Method and system for identifying accumulated water of tower footing of electric power tower by fusing multi-mode sensing technology

Abstract

The invention discloses an electric power tower footing ponding identification method and system fused with a multi-mode perception technology. The method comprises the steps that whether ponding exists at a tower footing of an electric power tower or not is identified through a sensor; and if the accumulated water is identified, acquiring an accumulated water level image at the tower footing of the power tower. The accuracy of subsequent accumulated water level recognition is improved by removing the noise in the accumulated water level image, the image is corrected, the pixel position relation before and after the image is determined on the basis of the correction coefficient and the parameter information of the image acquisition device, distortion correction of different degrees is achieved, the authenticity of the image is guaranteed, and the recognition accuracy of the accumulated water level is improved. The accuracy of subsequent accumulated water level recognition is further improved, the real-time water level height is obtained according to a preset calculation formula, the real-time water level height is compared with the preset threshold water level, if the real-time water level height exceeds the preset threshold water level, an alarm is directly given, and a worker is notified to eliminate hidden dangers in time through early warning according to the change of the real-time water level.

Description

Technical Field

[0001] The present invention relates to the technical field of water accumulation identification, and in particular to a method and system for identifying water accumulation at the base of a power tower by integrating multimodal sensing technology. Background Art

[0002] At present, there are hundreds of power plants in the Chinese market, with different sizes, mainly located in the three northern regions and the southeastern coast. This is mainly because these regions are rich in resources and densely distributed, which is conducive to the effective development of the power generation industry. At present, my country's support for power generation is very high, and the investment in power generation is also very huge. The investment in various equipment has been strengthened to ensure the hardware requirements for power generation. Many data show that my country's power generation industry has entered the stage of large-scale development. However, there are still some problems in the process of wind power generation.

[0003] In recent years, extreme weather has become more and more frequent, and generators are located in areas with strong winds. These areas are more severely affected by extreme weather. In these areas, water often accumulates, posing hidden dangers to generators. Existing water level detection and alarm devices for wind turbines are unable to promptly warn of hidden dangers caused by changes in water accumulation, resulting in hidden dangers and making it difficult to meet the needs of staff. Summary of the invention

[0004] In order to solve the above technical problems, a method and system for identifying water accumulation at the base of a power tower integrating multimodal sensing technology are provided. This technical solution solves the problem that the existing water level detection and alarm device of a wind turbine generator set proposed in the above background technology is unable to timely warn of hidden dangers caused by changes in water accumulation, thus causing hidden dangers.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] The method for identifying water accumulation at the base of a power tower by integrating multimodal sensing technology includes:

[0007] Use sensors to identify whether there is water accumulation at the base of the power tower;

[0008] If water accumulation is detected, an image of the water level at the base of the power tower is obtained;

[0009] Preprocessing the water level image, wherein the image preprocessing includes image denoising and image correction;

[0010] Analyze and identify the pre-processed images to obtain real-time height data of accumulated water level;

[0011] Compare and analyze the real-time water level height to determine whether an early warning is needed.

[0012] Preferably, the use of a sensor to identify whether there is water accumulation at the base of the power tower includes: the sensor is arranged on a water level gauge at the bottom of the base of the power tower, and the water level gauge is a sonic water level gauge.

[0013] Preferably, the image denoising specifically comprises the following steps:

[0014] Decompose the high-frequency and low-frequency information of noisy images;

[0015] The pixels of low-frequency information are processed by local pixel grouping, and the unbiased estimation of the error is used to approximate the similarity between the local pixel block and the target pixel block, and a sample set of similar local pixel blocks is obtained;

[0016] Traverse each sample set obtained, use the principal component analysis algorithm to denoise it in turn, calculate the covariance matrix, obtain the orthogonal transformation matrix, and combine it with the eigenvalue matrix to remove the dimensions containing a small amount of information in the sample set to obtain the reconstructed low-frequency components;

[0017] Decompose high-frequency information into overlapping blocks of the same size, calculate the Euclidean distance to construct similar blocks into groups, use singular value decomposition to learn the adaptive learning dictionary of each group, calculate sparse coding through the split Bregman iterative algorithm combined with convex optimization algorithm, and reconstruct the high-frequency components using sparse coding and adaptive learning dictionary;

[0018] The inverse wavelet transform aggregates the high-frequency components and the low-frequency components to obtain the denoised image.

