Intelligent early warning method for tunnel blasting construction safety risk

Abstract

The invention provides an intelligent early warning method for safety risks of tunnel blasting construction. The intelligent early warning method comprises the following steps: acquiring total blasting vibration energy corresponding to each group of blasting parameters; performing regression analysis on the blast vibration total energy data to construct a blast vibration total energy prediction formula; setting a corresponding safety distance for each blasting vibration total energy; obtaining an infrared image of a tunnel blasting construction site; preprocessing the infrared image to form a training sample; inputting the training sample into a neural network for training to obtain a constructor detection model; and using the constructor to detect whether a worker exists within a safe distance of the model, and if yes, giving out an early warning. According to the invention, the training of the early warning model is completed by adopting the thermal imaging technology and combining the neural network, and the accuracy of identifying the staff can be maintained in complex scenes such as a dust environment or a dark environment.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel blasting construction, and in particular to an intelligent early warning method for safety risks in tunnel blasting construction. Background Art

[0002] Tunnels are key projects in highway and railway construction. With the development of railway construction and technological advancements, tunnel excavation methods have rapidly evolved. Common excavation methods include drilling and blasting, shield tunneling, and tunnel boring machines. Drilling and blasting, due to its adaptability to geological conditions and low excavation costs, is particularly suitable for tunnels in hard and broken rock, as well as for the construction of numerous short tunnels. Therefore, drilling and blasting remains a commonly used tunnel excavation method both domestically and internationally. However, the blasting and excavation process inevitably disturbs the surrounding retained rock mass, causing rockfall within the tunnel and potentially harming workers. Traditional early warning technologies rely heavily on manual experience, which can be prone to misjudgments or missed detections. Summary of the Invention

[0003] To solve the above problems, the purpose of the present invention is to provide an intelligent early warning method for tunnel blasting construction safety risks.

[0004] The intelligent early warning method for tunnel blasting construction safety risks includes:

[0005] Step 1: Obtain the total blasting vibration energy corresponding to each set of blasting parameters;

[0006] Step 2: Perform regression analysis on the blasting vibration total energy data to construct a blasting vibration total energy prediction formula;

[0007] Step 3: Set the corresponding safety distance for each blasting vibration total energy;

[0008] Step 4: Acquire infrared images of the tunnel blasting construction site;

[0009] Step 5: Preprocess the infrared image to form a training sample;

[0010] Step 6: Input the training samples into the neural network for training to obtain the construction worker detection model;

[0011] Step 7: Use the construction personnel to detect whether there are any workers within the safe distance of the model. If there are any workers, an early warning is issued.

[0012] Preferably, the step 2: performing regression analysis on the blasting vibration total energy data to construct a blasting vibration total energy prediction formula includes:

[0013] The dimensional analysis method is used to conduct regression analysis on the total blasting vibration energy data, the distance from the blast center, and the maximum charge amount to construct a prediction formula for the total blasting vibration energy. The prediction formula for the total blasting vibration energy is:

[0014]

[0015] Among them, E represents the total energy of blasting vibration, Q represents the maximum charge, and R represents the distance from the explosion center.

[0016] Preferably, the step 5: pre-processing the infrared image to form a training sample includes:

[0017] Step 5.1: Build a sliding window on the infrared image, detect the noise points on each sliding window, and remove the corresponding noise points;

[0018] Step 5.2: Use the Gaussian function to perform edge detection on the infrared image to obtain the amplitude of each pixel;

[0019] Step 5.3: Calculate the edge detection threshold and take the pixels greater than the edge detection threshold as edge points;

[0020] Step 5.4: Calibrate the contour image composed of each edge point;

[0021] Step 5.5: Obtain the average temperature corresponding to the contour image;

[0022] Step 5.6: The contour images with average temperature within the preset range are used as staff and annotated to form training samples.

[0023] Preferably, the step 5.1: constructing a sliding window on the infrared image, detecting noise points on each sliding window, and removing the corresponding noise points includes:

[0024] Construct a 3*3 sliding window with any point as the center on the infrared image and calculate the variance of the pixels on the 3*3 sliding window;

[0025] Determine whether the difference between the central pixel and the pixel variance exceeds the threshold. Pixels exceeding the threshold are considered noise points, and the corresponding sliding window is denoised using a denoising algorithm. The sliding window is continuously moved until the entire infrared image is traversed. The noise point judgment formula is:

[0026]

[0027] Among them, P i,j represents variance, Mean(F i,j) represents the mean value of pixels in the sliding window, f(i+x,j+y) represents the pixel value at the position (i+x,j+y), and f(i,j) represents the pixel value at the position (i,j).

