Transformer substation inspection method and device, electronic equipment and storage medium

Abstract

The invention provides a substation inspection method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring current transformer insulator position data; determining an inspection route according to the current transformer insulator position data; according to the inspection route, controlling the inspection robot to inspect and collect three-dimensional point cloud data of the current transformer insulator; constructing a three-dimensional model corresponding to the current transformer insulator according to the three-dimensional point cloud data; and comparing the three-dimensional model with a preset standard model, and determining an inspection result. According to the invention, the technical problems of low accuracy and poor adaptability of a transformer substation inspection method provided in the related art are solved.

Description

Technical Field

[0001] The present disclosure relates to the fields of smart grid and robotics technology, and in particular to a substation inspection method, device, electronic equipment and storage medium. Background Art

[0002] In the power system, the status monitoring of the current transformer insulators in the substation is crucial to ensure the safe operation of the power equipment. Traditional manual inspection methods have problems such as low efficiency, high cost, and the detection results are affected by subjective factors. Especially in the complex and changeable outdoor environment, it is difficult to effectively identify the tiny cracks, damage or dirt on the surface of the insulator, which affects the stability and reliability of the power system. With the rapid development of robotics and three-dimensional point cloud processing technology, intelligent inspection robots have begun to be used in power equipment monitoring, but the current technology still faces the problem of insufficient computational efficiency when processing large-scale point cloud data and inaccurate feature extraction when processing point cloud data with complex shapes, uneven surfaces or noise.

[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present disclosure provide a substation inspection method, device, electronic device and storage medium to at least solve the technical problems of low accuracy and poor adaptability of the substation inspection method provided in the related art.

[0005] According to one aspect of an embodiment of the present disclosure, a substation inspection method is provided, including: obtaining position data of a current transformer insulator; determining an inspection route based on the position data of the current transformer insulator; controlling an inspection robot to perform inspections and collect three-dimensional point cloud data of the current transformer insulator based on the inspection route; constructing a three-dimensional model corresponding to the current transformer insulator based on the three-dimensional point cloud data; and comparing the three-dimensional model with a preset standard model to determine an inspection result.

[0006] Optionally, a three-dimensional model corresponding to the current transformer insulator is constructed based on the three-dimensional point cloud data, including: using a preset alignment algorithm to align the three-dimensional point cloud data to obtain target point cloud data; using a preset reconstruction algorithm to perform surface reconstruction on the target point cloud data to obtain a three-dimensional model.

[0007] Optionally, the preset registration algorithm includes a sampling consistency initial registration algorithm and an iterative closest point algorithm. The preset registration algorithm is used to align the three-dimensional point cloud data to obtain target point cloud data, including: using the sampling consistency initial registration algorithm to perform a first registration operation on the three-dimensional point cloud data to obtain a first registration result; using the iterative closest point algorithm to perform a second registration operation on the first registration result to obtain target point cloud data.

[0008] Optionally, a first registration operation is performed on the three-dimensional point cloud data using a sampling consistency initial registration algorithm to obtain a first registration result, including: selecting first three-dimensional point cloud data and second three-dimensional point cloud data from the three-dimensional point cloud data, wherein the first three-dimensional point cloud data is data collected by a first acquisition frame, and the second three-dimensional point cloud data is data collected by a second acquisition frame, and the first acquisition frame and the second acquisition frame are continuous frames; determining multiple sample points from the first three-dimensional data; determining multiple similar points from the second three-dimensional point cloud data based on the multiple sample points and preset similarity conditions; determining an initial point cloud transformation matrix based on the multiple sample points and the multiple similar points; and performing a first registration operation on the three-dimensional point cloud data based on the initial point cloud transformation matrix to obtain a first registration result.