[0019] Preferably, the image correction specifically includes the following steps:

[0020] Acquire a standard image preset in the system, and collect parameter information of the standard image;

[0021] According to the correction coefficient preset in the system and the parameter information, the pixel mapping relationship between the water level image and the standard image is determined, and the resolution of the water level image and the standard image is the same, according to the pixel mapping relationship;

[0022] The pixel value of the first pixel in the standard image is determined as the pixel value of the second pixel in the water level image to obtain a corrected second image, and the second pixel is any pixel in the water level image, and the second pixel corresponds to the first pixel in the pixel mapping relationship.

[0023] Preferably, the analysis and identification of the pre-processed image specifically comprises the following steps:

[0024] Performing contour detection on the water level image to determine the water level line in the water level image;

[0025] Acquire calibration information of a plurality of preset water point locations, the calibration information being a correspondence between a pixel point corresponding to each preset water point location in the water level image and a water level height of the preset water point location;

[0026] Determine the water level height based on the water level line and the calibration information of multiple preset water points.

[0027] Preferably, the step of obtaining calibration information of a plurality of preset water position points specifically comprises the following steps:

[0028] Determine scene information of the water level image;

[0029] According to the scene information, calling pre-stored calibration information of multiple preset water point locations corresponding to the scene information;

[0030] The plurality of preset water level points are a maximum water level point and a minimum water level point.

[0031] Preferably, the step of determining the water level height according to the calibration information of the water level line and a plurality of preset water level points specifically comprises the following steps:

[0032] In the water level image, determine the vertical pixel difference between the water level line and the pixel point at the highest water point as the first pixel value, and the vertical pixel difference between the water level line and the pixel point at the lowest water point as the second pixel value;

[0033] The water level height is calculated according to the calibration information of the highest water point, the calibration information of the lowest water point, the first pixel value and the second pixel value.

[0034] Preferably, the calculation formula for the water level is:

[0035]

[0036] Wherein, the highest water level point of the image is A, the lowest water level point is B, the vertical pixel value of the current water level line from position A is m, and the vertical pixel value of the current water level line from position B is n.

[0037] Preferably, the comparing and analyzing the obtained real-time water level height to determine whether an early warning needs to be issued specifically comprises the following steps:

[0038] Compare the obtained real-time water level height with the threshold value pre-set in the system;

[0039] If the real-time water level is greater than or equal to the preset threshold, an early warning message is issued;

[0040] If the real-time water level is less than the preset threshold, no action will be taken.

[0041] The power tower base water accumulation recognition system integrates multi-modal sensing technology, including:

[0042] An identification module, the identification module is used to identify whether there is water accumulation at the base of the power tower through a sensor;

[0043] A shooting module, wherein if the sensor identifies the presence of accumulated water, the shooting module is used to obtain an image of the accumulated water level at the base of the power tower;

[0044] An image processing module, wherein the image processing module internally integrates a first image processing unit and a second image processing unit;

[0045] A first image processing unit, wherein the first image processing unit is used to denoise the image;

[0046] A second image processing unit, wherein the second image processing unit is used to correct the image;

[0047] A water level height calculation module, which is used to analyze and identify the pre-processed image to obtain real-time height data of the accumulated water level;

[0048] The early warning module is used to compare and analyze the obtained real-time water level height to determine whether an early warning needs to be issued.

[0049] Compared with the prior art, the present invention provides a method and system for identifying water accumulation at the base of a power tower by integrating multimodal sensing technology, which has the following beneficial effects:

[0050] The present invention installs an acoustic wave water level meter at the base of the power tower. The acoustic wave water level meter can detect whether there is accumulated water. If so, an image of the accumulated water level is directly collected, and the noise in the accumulated water level image is removed to improve the accuracy of subsequent recognition of the accumulated water level. Then, the image is corrected, and the position relationship of pixels before and after the image is determined based on the correction coefficient and parameter information of the image acquisition device to achieve different degrees of distortion correction, ensure the authenticity of the image, and further improve the accuracy of subsequent recognition of the accumulated water level. Then, the water level in the image is identified through image processing technology, and the real-time water level height is obtained according to a preset calculation formula. The real-time water level height is compared with a preset threshold water level. If it exceeds the threshold, an alarm is directly issued. In life, the value of the threshold water level changes with factors such as seasons and is not a fixed value. The present invention notifies the staff to eliminate hidden dangers in time by issuing an early warning of the change in the real-time water level, and the water level change can be stored in the system for later maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A schematic diagram of a method for identifying water accumulation at the base of a power tower in the present invention;

