Object material detection method based on non-contact ultrasonic waves

Abstract

The invention relates to an object material detection method based on non-contact ultrasonic waves. The technical problems that an existing detection method is low in sensitivity and reliability and low in signal-to-noise ratio are solved. The invention provides an object material detection method based on non-contact ultrasonic waves. The object material detection method comprises the following steps: S1, acquiring ultrasonic echo signals of an object to be detected; s2, performing frequency domain filtering on the signal; s3, performing time domain filtering on the filtered ultrasonic echo signal by adopting empirical mode decomposition to obtain an intrinsic mode function of the signal, and selecting a main intrinsic mode function; s4, performing information entropy calculation on the selected main intrinsic mode function; and S5, establishing an ultrasonic echo signal information entropy template of objects made of different materials, and identifying the materials of the objects by using the calculated information entropy.

Description

Technical Field

[0001] The present invention relates to a non-destructive testing method for the material of an object, and particularly to a method for detecting the material of an object based on non-contact ultrasonic waves. Background Art

[0002] Most traditional methods for detecting the material of an object are contact detection methods, which usually rely on physical contact or contact sensors to obtain the state information of the object. In the detection process, they need to contact the object to be detected, so they may cause damage or wear to the surface of the object. Especially when detecting precision objects or vulnerable items, it often has an adverse impact on the object to be detected. In the prior art, to avoid the technical problems brought by traditional contact detection, non-contact detection methods are usually used to detect the material of an object.

[0003] The non-contact ultrasonic detection method is widely used because of its advantages such as simple operation, fast detection speed, no damage to the object to be detected, and low price. The non-contact ultrasonic detection method is a non-destructive detection method that uses ultrasonic waves to detect the material of an object. It emits ultrasonic waves to the surface of the object through a transmitting probe. When ultrasonic waves vertically enter from one medium into another medium, the changes in sound pressure and propagation direction have certain rules. By using this rule combined with specific algorithms, the material characteristics of the object to be detected can be detected without causing any damage to it.

[0004] When ultrasonic waves propagate in a medium, the medium has an obstructive effect on ultrasonic waves, so there will be acoustic impedance. If the acoustic impedances of two media are quite different, strong reflection and refraction phenomena of ultrasonic waves will occur. Acoustic impedance is related to the density of the material and the propagation speed of ultrasonic waves in the material. Since the acoustic impedances of different materials are different, the degrees of reflection and transmission of ultrasonic waves by different materials are also different. This characteristic enables ultrasonic echoes to be used to detect the material of an object. However, the ultrasonic echo signal is prone to contain noise signals, and these noise signals affect the feature extraction of the signal to a certain extent. How to efficiently extract and identify the effective signal from the signal containing noise, so as to improve the signal-to-noise ratio of the signal, and further improve the sensitivity and reliability of detection has always been an urgent problem to be solved. Summary of the Invention

[0005] The purpose of the present invention is to solve the technical problem that the existing ultrasonic detection method is prone to contain noise signals, resulting in low sensitivity and reliability, and to provide a method for detecting the material of an object based on non-contact ultrasonic waves.

[0006] To achieve the above purpose, the technical solution provided by the present invention is as follows:

[0007] A method for detecting the material of an object based on non-contact ultrasonic waves, characterized in that it includes the following steps:

[0008] S1. Obtain the ultrasonic echo signal of the object to be detected through a non-contact ultrasonic detection system;

[0009] S2. Perform frequency-domain filtering on the ultrasonic echo signal obtained in S1 to obtain the filtered ultrasonic echo signal;

[0010] S3. Perform time-domain filtering on the filtered ultrasonic echo signal obtained in S2 using empirical mode decomposition to obtain an ensemble of intrinsic mode functions;

[0011] S4. Extract the main intrinsic mode functions from the ensemble of intrinsic mode functions obtained in S3;

[0012] S5. Calculate the information entropy value of the main intrinsic mode functions obtained in S4 to obtain the information entropy value corresponding to the main intrinsic mode functions;

[0013] S6. According to the type of the object to be detected, empirically determine the different materials it may involve, and establish an information entropy value template for the ultrasonic echo signals of the different materials involved;

[0014] S7. Determine the material of the object to be detected based on the information entropy value obtained in S5 in combination with the information entropy value template established in S6.

