Axle load detection method based on optical fiber sensing and time-frequency scale diagram

Abstract

The invention discloses an axle load detection method based on optical fiber sensing and a time-frequency scale diagram. The axle load detection method comprises the following steps: (1) signal acquisition; and (2) data processing: de-noising, time-frequency scale map generation, axle identification and load estimation are carried out in sequence, and finally a corresponding axle load # imgabs0 # is obtained. The axle load detection method based on the optical fiber sensing and the time-frequency scale diagram has the beneficial effects that multi-resolution analysis of vehicle axle load signals is realized through the time-frequency scale diagram, and dynamic frequency components in the signals are effectively captured, so that the accuracy of axle recognition and load estimation is improved, the influence of environmental noise is effectively inhibited, and the detection accuracy is improved. The stability and robustness of signal features are enhanced; the method can effectively process non-stationary signals, is suitable for real-time monitoring and data analysis in various complex traffic environments, facilitates accurate identification of different axles in complex traffic scenes, and carries out accurate load statistics.

Description

Technical Field

[0001] The patent of this invention belongs to the field of road surface monitoring technology, and specifically relates to an axle load detection method based on optical fiber sensing and time-frequency scaling diagram. Background Art

[0002] In the field of traffic monitoring and management, vehicle axle load detection is a key link in road safety management. Traditional axle load detection methods usually rely on weighing devices installed on the road surface or sensors embedded in the road surface, which directly measure the weight of the vehicle to determine its load. However, these traditional methods are not only costly and complex to maintain, but also prone to errors when dealing with high-speed vehicles, and cannot meet the requirements of modern traffic management for high accuracy, real-time performance and cost-effectiveness.

[0003] In recent years, fiber optic sensing technology has gradually become a new method for axle load detection due to its advantages such as high sensitivity, resistance to electromagnetic interference and resistance to environmental changes. Fiber optic sensors can reflect the force exerted on the road surface when a vehicle passes by by detecting changes in light intensity, thereby indirectly estimating the vehicle's axle load. However, the above detection method has the following problems: (1) Limited frequency analysis capability: Time domain signals mainly provide amplitude information that changes over time and cannot effectively identify and analyze frequency components in complex signals. For multi-axle vehicles or high-speed passing, the dynamic changes of frequency components are difficult to capture, thus affecting the accuracy of detection.

[0004] (2) Poor noise suppression capability: Fiber optic sensors are susceptible to noise interference in complex traffic environments. Traditional time domain analysis is difficult to effectively filter out noise, resulting in unstable signal analysis results, which in turn affects the accuracy of axle identification and load estimation.

[0005] (3) Difficulty in extracting signal features: The original time domain signal is difficult to intuitively display the key features of a vehicle passing by. Especially in the face of complex scenarios where multiple vehicles pass by continuously, traditional methods are difficult to effectively distinguish and identify the load characteristics of different vehicles.

[0006] (4) Poor adaptability to complex signal processing: When the vehicle load signal exhibits non-stationary characteristics (such as vehicle acceleration, deceleration, vibration caused by different road conditions, etc.), traditional methods are difficult to provide high-precision analysis results.

[0007] In order to solve the above problems, the present invention proposes an axle load detection method based on fiber optic sensing and time-frequency scaling diagram. Through multi-resolution analysis, enhanced noise suppression capability and intuitive signal feature extraction, high-precision detection and load statistics of vehicle axle loads are achieved, which is particularly suitable for complex traffic monitoring scenarios. Summary of the invention

[0008] In view of this, the purpose of the present invention is to provide an axle load detection method based on optical fiber sensing and time-frequency scaling diagram. The detection method provided by the present invention improves the accuracy of axle identification and load estimation; can effectively process non-stationary signals, and is suitable for real-time monitoring and data analysis in various complex traffic environments; helps to accurately identify different axles in complex traffic scenarios and perform accurate load statistics.

