A discriminant method for the closing state of a disconnector based on point cloud data

Abstract

The present invention discloses a method for judging the closing state of a disconnector based on point cloud data, specifically: using a lidar to obtain the point cloud data of the disconnector in the monitoring area, extracting the region of interest (ROI), and obtaining the point cloud data of the conductive arm; selecting the point cloud of the target conductive arm for longitudinal slicing operation; using the Euclidean clustering segmentation algorithm to segment the sliced point cloud of the conductive arm to obtain the upper edge point cloud of the conductive arm; using the spatial line fitting algorithm based on RANSAC to fit the left and right arms of the upper edge of the conductive arm into a three-dimensional space line and calculating the included angle; calculating the closing angle threshold of the conductive arm based on the measured resistance value, and judging the closing state of the disconnector through the calculation of the fitting included angle and the threshold. The present invention realizes the judgment of the closing state of the disconnector based on the lidar, with high detection accuracy and strong anti-interference ability.

Description

Technical Field

[0001] The present invention relates to the technical field of disconnector closing state detection, and more specifically, to a method for discriminating the closing state of a disconnector based on point cloud data. Background Art

[0002] The disconnector plays the role of isolating voltage through the contact and separation of the moving and static contacts. It is an important device in the power system. Due to its long-term exposure to the outdoor environment, the disconnector may become stuck during the opening and closing processes, resulting in the situation that the disconnector cannot be opened or closed in place, leading to safety accidents. For traditional video monitoring, the judgment of the disconnector closing state mainly relies on the subjective experience of operation and maintenance personnel, and the reliability is not strong.

[0003] The existing intelligent disconnector monitoring methods mainly include the method based on pressure sensors, the method based on the principle of auxiliary contacts, etc. The monitoring method based on pressure sensor technology has problems such as signal transmission, and the monitoring method based on the principle of auxiliary contacts has problems of signal transmission in a high-voltage and strong magnetic environment and the function of the auxiliary switch. Therefore, it is necessary to develop a non-contact detection method under live conditions to realize the discrimination of the disconnector closing state. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to propose a method for discriminating the closing state of a disconnector based on point cloud data. The system uses a lidar to capture the three-dimensional point cloud image of the disconnector, and through the image processing technology based on point cloud data, realizes the accurate discrimination of the disconnector closing state, with strong anti-interference ability and wide application range.

[0005] A method for discriminating the closing state of a disconnector based on point cloud data according to the present invention includes the following steps:

[0006] Step 1: Use a lidar to obtain the point cloud data of the disconnector in the monitoring area, extract the region of interest (ROI) of the disconnector, and introduce aggregation factors τ x , τ y , τ z to obtain the best boundary tightening effect of the region of interest (ROI). Define the upper and lower limits of the xyz axes of the initial region of interest (ROI) boundary as x max , x min , y max , y min , z max , z min , and the calculation formula of the aggregation factor is:

[0007]

[0008]

[0009]

[0010] Obtain the point cloud data of the conductive arm;

[0011] Step 2: Select the point cloud of the target conductive arm for longitudinal slicing operation. The effect of longitudinal slicing is achieved by restricting the z-axis range within the region of interest (ROI) of the conductive arm. The z-axis range is obtained based on the point cloud data density of the target conductive arm, that is:

[0012]

[0013] In the formula, A is the number of point clouds received by the lidar per second, T is the frame time set by the lidar, and μ is the boundary contraction influence coefficient, and its value range is [1.45, 3.2];

[0014] Step 3: Use the Euclidean clustering segmentation algorithm to segment the point cloud of the conductive arm slice to obtain the upper edge point cloud of the conductive arm;

[0015] Step 4: Use the spatial line fitting algorithm based on RANSAC to fit the left and right arms of the upper edge of the conductive arm into a three-dimensional space line and calculate the included angle;

[0016] Step 5: Calculate the closing angle threshold of the conductive arm based on the measured resistance value, visualize and analyze the upper edge point cloud of the conductive arm, and judge whether the point cloud image is a convergent sample or a divergent sample. The convergent sample is that the included angle direction of the conductive arm in the image is downward, and the divergent sample is that the included angle direction in the image is upward. If the image is a positive sample, that is:

[0017] θ r = αe -0.0015R

[0018] In the formula, R is the measured resistance value of the disconnector contact, and α is the closing angle convergence coefficient, taking 201.25;

[0019] If the image is a negative sample, that is:

[0020] θ r = βln(R) - 1100

[0021] In the formula, R is the measured resistance value of the disconnector contact, and β is the closing angle divergence coefficient, taking 263.58;

[0022] Judge the closing state of the disconnector by calculating and discriminating the fitting included angle and the threshold. Define ε as the closing error coefficient, and its calculation formula is:

[0023]

[0024] In the formula, θ mLet ε be the included angle value of the straight-line fitting of the conductive arm. When ε < 0.3%, it indicates that the disconnector is in the fully closed state; when 0.3% ≤ ε < 1%, it indicates that the disconnector is in a state of contact jamming and a fault may occur; when ε ≥ 1%, it indicates that the disconnector is in the open state.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] The present invention uses lidar to realize the discrimination of the closing state of the disconnector, overcoming the problem of performance reduction of traditional video monitoring under low light; the present invention first longitudinally slices the point cloud of the disconnector and then performs spatial straight-line fitting based on RANSAC, optimizing the fitting effect; first distinguish positive and negative samples and then calculate the included angle threshold based on the contact resistance, improving the accuracy of discrimination. Description of the Drawings

[0027] Figure 1 It is the overall flowchart of the method for discriminating the closing state of the disconnector based on point cloud data of the present invention.

[0028] Figure 2 It is a schematic diagram of positive and negative samples of the point cloud image of the upper edge of the conductive arm. Detailed Embodiments

[0029] In order to make the technical solutions of the present invention clearer, the following further describes the present invention in detail with reference to the drawings and embodiments.

[0030] As Figure 1 shown, the present invention provides a method for discriminating the closing state of a disconnector based on point cloud data, including the following steps:

[0031] Step 1: Use lidar to obtain the point cloud data of the disconnector in the monitoring area, extract the region of interest (ROI) of the disconnector, and introduce aggregation factors τ x , τ y , τ z to obtain the best boundary tightening effect of the region of interest (ROI), and define the upper and lower limits of the xyz axes of the initial region of interest (ROI) boundary as x max , x min , y max , y min , z max , z min , and the calculation formula of the aggregation factor is:

[0032]

[0033]

[0034]

[0035] Obtain the point cloud data of the conductive arm;

[0036] Step 2: Select the target conductive arm point cloud for longitudinal slicing operation. By restricting the z-axis range within the region of interest (ROI) of the conductive arm, the effect of longitudinal slicing is achieved. The z-axis range is obtained based on the point cloud data density of the target conductive arm, that is:

[0037]

[0038] In the formula, A is the number of point clouds received by the lidar per second, T is the frame time set by the lidar, and μ is the boundary contraction influence coefficient, and its value range is [1.45, 3.2];

[0039] Step 3: Use the Euclidean clustering segmentation algorithm to segment the conductive arm sliced point cloud to obtain the upper edge point cloud of the conductive arm;

[0040] Step 4: Use the RANSAC-based spatial line fitting algorithm to fit the left and right arms of the upper edge of the conductive arm into a three-dimensional space line and calculate the included angle;

[0041] Step 5: Calculate the closing angle threshold of the conductive arm based on the measured resistance value, visualize and analyze the upper edge point cloud of the conductive arm, and determine whether the point cloud image is a convergent sample or a divergent sample. As Figure 2 shown, the convergent sample is that the included angle direction of the conductive arm in the image is downward, and the divergent sample is that the included angle direction in the image is upward. If the image is a positive sample, that is:

[0042] θ r = αe -0.0015R

[0043] In the formula, R is the measured resistance value of the disconnector contact, and α is the closing angle convergence coefficient, taking 201.25;

[0044] If the image is a negative sample, that is:

[0045] θ r = βln(R) - 1100

[0046] In the formula, R is the measured resistance value of the disconnector contact, and β is the closing angle divergence coefficient, taking 263.58;

[0047] Judge the closing state of the disconnector by calculating and discriminating the fitting included angle and the threshold. Define ε as the closing error coefficient, and its calculation formula is:

[0048]

[0049] In the formula, θ mIt is the included angle value of the straight line fitting of the conductive arm. When ε < 0.3%, it indicates that the disconnector is in the fully closed state; when 0.3% ≤ ε < 1%, it indicates that the disconnector is in the state of contact jamming and a fault may occur; when ε ≥ 1%, it indicates that the disconnector is in the open state.