[0015] Further, the specific process of step S3 is as follows:

[0016] S3.1. For the filtered ultrasonic echo signal x(t) obtained in S2, respectively determine the local maximum and local minimum from the time domain dimension, and construct a local maximum sequence and a local minimum sequence according to the local maximum and local minimum respectively;

[0017] S3.2. For the local maximum sequence and local minimum sequence obtained in S3.1, respectively construct the upper envelope e max (t) and the lower envelope e min (t) through cubic spline curves;

[0018] S3.3. Calculate the average value m(t) of the upper envelope e max (t) and the lower envelope e min (t):

[0019] m(t) = (e min (t) + e max (t)) / 2;

[0020] S3.4. Calculate the residual y(t):

[0021] y(t) = x(t) - m(t);

[0022] S3.5. Define the intrinsic mode function imf(t) = y(t), and construct a set of intrinsic mode functions based on the intrinsic mode function imf(t);

[0023] S3.6. Determine whether the number of extreme values of the intrinsic mode function imf(t) meets a preset value, or determine whether the average value m(t) of the upper envelope e max (t) and the lower envelope e min (t) is 0; if so, complete the construction of the set of intrinsic mode functions; if not, execute S3.7;

[0024] S3.7. Return to S3.1, and use the y(t) obtained in S3.4 as the new filtered ultrasonic echo signal x(t) until the construction of the set of intrinsic mode functions is completed.

[0025] Further, the specific process of step S4 is as follows:

[0026] From the set of intrinsic mode functions obtained in S3, select the intrinsic mode functions of the high-frequency components as the main intrinsic mode functions to be extracted.

[0027] Further, in step S5, the formula for calculating the information entropy value of the main intrinsic mode functions obtained in S4 is as follows:

[0028]

[0029] In the formula, H is the information entropy value, P(x i ) is the probability of the random event x i occurring, k is the number of data points, and i is the data point.

[0030] Further, in step S6, the process of establishing the information entropy value templates of the ultrasonic echo signals of different materials involved is as follows:

[0031] S6.1. Obtain the ultrasonic echo signals of different materials through a non-contact ultrasonic detection system respectively;

[0032] S6.2. Perform frequency-domain filtering on the ultrasonic echo signals of different materials obtained in S6.1 respectively to obtain the filtered ultrasonic echo signals of different materials;

[0033] S6.3. Perform time-domain filtering on the filtered ultrasonic echo signals of different materials obtained in S6.2 respectively by using empirical mode decomposition to obtain the sets of intrinsic mode functions of different materials;

[0034] S6.4. Extract the main intrinsic mode functions of different materials from the sets of intrinsic mode functions of different materials obtained in S6.3 respectively;

[0035] S6.5. Calculate the information entropy values for the main intrinsic mode functions of different materials obtained in S6.4 to obtain the information entropy values corresponding to different materials;

[0036] S6.6. According to the information entropy values corresponding to different materials obtained in S6.5, obtain the probability density curve functions of different materials by means of data fitting. The calculation formula is as follows:

[0037]

[0038] In the formula, σ 2 is the variance of the information entropy value of each material, and H avg is the average information entropy value;

[0039] S6.7. According to the probability density curve functions of different materials obtained in S6.6, establish the information entropy value templates of the ultrasonic echo signals of the different materials involved.

[0040] Further, in step S7, the method for determining the material of the object to be detected is as follows:

[0041] Determine the material of the object to be detected by judging the area of the information entropy value template into which the information entropy value of the object to be detected falls.

[0042] Further, in step S3.6, specifically judge whether the number of extreme values of the intrinsic mode function imf(t) meets the preset value:

[0043] When the number of extreme values of the intrinsic mode function imf(t) is less than or equal to 1, if so, it is considered that the number of extreme values of the intrinsic mode function imf(t) meets the preset value; if not, it is considered that the number of extreme values of the intrinsic mode function imf(t) does not meet the preset value.

[0044] Further, in step S2, perform frequency domain filtering on the ultrasonic echo signal obtained in S1 using a high-pass filter and a low-pass filter.

[0045] Advantages of the present invention:

[0046] 1. A method for detecting the material of an object based on non-contact ultrasonic waves according to the present invention filters the original ultrasonic echo signal in the frequency domain by using a band-pass filter and a band-stop filter to remove noise and retain effective frequency signals, and then performs time domain filtering by using the empirical mode decomposition method to further reduce the interference of low-frequency noise and improve the signal-to-noise ratio. This method can effectively extract the signal features reflecting the material characteristics and solves the technical problems of the low signal-to-noise ratio, detection sensitivity and reliability of the existing signals;

