High-precision material defect detection method based on nano sensor

Abstract

The invention provides a high-precision material defect detection method based on a nano sensor, and relates to the technical field of material defect detection.The method comprises the steps that S1, the nano sensor is provided and detects tiny defects on the surface or inside of a material, S2, the nano sensor is installed on the surface or inside of the material to be detected, and S3, the material to be detected is detected; and the material is monitored in real time through a sensor, and signals of potential defects in the material are obtained. According to the high-precision material defect detection method based on the nano sensor, the nano sensor and an advanced artificial intelligence algorithm are combined, so that high-precision identification of tiny defects in a material is realized. Defect type classification is carried out by using a random forest algorithm, and the image data and the signal feature data are subjected to conjoint analysis, so that accurate classification and positioning of multiple defect types are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of material defect detection, and in particular to a high-precision material defect detection method based on nanosensors. Background Art

[0002] With the continuous advancement of industrial technology, the quality requirements for materials in the aerospace, automotive, and high-tech sectors are becoming increasingly stringent. Material performance and quality play a crucial role in overall product safety and reliability. Therefore, material defect detection, as a key component of quality control, has attracted widespread attention. While traditional material defect detection methods, such as ultrasonic testing, X-ray testing, and infrared thermal imaging, can detect macroscopic surface defects to a certain extent, they still have many shortcomings in terms of accuracy, sensitivity, and real-time performance.

[0003] When defects are extremely small or their distribution is extremely complex, traditional detection methods often fail to detect tiny cracks, holes, and microstructural defects within materials. With the rapid development of nanotechnology, nanosensors, due to their high sensitivity, fast response speed, and small size, have gradually become a research hotspot for the next generation of material defect detection. Nanosensors can monitor material microstructures and subtle changes in real time, providing more accurate and detailed detection results. However, current nanosensor-based material defect detection methods still face the following challenges: how to improve the stability and reliability of sensors to ensure they are not affected by environmental changes or external factors during long-term use; and how to accurately identify various types of defects in complex material environments and conduct effective analysis and judgment. Summary of the Invention

[0004] In response to the shortcomings of the existing technology, the present invention provides a high-precision material defect detection method based on nanosensors. The technical problem to be solved by this invention is: how to achieve efficient and accurate defect identification and report generation in complex material environments through signal enhancement, defect classification, positioning and quantitative analysis.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: A high-precision material defect detection method based on nanosensors, comprising:

[0006] S1. Providing a nanosensor that detects tiny defects on the surface or inside a material;

[0007] S2. The nanosensor is mounted on the surface or inside of the material to be detected, and the material is monitored in real time by the sensor to obtain signals of potential defects in the material;

[0008] S3. The signal obtained by the sensor is processed by the data processing unit, and a signal enhancement algorithm is used to optimize the weak signal to improve the accuracy of identifying small defects and reduce interference from environmental noise;

[0009] S4. Analyze the defect type and location based on the signal processing results, and use artificial intelligence algorithms to automatically classify and quantitatively analyze defects, thereby accurately identifying different types of defects in the material and outputting a defect report;

[0010] S5. During the real-time monitoring process, the working status of the sensor is monitored and adjusted through the feedback control system to ensure its stability in long-term use and adapt to changes in the material environment.

[0011] Preferably, the nanosensor is a thin film sensor, which is used to capture signals of tiny defects and has a high surface area and response speed.

[0012] Preferably, the signal enhancement algorithm includes a multi-level signal filtering module, and the filtering module removes noise based on wavelet transform to improve the signal-to-noise ratio of the signal.

[0013] Preferably, S4 specifically includes:

[0014] S4.1 uses the support vector machine random forest algorithm to classify the types of material defects and conduct quantitative analysis based on the size and depth characteristics of the defects;

[0015] S4.2 adopts a deep learning-based target detection model, using image data and signal feature data to automatically classify material defects to achieve high-precision and high-efficiency defect identification.

[0016] Preferably, the defect position is located by combining ultrasonic imaging with nanosensor data to improve positioning accuracy in complex material environments.

[0017] Preferably, the defect report includes the defect type, location, size, severity, and impact on material properties.

[0018] Preferably, the specific steps of S5 include:

[0019] S5.1 monitors the working status of the sensor in real time and checks whether the sensor's performance meets the requirements by comparing the current working status with the preset standard value;

[0020] S5.2 dynamically adjusts the sensor's sensitivity and sampling frequency through an adaptive control algorithm to respond to changes in the material environment and maintain the sensor's optimal detection performance.

[0021] Preferably, the feedback control system adjusts the operating parameters of the sensor according to changes in temperature, humidity, pressure and vibration to ensure efficient operation of the sensor under different environmental conditions.