[0009] The technical solution of the present invention discloses an axle load detection method based on optical fiber sensing and time-frequency scalogram, which comprises the following steps: (1) Signal acquisition: The optical fiber sensor laid on the road captures the light intensity change signal under the axle load of the passing vehicle, and obtains the time domain electrical signal through photoelectric conversion. ; (2) Data processing: After denoising, time-frequency scalogram generation, axle identification and load estimation, the corresponding axle load is finally obtained. ; Among them, 2.1 Denoising: Use bandpass filter and wavelet denoising method to denoise the time domain electrical signal in step (1) Perform filtering to remove environmental noise , and obtain the purified signal ; 2.2 Time-frequency scale map generation: The signal in step 2.1 is transformed by short-time Fourier transform Perform time-frequency analysis to obtain time-frequency domain signals ; Then generate the time-frequency scale diagram according to the following formula ; ; 2.3 Axle identification and load estimation: Perform local maximum detection on the time-frequency scale map generated in step 2.2 to obtain the local maximum set, and then set the threshold , keep greater than The local maximum ; Then, the corresponding axle load is calculated based on the time-frequency domain signal of the retained local maximum value :

[0010]

[0011] in, is the axle load of the calibrated vehicle, is the maximum value of the corresponding optical fiber signal video scale diagram.

[0012] Furthermore, in step (1), two groups of optical fiber sensors are laid in sequence on the lane surface along the vehicle travel direction, the width of the optical fiber sensors laid on the lane is 1.5 meters to 3.5 meters, and the spacing between the two groups of optical fiber sensors is 3-10 meters.

[0013] Furthermore, in step 2.1, a bandpass filter is first used to perform preliminary noise removal, and then wavelet denoising is used to perform further noise removal.

[0014] Furthermore, the short-time Fourier transform calculation formula in step 2.2 is specifically: ; in, is the original signal, is a window function, is a frequency variable, It is the center of time. is the integration variable.

[0015] Furthermore, the method for detecting the local maximum in step 2.3 is: a) Define the local area: First, a rectangle is used to define the local area. The neighborhood is centered on the current pixel. The side length of the rectangular area is set to 200-400 sampling points, and the sampling frequency is 1000-2000HZ; b) Search for maximum values ​​in local areas: For each local area, the maximum point or peak point is detected by comparing the value in the local area with the values ​​of the surrounding neighborhood. If the value in the local area is greater than or equal to all the values ​​of the surrounding neighborhood, the point is considered to be a local maximum point.

[0016] Furthermore, the threshold in step 2.3 It is 0.02-0.06.

[0017] Advantages of the present invention: 1. The present invention discloses an axle load detection method based on optical fiber sensing and time-frequency scaling diagram. Through frequency domain filtering and time-frequency analysis methods, the influence of environmental noise is effectively suppressed and the stability and robustness of signal characteristics are enhanced. The method can effectively process non-stationary signals and is suitable for real-time monitoring and data analysis in various complex traffic environments.

[0018] 2. The present invention discloses an axle load detection method based on optical fiber sensing and time-frequency scaling diagram. The multi-resolution analysis of the vehicle axle load signal is realized through the time-frequency scaling diagram, and the dynamic frequency components in the signal are effectively captured, thereby improving the accuracy of axle identification and load estimation, with an accuracy of up to 94%.

[0019] 3. The present invention discloses an axle load detection method based on optical fiber sensing and time-frequency scaling diagram, wherein the time-frequency scaling diagram intuitively displays the load characteristics when a vehicle passes, which helps to accurately identify different axles in complex traffic scenarios and perform accurate load statistics. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0021] Figure 1 It is a specific flow chart of the vehicle axle load detection method of the present invention.

[0022] Figure 2 This is a time-frequency scale diagram of Example 1 of the present invention.

[0023] Figure 3 This is a diagram of the axle load detection results of Example 1 of the present invention. DETAILED DESCRIPTION

[0024] The present invention is further described in detail below by way of examples.

[0025] Example 1: Figure 1 As shown, a method for detecting axle load based on optical fiber sensing and time-frequency scaling diagram comprises the following steps: (1) Signal acquisition: The optical fiber sensor laid on the road captures the light intensity change signal under the axle load of the passing vehicle, and obtains the time domain electrical signal through photoelectric conversion. Two groups of optical fiber sensors are laid on the lane surface in sequence along the direction of vehicle travel. The width of the optical fiber sensors laid on the lane is 1.5 meters, and the spacing between the two groups of optical fiber sensors is 3m.

[0026] (2) Data processing: 2.1 Denoising: Use bandpass filter and wavelet denoising method to denoise the time domain electrical signal in step (1). Perform filtering to remove environmental noise , and obtain the purified signal The specific filtering process is to first use a bandpass filter to perform preliminary noise removal, and then use wavelet denoising to further remove noise.

[0027] 2.2 Time-frequency scale map generation: The signal in step 2.1 is transformed by short-time Fourier transform Perform time-frequency analysis to obtain time-frequency domain signals ; Then generate the time-frequency scale diagram according to the following formula ,like Figure 2 As shown; ; The short-time Fourier transform calculation formula is as follows: ; in, is the original signal, is a window function, is a frequency variable, It is the center of time. is the integration variable.

