (such as bidirectional long short-term memory network, multi-layer perceptron, etc.) all show a similar pattern:

[0114] The accuracy rate after using this method is higher than the result without using this method.

[0115] It can be seen that in a variable operating condition environment, the accuracy of the model using this method is generally improved compared to the model without using this method, verifying the positive effect of this method on optimizing the model training effect and proving that it can effectively enhance the performance of the model under complex operating conditions.

[0116] The above are only the preferred specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, making equivalent substitutions or changes should be covered within the protection scope of the present invention.

Claims

1. A vibration reliability test method for a vertical takeoff and landing aircraft, characterized in that, The following test steps are included: S1. Obtain the timing data of the aircraft during flight tests. The timing data includes vibration signals, flight environment variables, and the corresponding vibration reliabilities; S2. Convert each piece of timing data into the same type of data, and organize the converted data into corresponding grayscale matrices; S3. Calculate the weights of each piece of timing data according to the mutual information between the flight environment variables and each vibration signal. Weight the grayscale matrices by these weights to obtain weighted matrices, and fuse the weighted matrices in sequence to obtain a multi-channel fusion graph; S4. Use the multi-channel fusion graph to train a deep learning model to generate a vibration reliability prediction model with vibration signals as features and vibration reliabilities as labels; S5. Obtain the real-time timing data of the aircraft and input it into the vibration reliability prediction model to predict the real-time vibration reliability of the aircraft.

2. The vibration reliability test method for a vertical takeoff and landing aircraft according to claim 1, characterized in that The calculation formula for mutual information is as follows: ； In the formula, represents the mutual information between the flight environment variables and the vibration signal; represents the total number of data points selected from the time series data; represents the density ratio of the joint probability to the marginal probability; represents the th data point value of the vibration signal represents the th data point value of the flight environment variable represents the flight environment variable taking the value of and the vibration signal taking the value of at this time as the joint probability; represents the marginal probability when the vibration signal takes the value of ; represents the marginal probability when the flight environment variable takes the value of ; represents the sign of the density ratio; represents the logarithmic function.

3. A vertical takeoff and landing aircraft vibration reliability test method according to claim 2, characterized in that The calculation formula for the density ratio is as follows: ； In the formula, represents the digamma function; represents the vibration signal takes values of times; represents the flight environment variable takes values of times.

4. A vertical takeoff and landing aircraft vibration reliability test method according to claim 3, characterized in that The calculation formula for the weight is as follows: ； In the formula, represents the weight of the nth vibration signal; represents the total number of vibration signals; represents the mutual information between the flight environment variable and the nth vibration signal ; represents the natural constant.

5. A vertical take-off and landing aircraft vibration reliability test method according to claim 4, characterized in that, Convert each piece of timing data into the same type of data through a normalization operation, and the normalization operation is expressed as follows: ； In the formula, represents the value after normalization; represents the minimum value among all the values of the current vibration signal ; represents the maximum value among all the values of the current vibration signal .

6. A vertical take-off and landing aircraft vibration reliability test method according to claim 5, characterized in that The multi-channel fusion graph is expressed as follows: ； In the formula, represents a multi-channel fusion graph with channels; represents a multi-channel fusion operation; represents the first vibration signal ; represents the th vibration signal ; represents the th vibration signal ; represents a weight matrix composed of respective weights.

7. A vertical take-off and landing aircraft vibration reliability test method according to claim 6, characterized in that, The grayscale matrix is expressed as follows: ； In the formula, represents the current vibration signal gray matrix; represents the number of rows of the gray matrix; represents the number of columns of the gray matrix; represents the operation of generating the gray matrix.

8. A vertical takeoff and landing aircraft vibration reliability test method according to claim 7, characterized in that Multiply the weight of the vibration signal by each element in the grayscale matrix of the vibration signal, and the grayscale matrix can be converted into the corresponding weighted matrix.

9. A vibration reliability test mechanism for a vertical takeoff and landing aircraft, which is applied to a vibration reliability test method for a vertical takeoff and landing aircraft as described in claim 8, characterized in that, It includes a hexahedron-shaped frame (1), and a plurality of independently operable fans (2) are installed on each of the six faces of the frame. The fans (2) on the six faces enclose a test space (3) for testing the vibration reliability of a vertical takeoff and landing aircraft.

Images