[0022] The present invention provides a high-precision material defect detection method based on nanosensors. It has the following beneficial effects:

[0023] This nanosensor-based high-precision material defect detection method achieves high-precision identification of tiny defects in materials by combining nanosensors with advanced artificial intelligence algorithms. It utilizes a random forest algorithm to classify defect types and combines image data with signal signature data to ensure accurate classification and location of multiple defect types. This solution effectively improves the accuracy of material defect identification, demonstrating excellent classification and location capabilities, particularly in complex material environments.

[0024] Effective noise removal and signal enhancement: Through multi-band filtering using a four-layer wavelet transform, this technology significantly reduces high-frequency noise in the signal, retains the signal's key information, and significantly improves the signal-to-noise ratio. The original signal's signal-to-noise ratio is increased from 2dB to 15dB, making defect features more distinct in the processed signal and further enhancing the reliability and accuracy of defect detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 The present invention is a flowchart for implementing the present invention. DETAILED DESCRIPTION

[0026] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0027] like Figure 1 As shown, an embodiment of the present invention provides a high-precision material defect detection method based on a nanosensor, comprising: S1. providing a nanosensor, the nanosensor detecting tiny defects on the surface or inside of a material;

[0028] S2. Install the nanosensor on the surface or inside of the material to be tested, and use the sensor to monitor the material in real time to obtain signals of potential defects in the material. The nanosensor is a thin-film sensor. The thin-film sensor is used to capture signals of tiny defects and has a high surface area and response speed. The film thickness of the thin-film sensor is 20nm to 100nm, and the surface area is greater than 5cm 2, which can effectively increase the contact area with the material to be detected and improve the sensitivity and accuracy of signal reception. The structure of the nanosensor is designed as a multi-layer thin film structure, which includes a functionalized surface to enhance its interaction with the surface defects of the material. This functionalized surface is chemically modified to enhance the detection sensitivity of tiny defects. During operation, the nanosensor can capture the intensity range of defect signals of 10 -6 A to 10 -3 A. The signal change frequency range is 1Hz to 10kHz, which can accurately reflect the dynamic changes of tiny defects. The signal is transmitted to the main control system in real time through the wireless communication module for further processing and analysis. The sensitivity of the nanosensor can reach 0.01μm, which can detect tiny cracks, pores or other defects on the surface or inside the material. The response time is 10ms, ensuring a rapid response to defects during real-time monitoring, suitable for defect detection in high-speed production environments. The operating temperature range of the sensor is -40℃ to 150℃, ensuring its stable operation under a wide range of environmental conditions.

[0029] S3. The data processing unit processes the signals acquired by the sensor and uses a signal enhancement algorithm to optimize weak signals to improve the accuracy of identifying minor defects and reduce interference from environmental noise. The signal enhancement algorithm includes a multi-level signal filtering module that removes noise based on wavelet transform to improve the signal-to-noise ratio.

[0030] The multi-level signal filtering module performs a total of 4 layers of wavelet decomposition, and each layer of decomposition generates the following two parts:

[0031] Low-frequency part: contains the main information of the signal (such as defect characteristics) and usually has a lower frequency.

[0032] High-frequency part: contains the noise components in the signal (such as environmental noise, instrument noise, etc.), usually with a higher frequency.

[0033] Specific data processed by each layer

[0034] Original signal data: Assume that the original signal frequency range is 1 Hz to 10 kHz, the sampled signal resolution is 16-bit, the sampling frequency is 20 kHz, and the signal-to-noise ratio of the original signal is 2 dB.

[0035] First level decomposition:

[0036] Low-frequency part: After the first layer of decomposition, the frequency range of the low-frequency part is 1Hz to 1kHz, retaining the key characteristics of the signal.

[0037] High frequency part: The frequency range is 1kHz to 10kHz, which greatly reduces noise.

[0038] Filter weight: The low-frequency part has a weight of 0.8, and the high-frequency part has a weight of 0.2.

[0039] Second level decomposition:

[0040] Low-frequency part: The frequency range of the signal is further reduced to 1Hz to 500Hz, while the signal characteristics are still retained.

[0041] High frequency part: The frequency range is 500Hz to 2kHz, further reducing noise.

[0042] Filter weight: The low-frequency part has a weight of 0.85, and the high-frequency part has a weight of 0.15.

[0043] The third level of decomposition:

[0044] Low-frequency part: The frequency range is 1Hz to 250Hz, and it continues to retain the core information of the signal.

[0045] High frequency part: The frequency range is 250Hz to 1kHz, which further removes high frequency noise.

[0046] Filter weight: The low-frequency part has a weight of 0.9, and the high-frequency part has a weight of 0.1.

[0047] Fourth level decomposition:

[0048] Low frequency part: The frequency range is 1Hz to 100Hz, retaining the most important low-frequency information.

[0049] High frequency part: The frequency range is 100Hz to 500Hz, reducing subtle noise.

[0050] Filter weight: The low-frequency part has a weight of 0.95, and the high-frequency part has a weight of 0.05.

[0051] During each layer of decomposition, the low-frequency portion retains the key signal information, while the high-frequency portion compresses and attenuates the noise. The signal-to-noise ratio is increased from 2dB in the original signal to 15dB after denoising.