[0028] 2.3 Axle identification and load estimation: Perform local maximum detection on the time-frequency scale map generated in step 2.2 to obtain the local maximum set, and then set the threshold , keep greater than The local maximum ; Then, the corresponding axle load is calculated based on the time-frequency domain signal of the retained local maximum value :

[0029]

[0030] in, is the axle load of the calibrated vehicle, is the maximum value of the corresponding optical fiber signal video scale diagram.

[0031] The method for local maximum detection is: a) Define the local area: First, a rectangle is used to define the local area. The neighborhood is centered on the current pixel, the side length of the rectangular area is set to 200 sampling points, and the sampling frequency is 1000HZ; b) Search for maximum values ​​in local areas: For each local area, the maximum point or peak point is detected by comparing the value in the local area with the values ​​of the surrounding neighborhood. If the value in the local area is greater than or equal to all the values ​​of the surrounding neighborhood, the point is considered to be a local maximum point.

[0032] Example 2: Figure 1 As shown, a method for detecting axle load based on optical fiber sensing and time-frequency scaling diagram comprises the following steps: (1) Signal acquisition: The optical fiber sensor laid on the road captures the light intensity change signal under the axle load of the passing vehicle, and obtains the time domain electrical signal through photoelectric conversion. Two groups of optical fiber sensors are laid on the lane surface in sequence along the direction of vehicle travel. The width of the optical fiber sensors laid on the lane is 3.5 meters, and the spacing between the two groups of optical fiber sensors is 10m.

[0033] (2) Data processing: 2.1 Denoising: Use bandpass filter and wavelet denoising method to denoise the time domain electrical signal in step (1). Perform filtering to remove environmental noise , and obtain the purified signal The specific filtering process is to first use a bandpass filter to perform preliminary noise removal, and then use wavelet denoising to further remove noise.

[0034] 2.2 Time-frequency scale map generation: The signal in step 2.1 is transformed by short-time Fourier transform Perform time-frequency analysis to obtain time-frequency domain signals ; Then generate the time-frequency scale diagram according to the following formula ,like Figure 2 As shown; ; The short-time Fourier transform calculation formula is as follows: ; in, is the original signal, is a window function, is a frequency variable, It is the center of time. is the integration variable.

[0035] 2.3 Axle identification and load estimation: Perform local maximum detection on the time-frequency scale map generated in step 2.2 to obtain the local maximum set, and then set the threshold , keep greater than The local maximum ; Then, the corresponding axle load is calculated based on the time-frequency domain signal of the retained local maximum value :

[0036]

[0037] in, is the axle load of the calibrated vehicle, is the maximum value of the corresponding optical fiber signal video scale diagram.

[0038] The method for local maximum detection is: a) Define the local area: First, a rectangle is used to define the local area. The neighborhood is centered on the current pixel, the side length of the rectangular area is set to 400 sampling points, and the sampling frequency is 2000HZ; b) Search for maximum values ​​in local areas: For each local area, the maximum point or peak point is detected by comparing the value in the local area with the values ​​of the surrounding neighborhood. If the value in the local area is greater than or equal to all the values ​​of the surrounding neighborhood, the point is considered to be a local maximum point.

[0039] Example 3: Figure 1 As shown, a method for detecting axle load based on optical fiber sensing and time-frequency scaling diagram comprises the following steps: (1) Signal acquisition: The optical fiber sensor laid on the road captures the light intensity change signal under the axle load of the passing vehicle, and obtains the time domain electrical signal through photoelectric conversion. Two groups of optical fiber sensors are laid on the lane surface in sequence along the direction of vehicle travel. The width of the optical fiber sensors laid on the lane is 2.5 meters, and the spacing between the two groups of optical fiber sensors is 4 meters.

[0040] (2) Data processing: 2.1 Denoising: Use bandpass filter and wavelet denoising method to denoise the time domain electrical signal in step (1). Perform filtering to remove environmental noise , and obtain the purified signal The specific filtering process is to first use a bandpass filter to perform preliminary noise removal, and then use wavelet denoising to further remove noise.

[0041] 2.2 Time-frequency scale map generation: The signal in step 2.1 is transformed by short-time Fourier transform Perform time-frequency analysis to obtain time-frequency domain signals ; Then generate the time-frequency scale diagram according to the following formula ,like Figure 2 As shown; ; The short-time Fourier transform calculation formula is as follows: ; in, is the original signal, is a window function, is a frequency variable, It is the center of time. is the integration variable.