[0052] Figure 2 A schematic diagram of a method for denoising an image in the present invention;

[0053] Figure 3 A schematic diagram of a method for correcting an image in the present invention;

[0054] Figure 4 A schematic diagram of a method for analyzing and identifying a pre-processed image in the present invention;

[0055] Figure 5 A schematic diagram of a method for obtaining calibration information of multiple preset water position points in the present invention. DETAILED DESCRIPTION

[0056] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0057] Example 1

[0058] Please refer to Figure 1-Figure 5 As shown, the method for identifying water accumulation at the base of a power tower integrating multimodal sensing technology includes:

[0059] Identifying whether there is water accumulation at the base of the power tower by using a sensor, and identifying whether there is water accumulation at the base of the power tower by using a sensor, including: the sensor is arranged on a water level meter at the bottom of the base of the power tower, and the water level meter is a sonic water level meter;

[0060] If water accumulation is detected, an image of the water level at the base of the power tower is obtained;

[0061] Preprocessing the water level image, including image denoising and image correction;

[0062] Analyze and identify the pre-processed images to obtain real-time height data of accumulated water level;

[0063] Compare and analyze the real-time water level height to determine whether an early warning is needed.

[0064] It can be understood by those skilled in the art that the present invention installs an acoustic water level meter at the base of the power tower, and the acoustic water level meter can detect whether there is water accumulation. If there is, an image of the water accumulation level is directly collected, and the noise in the water accumulation level image is removed to improve the accuracy of subsequent water accumulation level recognition. Then, the image is corrected, and the pixel position relationship before and after the image is determined based on the correction coefficient and the parameter information of the image acquisition device to achieve different degrees of distortion correction, ensure the authenticity of the image, and further improve the accuracy of subsequent water accumulation level recognition. Then, the water level in the image is identified through image processing technology, and the real-time water level height is obtained according to a preset calculation formula. The real-time water level height is compared with the preset threshold water level. If it exceeds, an alarm is directly issued. In life, the value of the threshold water level changes with factors such as seasons and is not a fixed value. The present invention notifies the staff to eliminate hidden dangers in time by issuing an early warning of the change in the real-time water level, and the water level change can be stored in the system for later maintenance.

[0065] Image denoising specifically includes the following steps:

[0066] Decompose the high-frequency and low-frequency information of noisy images;

[0067] The pixels of low-frequency information are processed by local pixel grouping, and the unbiased estimation of the error is used to approximate the similarity between the local pixel block and the target pixel block, and a sample set of similar local pixel blocks is obtained;

[0068] Traverse each sample set obtained, use the principal component analysis algorithm to denoise it in turn, calculate the covariance matrix, obtain the orthogonal transformation matrix, and combine it with the eigenvalue matrix to remove the dimensions containing a small amount of information in the sample set to obtain the reconstructed low-frequency components;

[0069] Decompose high-frequency information into overlapping blocks of the same size, calculate the Euclidean distance to construct similar blocks into groups, use singular value decomposition to learn the adaptive learning dictionary of each group, calculate sparse coding through the split Bregman iterative algorithm combined with convex optimization algorithm, and reconstruct the high-frequency components using sparse coding and adaptive learning dictionary;

[0070] The inverse wavelet transform aggregates the high-frequency components and the low-frequency components to obtain the denoised image.

[0071] Image correction specifically includes the following steps:

[0072] Obtaining the preset standard image in the system and the parameter information of acquiring the standard image;

[0073] According to the correction coefficient and parameter information pre-set in the system, the pixel mapping relationship between the water level image and the standard image is determined, and the resolution of the water level image and the standard image is the same, according to the pixel mapping relationship;

[0074] The pixel value of the first pixel in the standard image is determined as the pixel value of the second pixel in the water level image to obtain a corrected second image, and the second pixel is any pixel in the water level image, and the second pixel corresponds to the first pixel in the pixel mapping relationship.

[0075] The analysis and recognition of the preprocessed image specifically includes the following steps:

[0076] Performing contour detection on the water level image to determine the water level line in the water level image;

[0077] Acquire calibration information of a plurality of preset water point locations, the calibration information being a correspondence between a pixel point corresponding to each preset water point location in the water level image and a water level height of the preset water point location;

[0078] Determine the water level height based on the water level line and the calibration information of multiple preset water points.