[0028] Preferably, step 5.2: performing edge detection on the infrared image using a Gaussian function to obtain the amplitude of each pixel includes:

[0029] Step 5.2.1: Use the Gaussian function to convolve the infrared image to obtain the convolved infrared image; wherein the convolved infrared image is:

[0030]

[0031] S(x,y)=G*I

[0032] Where S(x,y) represents the convolved infrared image, G represents the Gaussian function, σ represents the standard deviation of the Gaussian function, x represents the value of the horizontal coordinate on the Gaussian function, y represents the value of the vertical coordinate on the Gaussian function, and I represents the original infrared image;

[0033] Step 5.2.2: Calculate the amplitude of the convolved infrared image. The amplitude calculation formula is:

[0034]

[0035] Among them, S(x+1,y) represents the value of the convolved infrared image at the position (x+1,y), S(x-1,y) represents the value of the convolved infrared image at the position (x-1,y), S(x,y+1) represents the value of the convolved infrared image at the position (x,y+1), S(x,y-1) represents the value of the convolved infrared image at the position (x-1,y), g x Indicates the magnitude of the x-direction, g y Indicates the magnitude of the y-direction, g θ Indicates the total amplitude.

[0036] Preferably, in step 5.3, the edge detection threshold is calculated as follows:

[0037] The amplitude in the x-direction is used as the horizontal axis, and the amplitude in the y-direction is used as the y-axis to construct an amplitude image. The edge detection threshold is solved according to the mean and standard deviation of the amplitude image. The edge detection threshold calculation formula is:

[0038]

[0039] Where P represents the mean of the amplitude image, σ represents the standard deviation of the amplitude image, and H represents the edge detection threshold.

[0040] Preferably, the step 6: inputting the training samples into the neural network for training to obtain a construction worker detection model includes:

[0041] Inputting the training samples into the neural network and optimizing the neural network using a loss function to obtain a construction worker detection model;

[0042]

[0043] Among them, N represents the number of training samples, C represents the category to which the training samples belong, and y ij represents the true category label of the training sample, Represents the predicted value of the neural network.

[0044] The present invention also provides an intelligent early warning system for tunnel blasting construction safety risks, comprising:

[0045] The blasting parameter acquisition module is used to obtain the total blasting vibration energy corresponding to each set of blasting parameters;

[0046] Regression analysis module, used to perform regression analysis on blasting vibration total energy data to construct a blasting vibration total energy prediction formula;

[0047] A safety distance setting module is used to set a corresponding safety distance for each blasting vibration total energy;

[0048] Infrared image acquisition module, used to acquire infrared images of the tunnel blasting construction site;

[0049] A preprocessing module is used to preprocess infrared images to form training samples;

[0050] A training module is used to input training samples into a neural network for training to obtain a construction worker detection model;

[0051] The early warning module is used to use the construction personnel to detect whether there are workers within the safe distance of the model, and to issue an early warning if there are workers.

[0052] The present invention also provides an electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus. The electronic device is characterized in that when the computer program is executed by the processor, the steps of the above-mentioned intelligent early warning method for safety risks in tunnel blasting construction are implemented.

[0053] The present invention also provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps in the above-mentioned intelligent early warning method for safety risks in tunnel blasting construction are implemented.

[0054] The beneficial effect of the intelligent early warning method and system for tunnel blasting construction safety risks provided by the present invention is that, compared with the existing technology, the present invention uses thermal imaging technology combined with a neural network to complete the training of the early warning model, and can maintain the accuracy of worker identification in complex scenarios such as dusty or dark environments.

[0055] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. 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 any creative work.