[0047] 2. The object material detection method based on non-contact ultrasonic waves in the present invention establishes characteristic templates for objects of different materials, simplifying the process of object material identification; calculates the information entropy value of the signal using the intrinsic mode functions obtained by empirical mode decomposition, and uses this as the characteristic value of the material. By measuring objects of different materials multiple times to obtain their information entropy values, the corresponding probability density curve function is obtained using the data fitting method. In this way, the material type can be detected using a single information entropy value, simplifying the detection process. Description of the Drawings

[0048] Figure 1 Schematic diagram of the non-contact ultrasonic detection system used in the embodiment of the object material detection method based on non-contact ultrasonic waves in the present invention;

[0049] Figure 2 Flowchart of the object material detection method based on non-contact ultrasonic waves in the present invention;

[0050] Figure 3 Ultrasonic echo signal diagram in step S1 of the embodiment of the object material detection method based on non-contact ultrasonic waves in the present invention;

[0051] Figure 4 Filtered ultrasonic echo signal diagram in step S2 of the embodiment of the object material detection method based on non-contact ultrasonic waves in the present invention;

[0052] Figure 5 Schematic diagram of the set of intrinsic mode functions obtained using empirical mode decomposition in step S3 of the embodiment of the object material detection method based on non-contact ultrasonic waves in the present invention;

[0053] Figure 6 Information entropy value template of the ultrasonic echo signals of objects of different materials established in step S6 of the embodiment of the object material detection method based on non-contact ultrasonic waves in the present invention.

[0054] Description of the reference numerals: 1 - the plate to be measured, 2 - ultrasonic probe, 3 - ultrasonic signal generator, 4 - oscilloscope, 5 - computer. Detailed Embodiment

[0055] The present invention will be further described below with reference to the drawings and embodiments.

[0056] The object material detection method based on non-contact ultrasonic waves in this embodiment, as Figure 2 shown, the specific implementation steps are as follows:

[0057] Step S1: Obtain the ultrasonic echo signal of the object to be detected through a non-contact ultrasonic detection system;

[0058] AsFigure 1 As shown in the figure, the non-contact ultrasonic detection system includes an ultrasonic probe 2, an ultrasonic signal generator 3, an oscilloscope 4, and a computer 5. The signal transmission ends of the ultrasonic probe 2 and the ultrasonic signal generator 3 are connected for two-way communication, and are used to generate ultrasonic waves with corresponding frequencies under the excitation of the ultrasonic signal generator 3. The detection end of the ultrasonic probe 2 is used to send ultrasonic waves to the object 1 to be detected, receive the ultrasonic signal reflected by the detected object 1, and at the same time send the received reflected ultrasonic signal to the ultrasonic signal generator 3. The signal sending end of the ultrasonic signal generator 3 is connected to the signal receiving end of the oscilloscope 4, and is used to send the received reflected ultrasonic signal to the oscilloscope 4. The signal sending end of the oscilloscope 4 is connected to the signal receiving end of the computer 5, and is used to send the received reflected ultrasonic signal to the computer 5. The computer 5 collects and analyzes the received reflected ultrasonic signal; as Figure 3 shown is the original ultrasonic echo signal of the object to be detected received;

[0059] Step S2: Perform frequency-domain filtering on the original ultrasonic echo signal to obtain the filtered ultrasonic echo signal. The original ultrasonic echo signal contains low-frequency noise and high-frequency noise signals, and its noise components are filtered out by using a high-pass filter and a low-pass filter. As Figure 4 shown is the filtered ultrasonic echo signal;

[0060] Step S3: Perform time-domain filtering on the filtered ultrasonic echo signal by using empirical mode decomposition to obtain an ensemble of intrinsic mode functions;

[0061] S3.1: Respectively determine the local maximum and local minimum of the filtered ultrasonic echo signal x(t) from the time domain dimension, and construct a local maximum sequence and a local minimum sequence according to the local maximum and local minimum respectively;

[0062] S3.2: For the local maximum sequence and the local minimum sequence obtained in S3.1, respectively construct an upper envelope e max (t) and a lower envelope e min (t) through a cubic spline curve;

[0063] S3.3: Calculate the average value m(t) of the upper envelope e max (t) and the lower envelope e min (t):

[0064] m(t) = (e min (t) + e max (t)) / 2;

[0065] S3.4: Calculate the first residual y1(t):

[0066] y1(t) = x(t) - m(t);

[0067] S3.5. Define the first intrinsic mode function as imf1(t) = y1(t), and construct an intrinsic mode function set based on the first intrinsic mode function imf1(t);