[0079] Obtaining calibration information of multiple preset water level points specifically includes the following steps:

[0080] Determine scene information of the water level image;

[0081] According to the scene information, calling pre-stored calibration information of a plurality of preset water point locations corresponding to the scene information;

[0082] The plurality of preset water level points are a maximum water level point and a minimum water level point.

[0083] According to the calibration information of the water level line and multiple preset water point points, determining the water level height specifically includes the following steps:

[0084] In the water level image, determine the vertical pixel difference between the water level line and the pixel point at the highest water point as the first pixel value, and the vertical pixel difference between the water level line and the pixel point at the lowest water point as the second pixel value;

[0085] The water level height is calculated according to the calibration information of the highest water point, the calibration information of the lowest water point, the first pixel value and the second pixel value.

[0086] The water level calculation formula is:

[0087]

[0088] Wherein, the highest water level point of the image is A, the lowest water level point is B, the vertical pixel value of the current water level line from position A is m, and the vertical pixel value of the current water level line from position B is n.

[0089] Comparing and analyzing the obtained real-time water level height to determine whether an early warning is needed includes the following steps:

[0090] Compare the obtained real-time water level height with the threshold value pre-set in the system;

[0091] If the real-time water level is greater than or equal to the preset threshold, an early warning message is issued;

[0092] If the real-time water level is less than the preset threshold, no action will be taken.

[0093] The power tower base water accumulation recognition system integrates multi-modal sensing technology, including:

[0094] An identification module, which is used to identify whether there is water accumulation at the base of the power tower through a sensor;

[0095] A shooting module, which is used to obtain an image of the water level at the base of the power tower if the sensor recognizes that there is water accumulation;

[0096] An image processing module, wherein the image processing module internally integrates a first image processing unit and a second image processing unit;

[0097] A first image processing unit, wherein the first image processing unit is used to denoise the image;

[0098] A second image processing unit, wherein the second image processing unit is used to correct the image;

[0099] A water level height calculation module, which is used to analyze and identify the pre-processed image to obtain real-time height data of the accumulated water level;

[0100] The early warning module is used to compare and analyze the obtained real-time water level height to determine whether an early warning needs to be issued.

[0101] In summary, the present invention installs an acoustic wave water level meter at the base of the power tower. The acoustic wave water level meter can detect whether there is water accumulation. If there is, the image of the water accumulation level is directly collected, and the noise in the water accumulation level image is removed to improve the accuracy of subsequent water accumulation level recognition. Then, the image is corrected, and the pixel position relationship before and after the image is determined based on the correction coefficient and the parameter information of the image acquisition device to achieve different degrees of distortion correction, ensure the authenticity of the image, and further improve the accuracy of subsequent water accumulation level recognition. Then, the water level in the image is identified through image processing technology, and the real-time water level height is obtained according to a preset calculation formula. The real-time water level height is compared with the preset threshold water level. If it exceeds, an alarm is directly issued. In life, the value of the threshold water level changes with factors such as seasons and is not a fixed value. The present invention notifies the staff to eliminate hidden dangers in time by issuing an early warning of the change in the real-time water level, and the water level change can be stored in the system for later maintenance.

[0102] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for identifying water accumulation at the base of a power tower by integrating multimodal sensing technology, characterized in that: include: Use sensors to identify whether there is water accumulation at the base of the power tower; If water accumulation is detected, an image of the water level at the base of the power tower is obtained; Preprocessing the water level image, wherein the image preprocessing includes image denoising and image correction; Analyze and identify the pre-processed images to obtain real-time height data of accumulated water level; Compare and analyze the real-time water level height to determine whether an early warning is needed.

2. The method for identifying water accumulation at the foundation of a power tower by integrating multimodal sensing technology according to claim 1 is characterized in that: The method uses a sensor to identify whether there is water accumulation at the base of a power tower, including: the sensor is arranged on a water level meter at the bottom of the base of the power tower, and the water level meter is a sonic water level meter.