[0052] S4. Based on the signal processing results, the defect type and location are analyzed. Combined with artificial intelligence algorithms, automated defect classification and quantitative analysis are implemented to accurately identify different types of defects in the material and output a defect report. S4 specifically includes:

[0053] S4.1 uses the support vector machine random forest algorithm to classify the types of material defects and conducts quantitative analysis based on the size and depth characteristics of the defects. The random forest uses a multi-decision tree model to vote on the classification of defects during the training phase to enhance the stability and accuracy of the classification. The construction of the decision tree depends on the combination of signal characteristics and material properties (such as density, hardness, and elastic modulus). The size of the training data set is 10,000 sample data, which includes 6 types of defects (such as cracks, holes, bubbles, corrosion, missing materials, and deformation), and the number of samples of each type is equal. In the end, the classification accuracy of the support vector machine and random forest was 98%, and the quantitative analysis error of the defect size and depth was ±0.2mm;

[0054] S4.2 uses a deep learning-based target detection model to automatically classify material defects using both image data and signal feature data to achieve high-precision and high-efficiency defect identification.

[0055] Defect location is determined by combining ultrasonic imaging with nanosensor data to improve positioning accuracy in complex material environments. Defect reports include defect type, location, size, severity, and impact on material properties. Ultrasonic sensors emit high-frequency sound waves and measure the echo time to form a reflected image. With a resolution of 0.1mm, the system can clearly display even tiny cracks and bubbles within the material. The ultrasonic imaging image is combined with the signal data provided by the nanosensor, and a data fusion algorithm is used to further improve positioning accuracy. After data fusion, the defect location accuracy reaches 0.05mm.

[0056] S5. During the real-time monitoring process, the sensor's operating status is monitored and adjusted through a feedback control system to ensure its stability in long-term use and adapt to changes in the material environment. The specific steps of S5 include:

[0057] S5.1 monitors the working status of the sensor in real time and checks whether the sensor's performance meets the requirements by comparing the current working status with the preset standard value;

[0058] S5.2 dynamically adjusts the sensor's sensitivity and sampling frequency through an adaptive control algorithm to respond to changes in the material environment and maintain the sensor's optimal detection performance. The feedback control system adjusts the sensor's operating parameters based on changes in temperature, humidity, pressure, and vibration to ensure the sensor's efficient operation under different environmental conditions.

[0059] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A high-precision material defect detection method based on nanosensors, characterized in that: The following steps are involved: S1. Providing a nanosensor that detects tiny defects on the surface or inside a material; S2. The nanosensor is mounted on the surface or inside of the material to be detected, and the material is monitored in real time by the sensor to obtain signals of potential defects in the material; S3. The signal obtained by the sensor is processed by the data processing unit, and a signal enhancement algorithm is used to optimize the weak signal to improve the accuracy of identifying small defects and reduce interference from environmental noise; S4. Analyze the defect type and location based on the signal processing results, and use artificial intelligence algorithms to automatically classify and quantitatively analyze defects, thereby accurately identifying different types of defects in the material and outputting a defect report; S5. During the real-time monitoring process, the working status of the sensor is monitored and adjusted through the feedback control system to ensure its stability in long-term use and adapt to changes in the material environment.

2. The high-precision material defect detection method based on nanosensors according to claim 1, characterized in that: The nanosensor is a thin film sensor, which is used to capture signals of tiny defects.

3. The high-precision material defect detection method based on nanosensors according to claim 1, characterized in that: The signal enhancement algorithm includes a multi-level signal filtering module, which removes noise based on wavelet transform to improve the signal-to-noise ratio of the signal.

4. The high-precision material defect detection method based on nanosensors according to claim 3, characterized in that: S4 specifically includes: S4.1 uses the support vector machine random forest algorithm to classify the types of material defects and conduct quantitative analysis based on the size and depth characteristics of the defects; S4.2 uses a deep learning-based target detection model to automatically classify material defects using image data and signal feature data.

5. The high-precision material defect detection method based on nanosensors according to claim 4, characterized in that: The defect position is located by combining ultrasonic imaging with nanosensor data.

6. The high-precision material defect detection method based on nanosensors according to claim 5, characterized in that: The defect report includes the type, location, size, severity, and impact of the defect on material properties.

7. The high-precision material defect detection method based on nanosensors according to claim 1, characterized in that: The specific steps of S5 include: S5.1 monitors the working status of the sensor in real time and checks whether the sensor's performance meets the requirements by comparing the current working status with the preset standard value; S5.2 dynamically adjusts the sensor's sensitivity and sampling frequency through an adaptive control algorithm to respond to changes in the material environment.

8. The high-precision material defect detection method based on nanosensors according to claim 1, characterized in that: The feedback control system adjusts the operating parameters of the sensor according to changes in temperature, humidity, pressure and vibration.

Images