[0050] To verify the credibility of this method, on-site verification is carried out on the GW4-type disconnector of a certain substation. In step 1, τ x = 2.25, τ y = 0.5, τ z = 1.5. In step 2, l z = 1.62. In this example, the point cloud image of the upper edge of the conductive arm is a positive sample. When closing, the measured resistance value is 98.2 Ω, θ r = 173.69°, the fitting angle θ m = 174.1°, ε = 0.24%, indicating that it has been fully closed.

[0051] The present invention is not limited to the above specific embodiments. Under the inspiration of the present invention, without departing from the purpose of the present invention and the scope protected by the claims, these features and embodiments can be appropriately modified to adapt to the specific working scenarios, and these all fall within the protection scope of the present invention.

Claims

1. A method for judging the closing state of a disconnector based on point cloud data, comprising the following steps: Step 1: Use a lidar to obtain the point cloud data of the disconnector in the monitoring area, extract the region of interest (ROI) of the disconnector, and obtain the point cloud data of the conductive arm; Step 2: Select the point cloud of the target conductive arm for longitudinal slicing operation; Step 3: Use the Euclidean clustering segmentation algorithm to segment the sliced point cloud of the conductive arm to obtain the upper edge point cloud of the conductive arm; Step 4: Use the spatial line fitting algorithm based on RANSAC to fit the left and right arms of the upper edge of the conductive arm into a three-dimensional space line and calculate the included angle; Step 5: Calculate the closing angle threshold of the conductive arm based on the measured resistance value, and judge the closing state of the disconnector by calculating the fitting included angle and the threshold; In the step 1, aggregation factors τ x , τ y , τ z are introduced to the x, y, and z axes respectively during the extraction of the region of interest (ROI) to obtain the best boundary tightening effect of the region of interest (ROI). The upper and lower limits of the xyz axes defining the boundary of the initial region of interest (ROI) are x max , x min , y max , y min , z max , z min , and the calculation formula for the aggregation factor is: In the said Step 2, the effect of longitudinal slicing is achieved by restricting the z-axis range within the region of interest (ROI) of the conductive arm, and the z-axis range is obtained based on the point cloud data density of the target conductive arm, that is: In the formula, A is the number of point clouds received by the lidar per second, T is the frame time set by the lidar, and μ is the boundary contraction influence coefficient, and its value range is [1.45, 3.2].

2. The isolation switch closing state discrimination method based on point cloud data according to claim 1, characterized in that In the step 5, the closing angle threshold θ of the conductive arm is obtained based on the measured resistance value r , the upper edge point cloud of the conductive arm is visualized and analyzed to determine whether the point cloud image is a convergent sample or a divergent sample. The convergent sample is that the included angle direction of the conductive arm in the image is downward, and the divergent sample is that the included angle direction in the image is upward. If the image is a positive sample, that is: θ r = αe -0.0015R In the formula, R is the measured resistance value of the disconnector contact, and α is the closing angle convergence coefficient, taking 201.25; If the image is a negative sample, that is: θ r = βln(R) - 1100 In the formula, R is the measured resistance value of the disconnector contact, and β is the closing angle divergence coefficient, taking 263.58; Judge the closing state of the disconnector by fitting the included angle and the closing angle threshold, and define ε as the closing error coefficient, and its calculation formula is: where θ m is the linear fitting included angle value of the conductive arm, and θ r is the closing angle threshold of the conductive arm; When ε < 0.3%, it means that the disconnector is in the fully closed state; when 0.3% ≤ ε < 1%, it means that the disconnector is in the state of contact jamming and a fault may occur; when ε ≥ 1%, it means that the disconnector is in the open state.

Images