3. The method for identifying water accumulation at the foundation of a power tower by integrating multimodal sensing technology according to claim 1 is characterized in that: The image denoising specifically comprises the following steps: Decompose the high-frequency and low-frequency information of noisy images; The pixels of low-frequency information are processed by local pixel grouping, and the unbiased estimation of the error is used to approximate the similarity between the local pixel block and the target pixel block, and a sample set of similar local pixel blocks is obtained; Traverse each sample set obtained, use the principal component analysis algorithm to denoise it in turn, calculate the covariance matrix, obtain the orthogonal transformation matrix, and combine it with the eigenvalue matrix to remove the dimensions containing a small amount of information in the sample set to obtain the reconstructed low-frequency components; Decompose high-frequency information into overlapping blocks of the same size, calculate the Euclidean distance to construct similar blocks into groups, use singular value decomposition to learn the adaptive learning dictionary of each group, calculate sparse coding through the split Bregman iterative algorithm combined with convex optimization algorithm, and reconstruct the high-frequency components using sparse coding and adaptive learning dictionary; The inverse wavelet transform aggregates the high-frequency components and the low-frequency components to obtain the denoised image.

4. The method for identifying water accumulation at the foundation of a power tower by integrating multimodal sensing technology according to claim 3 is characterized in that: The correction of the image specifically comprises the following steps: Acquire a standard image preset in the system, and collect parameter information of the standard image; According to the correction coefficient preset in the system and the parameter information, the pixel mapping relationship between the water level image and the standard image is determined, and the resolution of the water level image and the standard image is the same, according to the pixel mapping relationship; The pixel value of the first pixel in the standard image is determined as the pixel value of the second pixel in the water level image to obtain a corrected second image, and the second pixel is any pixel in the water level image, and the second pixel corresponds to the first pixel in the pixel mapping relationship.

5. The method for identifying water accumulation at the foundation of a power tower by integrating multimodal sensing technology according to claim 4 is characterized in that: The analysis and identification of the pre-processed image specifically comprises the following steps: Performing contour detection on the water level image to determine the water level line in the water level image; Acquire calibration information of a plurality of preset water point locations, the calibration information being a correspondence between a pixel point corresponding to each preset water point location in the water level image and a water level height of the preset water point location; Determine the water level height based on the water level line and the calibration information of multiple preset water points.

6. The method for identifying water accumulation at the base of a power tower incorporating multimodal sensing technology according to claim 5 is characterized in that: The step of obtaining calibration information of a plurality of preset water position points specifically comprises the following steps: Determine scene information of the water level image; According to the scene information, calling pre-stored calibration information of a plurality of preset water point locations corresponding to the scene information; The plurality of preset water level points are a maximum water level point and a minimum water level point.

7. The method for identifying water accumulation at the base of a power tower incorporating multimodal sensing technology according to claim 6 is characterized in that: Determining the water level height according to the calibration information of the water level line and multiple preset water points specifically includes the following steps: In the water level image, determine the vertical pixel difference between the water level line and the pixel point at the highest water point as the first pixel value, and the vertical pixel difference between the water level line and the pixel point at the lowest water point as the second pixel value; The water level height is calculated according to the calibration information of the highest water point, the calibration information of the lowest water point, the first pixel value and the second pixel value.

8. The method for identifying water accumulation at the foundation of a power tower by integrating multimodal sensing technology according to claim 7 is characterized in that: The calculation formula of the water level is: Wherein, the highest water level point of the image is A, the lowest water level point is B, the vertical pixel value of the current water level line from position A is m, and the vertical pixel value of the current water level line from position B is n.

9. The method for identifying water accumulation at the foundation of a power tower by integrating multimodal sensing technology according to claim 8 is characterized in that: The comparison and analysis of the obtained real-time water level height to determine whether an early warning needs to be issued specifically includes the following steps: Compare the obtained real-time water level height with the threshold value pre-set in the system; If the real-time water level is greater than or equal to the preset threshold, an early warning message is issued; If the real-time water level is less than the preset threshold, no action will be taken.

10. A system for identifying water accumulation at the base of a power tower integrating multimodal sensing technology, which is used to implement a method for identifying water accumulation at the base of a power tower integrating multimodal sensing technology as described in any one of claims 1 to 9, characterized in that: include: An identification module, the identification module is used to identify whether there is water accumulation at the base of the power tower through a sensor; A shooting module, wherein if the sensor identifies the presence of accumulated water, the shooting module is used to obtain an image of the accumulated water level at the base of the power tower; An image processing module, wherein the image processing module internally integrates a first image processing unit and a second image processing unit; A first image processing unit, wherein the first image processing unit is used to denoise the image; A second image processing unit, wherein the second image processing unit is used to correct the image; A water level height calculation module, which is used to analyze and identify the pre-processed image to obtain real-time height data of the accumulated water level; The early warning module is used to compare and analyze the obtained real-time water level height to determine whether an early warning needs to be issued.

Images