[0009] Optionally, an iterative closest point algorithm is used to perform a second registration operation on the first registration result to obtain target point cloud data, including: determining target point cloud data and reference point cloud data based on the three-dimensional point cloud data, wherein the target point cloud data is the point cloud data corresponding to the first acquisition frame, the reference point cloud data is the point cloud data corresponding to the second acquisition frame, and the first acquisition frame and the second acquisition frame are continuous frames; determining a rotation matrix and a translation vector based on the target point cloud data and the reference point cloud data; determining matching point pairs in the target point cloud data and the reference point cloud data based on the rotation matrix and the translation vector; determining a target point cloud transformation matrix based on the matching point pairs; and performing a second registration operation on the first registration result based on the target point cloud transformation matrix to obtain the target point cloud data.

[0010] Optionally, a preset reconstruction algorithm is used to perform surface reconstruction on the target point cloud data to obtain a three-dimensional model, including: constructing a vector field based on the normal vector of the target point cloud data; determining the isosurface of the indicator function based on the vector field and a preset indicator function; and post-processing the isosurface according to a preset post-processing strategy to obtain a three-dimensional model.

[0011] According to another aspect of the embodiment of the present disclosure, a substation inspection device is also provided, including: an acquisition module for acquiring position data of current transformer insulators; a determination module for determining an inspection route based on the position data of the current transformer insulators; a collection module for controlling an inspection robot to perform inspections and collect three-dimensional point cloud data of current transformer insulators according to the inspection route; a construction module for constructing a three-dimensional model corresponding to the current transformer insulator based on the three-dimensional point cloud data; and a comparison module for comparing the three-dimensional model with a preset standard model to determine the inspection result.

[0012] Optionally, the construction module is also used to: use a preset registration algorithm to align the three-dimensional point cloud data to obtain target point cloud data; use a preset reconstruction algorithm to perform surface reconstruction on the target point cloud data to obtain a three-dimensional model.

[0013] Optionally, the construction module is also used to: use a sampling consistency initial registration algorithm to perform a first registration operation on the three-dimensional point cloud data to obtain a first registration result; use an iterative closest point algorithm to perform a second registration operation on the first registration result to obtain target point cloud data.

[0014] Optionally, the construction module is also used to: select first three-dimensional point cloud data and second three-dimensional point cloud data from the three-dimensional point cloud data, wherein the first three-dimensional point cloud data is data collected by a first acquisition frame, the second three-dimensional point cloud data is data collected by a second acquisition frame, and the first acquisition frame and the second acquisition frame are continuous frames; determine multiple sample points from the first three-dimensional data; determine multiple similar points from the second three-dimensional point cloud data based on the multiple sample points and preset similarity conditions; determine an initial point cloud transformation matrix based on the multiple sample points and the multiple similar points; and perform a first registration operation on the three-dimensional point cloud data based on the initial point cloud transformation matrix to obtain a first registration result.

[0015] Optionally, the construction module is also used to: determine target point cloud data and reference point cloud data based on the three-dimensional point cloud data, wherein the target point cloud data is the point cloud data corresponding to the first acquisition frame, the reference point cloud data is the point cloud data corresponding to the second acquisition frame, and the first acquisition frame and the second acquisition frame are continuous frames; determine the rotation matrix and the translation vector based on the target point cloud data and the reference point cloud data; determine the matching point pairs in the target point cloud data and the reference point cloud data based on the rotation matrix and the translation vector; determine the target point cloud transformation matrix based on the matching point pairs; and perform a second registration operation on the first registration result based on the target point cloud transformation matrix to obtain the target point cloud data.

[0016] Optionally, the construction module is also used to: construct a vector field based on the normal vector of the target point cloud data; determine the isosurface of the indicator function based on the vector field and the preset indicator function; and post-process the isosurface according to a preset post-processing strategy to obtain a three-dimensional model.

[0017] According to another aspect of the embodiments of the present disclosure, an electronic device is further provided, including: a memory storing an executable program; and a processor for running the program, wherein the method in each embodiment of the present disclosure is executed when the program is running.

[0018] According to another aspect of an embodiment of the present disclosure, a computer-readable storage medium is further provided. The computer-readable storage medium includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in various embodiments of the present disclosure.