[0068] S3.6. Take y1(t) as the new filtered ultrasonic echo signal x(t), return to S3.1, and repeat the calculation to obtain the second intrinsic mode function imf2(t), the third intrinsic mode function imf3(t),..., when the k-th intrinsic mode function imf k (t) satisfies that the difference between the number of extrema and 0 does not exceed 1, or when the average value m(t) of the upper envelope e max (t) and the lower envelope e min (t) is 0, the calculation terminates. As Figure 5 shown, obtain the final intrinsic mode function set [imf1(t), imf2(t), imf3(t),... imf k (t)];

[0069] Step S4. Extract the main intrinsic mode functions from the intrinsic mode function set [imf1(t), imf2(t), imf3(t),... imf k (t)];

[0070] Each obtained imf k (t) corresponds to a vibration mode in a specific frequency range of the signal, where k = 1 represents the high-frequency component in the signal and k = n represents the low-frequency component. Compared with other imf k (t), k = 2,..., n, the corresponding signal component of imf1(t) has a greater energy spectral density. In addition, the acoustic characteristics of different materials are mainly reflected in the high-frequency signals. Therefore, use imf1(t) to extract features;

[0071] Step S5. Calculate the information entropy value of the main intrinsic mode function to obtain the information entropy value corresponding to the main intrinsic mode function; The calculation formula for the information entropy value corresponding to the intrinsic mode function is as follows:

[0072]

[0073] In the formula, H is the information entropy value, P(x i ) is the probability of the random event x i occurring, k is the number of data points, and i is the data point;

[0074] Step S6. Establish an information entropy value template for the ultrasonic echo signals of objects with different materials, and determine the material of the object to be detected according to the information entropy value obtained in S5 in combination with the information entropy value template;

[0075] S6.1. Obtain the ultrasonic echo signals of objects made of different materials respectively through a non-contact ultrasonic detection system;

[0076] S6.2. Filter the ultrasonic echo signals of objects made of different materials respectively to obtain the filtered ultrasonic echo signals of objects made of different materials;

[0077] S6.3. Perform empirical mode decomposition on the filtered ultrasonic echo signals of objects made of different materials respectively to obtain the intrinsic mode function sets of objects made of different materials;

[0078] S6.4. Extract the main intrinsic mode functions of objects made of different materials respectively from the intrinsic mode function sets of objects made of different materials;

[0079] S6.5. Calculate the information entropy values for the main intrinsic mode functions of objects made of different materials respectively to obtain the corresponding information entropy values of objects made of different materials;

[0080] S6.6. According to the information entropy values corresponding to objects made of different materials, obtain the probability density curve function of objects made of different materials by means of data fitting, and its calculation formula is as follows:

[0081]

[0082] In the formula, σ 2 is the variance of the information entropy value of each material, and H avg is the average information entropy value;

[0083] S6.7. Establish an information entropy value template according to the probability density curve function of objects made of different materials. In this embodiment, as Figure 6 shown is the information entropy value template established based on iron plates, aluminum plates, wooden boards, cardboard and foam boards; The present invention is applicable to the identification of a variety of materials, and has tested a variety of materials including metals and non-metals, indicating that this method has good universality and can be applicable to the identification of different types of materials; It enriches the application fields of ultrasonic detection. Traditional non-contact ultrasonic detection methods are limited in some fields (such as wood products, porous hydrophobic materials, etc.), while this method can effectively solve these problems and expand the application scenarios of ultrasonic detection;

[0084] S7. By using this statistical method, determine the material of the object to be detected through the area of the information entropy value template into which the information entropy value of the object to be detected falls.

[0085] When performing step S7, if a certain material of the object to be detected is not found in the information entropy value template, supplementary detection should be carried out on the different materials that may be involved in step S6, and the information entropy value template should be updated.

[0086] The above are only the preferred specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present application should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for detecting the material of an object based on non-contact ultrasonic waves, characterized in that, It includes the following steps: S1. Obtain the ultrasonic echo signal of the object to be detected through a non-contact ultrasonic detection system; S2. Perform frequency-domain filtering on the ultrasonic echo signal obtained in S1 to obtain the filtered ultrasonic echo signal; S3. Perform time-domain filtering on the filtered ultrasonic echo signal obtained in S2 using empirical mode decomposition to obtain an intrinsic mode function set; S4. Extract the main intrinsic mode functions from the intrinsic mode function set obtained in S3; S5. Calculate the information entropy value of the main intrinsic mode functions obtained in S4 to obtain the information entropy value corresponding to the main intrinsic mode functions; S6. According to the type of the object to be detected, empirically determine the different materials it may involve, and establish an information entropy value template for the ultrasonic echo signals of the different materials involved; S7. Determine the material of the object to be detected according to the information entropy value obtained in S5 and in combination with the information entropy value template established in S6.