[0057] Figure 1 A flow chart of an intelligent early warning method for tunnel blasting construction safety risks provided by an embodiment of the present invention is shown;

[0058] Figure 2 A blasting parameter data diagram provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0059] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like to indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0060] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature identified as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0061] In the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," "connect," "fixed," etc. should be understood broadly. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediary; or internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0062] See also Figure 1 , the intelligent early warning method for tunnel blasting construction safety risks includes:

[0063] Step 1: Obtain the total blasting vibration energy corresponding to each set of blasting parameters;

[0064] The damage caused by blasting vibration to surrounding rock, supporting structures, and the surrounding environment is essentially the conversion of explosive chemical energy into mechanical vibration energy. Total energy directly reflects the potential destructive intensity of a blast and more comprehensively characterizes the energy distribution of the shock wave than single peak vibration velocity (PPV) or frequency.

[0065] Step 2: Perform regression analysis on the blasting vibration total energy data to construct a blasting vibration total energy prediction formula;

[0066] like Figure 2 As shown in the figure, the present invention uses dimensional analysis to perform regression analysis on the total blasting vibration energy data, the distance from the explosion center, and the maximum charge to construct a prediction formula for the total blasting vibration energy; wherein, the prediction formula for the total blasting vibration energy is:

[0067]

[0068] Among them, E represents the total energy of blasting vibration, Q represents the maximum charge, and R represents the distance from the explosion center.

[0069] Step 3: Set the corresponding safety distance for each blasting vibration total energy;

[0070] Step 4: Acquire infrared images of the tunnel blasting construction site;

[0071] Step 5: Preprocess the infrared image to form a training sample;

[0072] Furthermore, step 5 includes:

[0073] Step 5.1: Build a sliding window on the infrared image, detect the noise points on each sliding window, and remove the corresponding noise points;

[0074] In step 5.1, a 3*3 sliding window is constructed with an arbitrary point as the center on the infrared image, and the variance of the pixels on the 3*3 sliding window is calculated;

[0075] Determine whether the difference between the central pixel and the pixel variance exceeds the threshold. Pixels exceeding the threshold are considered noise points, and the corresponding sliding window is denoised using a denoising algorithm. The sliding window is continuously moved until the entire infrared image is traversed. The noise point judgment formula is:

[0076]

[0077] Among them, P i,j represents variance, Mean(F i,j ) represents the mean value of pixels in the sliding window, f(i+x,j+y) represents the pixel value at the position (i+x,j+y), and f(i,j) represents the pixel value at the position (i,j). It should be noted that for a sliding window with all pixels being the same, it is generally considered to be a non-noise point.

[0078] Step 5.2: Use the Gaussian function to perform edge detection on the infrared image to obtain the amplitude of each pixel;

[0079] Furthermore, step 5.2 includes:

[0080] Step 5.2.1: Use the Gaussian function to convolve the infrared image to obtain the convolved infrared image; wherein the convolved infrared image is:

[0081]

[0082] S(x,y)=G*I

[0083] Where S(x,y) represents the convolved infrared image, G represents the Gaussian function, σ represents the standard deviation of the Gaussian function, x represents the value of the horizontal coordinate on the Gaussian function, y represents the value of the vertical coordinate on the Gaussian function, and I represents the original infrared image;

[0084] The Gaussian function is a linear filter that is obtained by sampling and normalizing a two-dimensional Gaussian function. By applying this filter to the original infrared image, it is possible to suppress noise while retaining the edge features of the image, providing a more accurate basis for subsequent edge detection.

[0085] Step 5.2.2: Calculate the amplitude of the convolved infrared image. The amplitude calculation formula is:

[0086]

[0087] Among them, S(x+1,y) represents the value of the convolved infrared image at the position (x+1,y), S(x-1,y) represents the value of the convolved infrared image at the position (x-1,y), S(x,y+1) represents the value of the convolved infrared image at the position (x,y+1), S(x,y-1) represents the value of the convolved infrared image at the position (x-1,y), g x Indicates the magnitude of the x-direction, g y Indicates the magnitude of the y-direction, g θ Indicates the total amplitude.

[0088] Step 5.3: Calculate the edge detection threshold and take the pixels greater than the edge detection threshold as edge points;

[0089] In step 5.3, the edge detection threshold is calculated as follows:

[0090] The amplitude in the x-direction is used as the horizontal axis, and the amplitude in the y-direction is used as the y-axis to construct an amplitude image. The edge detection threshold is solved according to the mean and standard deviation of the amplitude image. The edge detection threshold calculation formula is:

[0091]

[0092] Where P represents the mean of the amplitude image, σ represents the standard deviation of the amplitude image, and H represents the edge detection threshold.