[0042] 2.3 Axle identification and load estimation: Perform local maximum detection on the time-frequency scale map generated in step 2.2 to obtain the local maximum set, and then set the threshold , keep greater than The local maximum ; Then, the corresponding axle load is calculated based on the time-frequency domain signal of the retained local maximum value :

[0043]

[0044] in, is the axle load of the calibrated vehicle, is the maximum value of the corresponding optical fiber signal video scale diagram.

[0045] The method for local maximum detection is: a) Define the local area: First, a rectangle is used to define the local area. The neighborhood is centered on the current pixel, the side length of the rectangular area is set to 300 sampling points, and the sampling frequency is 1500HZ; b) Search for maximum values ​​in local areas: For each local area, the maximum point or peak point is detected by comparing the value in the local area with the values ​​of the surrounding neighborhood. If the value in the local area is greater than or equal to all the values ​​of the surrounding neighborhood, the point is considered to be a local maximum point.

[0046] Experiment: The actual axle loads of the front and rear axles of a standard car are 12.74 kN and 14.70 kN respectively. After the car passes through the optical fiber sensor, the data of Example 1 is processed and the following is obtained: Figure 3 ,like Figure 3 As shown, the red frame 15 shows the local maximum of the front axle is 0.061, so we have ; The red frame 16 shows the local maximum of the rear axle is 0.066, put it into the formula The rear axle load calculation value is obtained , so the accuracy of Example 1 of the present invention is: 13.8÷14.7×100%=94% The above are preferred embodiments of the present invention. For ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for detecting axle load based on optical fiber sensing and time-frequency scalogram, comprising the following steps: (1) Signal acquisition: The optical fiber sensor laid on the road captures the light intensity change signal under the axle load of the passing vehicle, and obtains the time domain electrical signal through photoelectric conversion. ; It is characterized in that it also includes the following steps: (2) Data processing: After denoising, time-frequency scalogram generation, axle identification and load estimation, the corresponding axle load is finally obtained. ; in, 2.1 Denoising: Use bandpass filter and wavelet denoising method to denoise the time domain electrical signal in step (1). Perform filtering to remove environmental noise , and obtain the purified signal ; 2.2 Time-frequency scale map generation: The signal in step 2.1 is transformed by short-time Fourier transform Perform time-frequency analysis to obtain time-frequency domain signals ; Then generate the time-frequency scale diagram according to the following formula ; ； 2.3 Axle identification and load estimation: Perform local maximum detection on the time-frequency scale map generated in step 2.2 to obtain the local maximum set, and then set the threshold , keep greater than The local maximum ; Then, the corresponding axle load is calculated based on the time-frequency domain signal of the retained local maximum value : in, is the axle load of the calibrated vehicle, is the maximum value of the corresponding optical fiber signal video scale diagram.

2. The axle load detection method based on optical fiber sensing and time-frequency scalogram according to claim 1 is characterized in that: In step (1), two groups of optical fiber sensors are laid in sequence on the lane surface along the direction of vehicle travel, the width of the optical fiber sensors laid on the lane is 1.5 meters to 3.5 meters, and the spacing between the two groups of optical fiber sensors is 3-10 meters.

3. The axle load detection method based on optical fiber sensing and time-frequency scale diagram according to claim 1 is characterized in that: In step 2.1, a bandpass filter is first used to perform preliminary noise removal, and then wavelet denoising is used to perform further noise removal.

4. The axle load detection method based on optical fiber sensing and time-frequency scaling diagram according to claim 1 is characterized in that: The short-time Fourier transform calculation formula in step 2.2 is specifically: ； in, is the original signal, is a window function, is a frequency variable, It is the center of time. is the integration variable.

5. The axle load detection method based on optical fiber sensing and time-frequency scalogram according to claim 1 is characterized in that: The method for local maximum detection in step 2.3 is: a) Define the local area: First, a rectangle is used to define the local area. The neighborhood is centered on the current pixel. The side length of the rectangular area is set to 200-400 sampling points, and the sampling frequency is 1000-2000HZ; b) Search for maximum values ​​in local areas: For each local area, the maximum point or peak point is detected by comparing the value in the local area with the values ​​of the surrounding neighborhood. If the value in the local area is greater than or equal to all the values ​​of the surrounding neighborhood, the point is considered to be a local maximum point.

6. The axle load detection method based on optical fiber sensing and time-frequency scalogram according to claim 1 is characterized in that: Step 2.3: Queue It is 0.02-0.06.

Images