2. The method for detecting the material of an object based on non-contact ultrasonic waves according to claim 1, wherein The specific process of step S3 is as follows: S3.

1. For the filtered ultrasonic echo signal x(t) obtained in S2, respectively determine the local maximum and local minimum from the time domain dimension, and construct a local maximum sequence and a local minimum sequence according to the local maximum and local minimum respectively; S3.

2. Construct the upper envelope e max (t) and the lower envelope e min (t) respectively for the local maximum value sequence and the local minimum value sequence obtained in S3.1 through cubic spline curves; S3.

3. Calculate the average value m(t) of the upper envelope e max (t) and the lower envelope e min (t): m(t) = (e min (t) + e max (t)) / 2; S3.

4. Calculate the residual y(t): y(t) = x(t) - m(t); S3.

5. Define the intrinsic mode function imf(t) = y(t), and construct an intrinsic mode function set according to the intrinsic mode function imf(t); S3.

6. Determine whether the number of extreme values of the intrinsic mode function imf(t) meets a preset value, or determine whether the average value m(t) of the upper envelope e max (t) and the lower envelope e min (t) is 0; if so, complete the construction of the set of intrinsic mode functions, if not, execute S3.7; S3.

7. Return to S3.1, and use the y(t) obtained in S3.4 as the new filtered ultrasonic echo signal x(t) until the construction of the intrinsic mode function set is completed.

3. The method for detecting the material of an object based on non-contact ultrasonic waves according to claim 2, characterized in that, The specific process of step S4 is as follows: From the intrinsic mode function set obtained in S3, select the intrinsic mode functions of the high-frequency components as the extracted main intrinsic mode functions.

4. The object material detection method based on non-contact ultrasonic according to claim 3, characterized in that, In step S5, the formula for calculating the information entropy value of the main intrinsic mode functions obtained in S4 is as follows: where H is the information entropy value, P(x i ) is the probability of the random event x i occurring, k is the number of data points, and i is the data point.

5. The object material detection method based on non-contact ultrasonic according to claim 4, characterized in that, In step S6, the process of establishing the information entropy value template for the ultrasonic echo signals of the different materials involved is as follows: S6.

1. Respectively obtain the ultrasonic echo signals of different materials through a non-contact ultrasonic detection system; S6.

2. Perform frequency-domain filtering on the ultrasonic echo signals of different materials obtained in S6.1 to obtain the filtered ultrasonic echo signals of different materials; S6.

3. Perform time-domain filtering on the filtered ultrasonic echo signals of different materials obtained in S6.2 using empirical mode decomposition to obtain the intrinsic mode function sets of different materials; S6.

4. Respectively extract the main intrinsic mode functions of different materials from the intrinsic mode function sets of different materials obtained in S6.3; S6.

5. Calculate the information entropy values of the main intrinsic mode functions of different materials obtained in S6.4 to obtain the information entropy values corresponding to different materials; S6.

6. According to the information entropy values corresponding to different materials obtained in S6.5, obtain the probability density curve function of different materials by means of data fitting, and its calculation formula is as follows: where, σ 2 is the variance of the information entropy value of each material, and H avg is the average information entropy value; S6.

7. According to the probability density curve functions of different materials obtained in S6.6, establish an information entropy value template for the ultrasonic echo signals of the different materials involved.

6. The method for detecting the material of an object based on non-contact ultrasonic waves according to claim 5, wherein In step S7, the method for determining the material of the object to be detected is as follows: By judging the region of the information entropy value template into which the information entropy value of the object to be detected falls, the material of the object to be detected is determined.

7. A method for detecting the material of an object based on non-contact ultrasonic waves according to claim 6, characterized in that, In step S3.6, specifically, it is judged whether the number of extreme values of the intrinsic mode function imf(t) meets the preset value: When the number of extreme values of the intrinsic mode function imf(t) is less than or equal to 1, if so, it is considered that the number of extreme values of the intrinsic mode function imf(t) meets the preset value; if not, it is considered that the number of extreme values of the intrinsic mode function imf(t) does not meet the preset value.

8. The method for detecting the material of an object based on non-contact ultrasonic waves according to claim 7, wherein: In step S2, the ultrasonic echo signal obtained in S1 is subjected to frequency-domain filtering using a high-pass filter and a low-pass filter.

Images