[0093] Step 5.4: Calibrate the contour image composed of each edge point;

[0094] Step 5.5: Obtain the average temperature corresponding to the contour image;

[0095] Step 5.6: The contour images with average temperature within the preset range are used as staff and annotated to form training samples.

[0096] Step 6: Input the training samples into the neural network for training to obtain the construction worker detection model;

[0097] Furthermore, step 6 includes:

[0098] Inputting the training samples into the neural network and optimizing the neural network using a loss function to obtain a construction worker detection model;

[0099]

[0100] Among them, N represents the number of training samples, C represents the category to which the training samples belong, and y ij represents the true category label of the training sample, Represents the predicted value of the neural network.

[0101] Step 7: Use the construction personnel to detect whether there are any workers within the safe distance of the model. If there are any workers, an early warning is issued.

[0102] The present invention uses thermal imaging technology in combination with a neural network to complete the training of the early warning model, and can maintain the accuracy of identifying workers in complex scenarios such as dusty environments or dark environments.

[0103] The present invention also provides an intelligent early warning system for tunnel blasting construction safety risks, comprising:

[0104] The blasting parameter acquisition module is used to obtain the total blasting vibration energy corresponding to each set of blasting parameters;

[0105] Regression analysis module, used to perform regression analysis on blasting vibration total energy data to construct a blasting vibration total energy prediction formula;

[0106] A safety distance setting module is used to set a corresponding safety distance for each blasting vibration total energy;

[0107] Infrared image acquisition module, used to acquire infrared images of the tunnel blasting construction site;

[0108] A preprocessing module is used to preprocess infrared images to form training samples;

[0109] A training module is used to input training samples into a neural network for training to obtain a construction worker detection model;

[0110] The early warning module is used to use the construction personnel to detect whether there are workers within the safe distance of the model, and to issue an early warning if there are workers.

[0111] The present invention also provides an electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus. The invention is characterized in that when the computer program is executed by the processor, the steps of the above-mentioned intelligent early warning method for safety risks in tunnel blasting construction are implemented. Compared with the prior art, the beneficial effects of the electronic device provided by the present invention are the same as those of the intelligent early warning method for safety risks in tunnel blasting construction described in the above-mentioned technical solution, and are not further elaborated here.

[0112] The present invention also provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps in the above-mentioned intelligent early warning method for safety risks in tunnel blasting construction are implemented. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present invention are the same as the beneficial effects of the intelligent early warning method for safety risks in tunnel blasting construction described in the above-mentioned technical solution, and will not be elaborated here.

[0113] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technical solution that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. An intelligent early warning method for tunnel blasting construction safety risks, characterized by: include: Step 1: Obtain the total blasting vibration energy corresponding to each set of blasting parameters; Step 2: Perform regression analysis on the blasting vibration total energy data to construct a blasting vibration total energy prediction formula; Step 3: Set the corresponding safety distance for each blasting vibration total energy; Step 4: Acquire infrared images of the tunnel blasting construction site; Step 5: Preprocess the infrared image to form a training sample; Step 6: Input the training samples into the neural network for training to obtain the construction worker detection model; Step 7: Use the construction personnel to detect whether there are any workers within the safe distance of the model. If there are any workers, an early warning is issued.

2. The intelligent early warning method for tunnel blasting construction safety risks according to claim 1 is characterized in that: The step 2: performing regression analysis on the blasting vibration total energy data to construct a blasting vibration total energy prediction formula, including: The dimensional analysis method is used to conduct regression analysis on the total blasting vibration energy data, the distance from the blast center, and the maximum charge amount to construct a prediction formula for the total blasting vibration energy. The prediction formula for the total blasting vibration energy is: Among them, E represents the total energy of blasting vibration, Q represents the maximum charge, and R represents the distance from the explosion center.

3. The intelligent early warning method for tunnel blasting construction safety risks according to claim 2 is characterized in that: The step 5: pre-processing the infrared image to form a training sample, including: Step 5.1: Build a sliding window on the infrared image, detect the noise points on each sliding window, and remove the corresponding noise points; Step 5.2: Use the Gaussian function to perform edge detection on the infrared image to obtain the amplitude of each pixel; Step 5.3: Calculate the edge detection threshold and take the pixels greater than the edge detection threshold as edge points; Step 5.4: Calibrate the contour image composed of each edge point; Step 5.5: Obtain the average temperature corresponding to the contour image; Step 5.6: The contour images with average temperature within the preset range are used as staff and annotated to form training samples.

4. The intelligent early warning method for tunnel blasting construction safety risks according to claim 3 is characterized in that: Step 5.1: constructing a sliding window on the infrared image, detecting noise points on each sliding window, and removing the corresponding noise points, including: Construct a 3*3 sliding window with any point as the center on the infrared image and calculate the variance of the pixels on the 3*3 sliding window; Determine whether the difference between the central pixel and the pixel variance exceeds the threshold. Pixels exceeding the threshold are considered noise points, and the corresponding sliding window is denoised using a denoising algorithm. The sliding window is continuously moved until the entire infrared image is traversed. The noise point judgment formula is: Among them, P i,j represents variance, Mean(F i,j ) represents the mean value of pixels in the sliding window, f(i+x,j+y) represents the pixel value at the position (i+x,j+y), and f(i,j) represents the pixel value at the position (i,j).

5. The intelligent early warning method for tunnel blasting construction safety risks according to claim 4 is characterized in that: Step 5.2: Use the Gaussian function to perform edge detection on the infrared image to obtain the amplitude of each pixel, including: Step 5.2.1: Use the Gaussian function to convolve the infrared image to obtain the convolved infrared image; wherein the convolved infrared image is: S(x,y)=G*I Where S(x,y) represents the convolved infrared image, G represents the Gaussian function, σ represents the standard deviation of the Gaussian function, x represents the value of the horizontal coordinate on the Gaussian function, y represents the value of the vertical coordinate on the Gaussian function, and I represents the original infrared image; Step 5.2.2: Calculate the amplitude of the convolved infrared image. The amplitude calculation formula is: Among them, S(x+1,y) represents the value of the convolved infrared image at the position (x+1,y), S(x-1,y) represents the value of the convolved infrared image at the position (x-1,y), S(x,y+1) represents the value of the convolved infrared image at the position (x,y+1), S(x,y-1) represents the value of the convolved infrared image at the position (x-1,y), g x Indicates the magnitude of the x-direction, g y Indicates the magnitude of the y-direction, g θ Indicates the total amplitude.

6. The intelligent early warning method for tunnel blasting construction safety risks according to claim 5 is characterized in that: In step 5.3, the edge detection threshold is calculated as follows: The amplitude in the x-direction is used as the horizontal axis, and the amplitude in the y-direction is used as the y-axis to construct an amplitude image. The edge detection threshold is solved according to the mean and standard deviation of the amplitude image. The edge detection threshold calculation formula is: Where P represents the mean of the amplitude image, σ represents the standard deviation of the amplitude image, and H represents the edge detection threshold.

7. The intelligent early warning method for tunnel blasting construction safety risks according to claim 6 is characterized in that: Step 6: inputting the training samples into the neural network for training to obtain a construction worker detection model, including: Inputting the training samples into the neural network and optimizing the neural network using a loss function to obtain a construction worker detection model; Among them, N represents the number of training samples, C represents the category to which the training samples belong, and y ij represents the true category label of the training sample, Represents the predicted value of the neural network.

8. An intelligent early warning system for tunnel blasting construction safety risks, characterized by: include: The blasting parameter acquisition module is used to obtain the total blasting vibration energy corresponding to each set of blasting parameters; Regression analysis module, used to perform regression analysis on blasting vibration total energy data to construct a blasting vibration total energy prediction formula; A safety distance setting module is used to set a corresponding safety distance for each blasting vibration total energy; Infrared image acquisition module, used to acquire infrared images of the tunnel blasting construction site; A preprocessing module is used to preprocess infrared images to form training samples; A training module is used to input training samples into a neural network for training to obtain a construction worker detection model; The early warning module is used to use the construction personnel to detect whether there are workers within the safe distance of the model, and to issue an early warning if there are workers.

9. An electronic device comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, wherein: When the computer program is executed by the processor, the steps of the intelligent early warning method for tunnel blasting construction safety risks according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent early warning method for tunnel blasting construction safety risks according to any one of claims 1 to 7 are implemented.

Images