[0019] According to another aspect of the embodiments of the present disclosure, a computer program product is also provided, including a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present disclosure is implemented.

[0020] In the disclosed embodiment, the inspection robot is controlled to perform inspections and collect three-dimensional point cloud data of current transformer insulators, and then a three-dimensional model corresponding to the current transformer insulator is constructed based on the three-dimensional point cloud data. Finally, the three-dimensional model is compared with a preset standard model to determine the inspection result, thereby achieving the purpose of comprehensive and automated monitoring of the status of current transformer insulators in the substation, thereby achieving the technical effect of improving the accuracy and adaptability of substation inspections, and thus solving the technical problems of low accuracy and poor adaptability of the substation inspection method provided in the related art. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings described herein are used to provide a further understanding of the present disclosure and constitute a part of the present disclosure. The illustrative embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation on the present disclosure. In the drawings:

[0022] Figure 1 is a flow chart of a substation inspection method according to one embodiment of the present disclosure;

[0023] Figure 2 is a schematic diagram of a point cloud data registration result according to one embodiment of the present disclosure;

[0024] Figure 3 is a schematic diagram of a surface reconstruction method according to one embodiment of the present disclosure;

[0025] Figure 4 is a schematic diagram of a three-dimensional model according to one embodiment of the present disclosure;

[0026] Figure 5 It is a structural block diagram of a substation inspection device according to one embodiment of the present disclosure. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the scheme of the present disclosure, the technical scheme in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present disclosure.

[0028] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0029] In the power system, the status monitoring of the current transformer insulators in the substation is crucial to ensure the safe operation of the power equipment. Traditional manual inspection methods have problems such as low efficiency, high cost, and the detection results are affected by subjective factors. Especially in the complex and changeable outdoor environment, it is difficult to effectively identify the tiny cracks, damage or dirt on the surface of the insulator, which affects the stability and reliability of the power system. With the rapid development of robotics and three-dimensional point cloud processing technology, intelligent inspection robots have begun to be used in power equipment monitoring, but the current technology still faces the problem of insufficient computational efficiency when processing large-scale point cloud data and inaccurate feature extraction when processing point cloud data with complex shapes, uneven surfaces or noise.

[0030] Specifically, insulators not only play the role of electrical isolation, prevent accidental current leakage, and protect the safety of equipment and personnel, but also assume the function of mechanical support to ensure the stable operation of current transformers. Through three-dimensional dynamic modeling technology, it is possible to accurately identify tiny defects on the surface of insulators, monitor their size changes, analyze the impact of environmental factors (such as temperature, humidity, and chemical corrosion) on the performance of insulators, optimize the installation location, and support regular inspection and maintenance decisions of equipment through intelligent algorithms, and provide visual maintenance records, thereby significantly improving the efficiency and safety of equipment management. However, the current technology does not perform well in terms of density and uniformity when dealing with complex and changeable outdoor environments, such as changes in light, rain, dust, and temperature differences, resulting in unsatisfactory three-dimensional modeling results. In addition, for point cloud data with complex shapes, uneven surfaces, or noise, the existing feature extraction algorithms have low accuracy and may extract erroneous features, thereby affecting the analysis accuracy. Furthermore, with the expansion of the scale of the power system and the increase in equipment, some high-precision algorithms have too long a calculation time when processing large-scale data, which limits the real-time application capability of the system and makes it difficult to adapt to the needs of rapid inspection and maintenance in the power industry.

[0031] According to an embodiment of the present disclosure, a method embodiment of a substation inspection method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0032] The method embodiment can be executed in an electronic device or a similar computing device including a memory and a processor. Taking running on a computer terminal as an example, the computer terminal may include one or more processors (processors may include but are not limited to central processing units (CPU), graphics processing units (GPU), digital signal processing (DSP) chips, microprocessors (MCU), programmable logic devices (Field Programmable Gate Array, FPGA), neural network processors (Neural-network Processor Unit, NPU), tensor processors (TensorProcessing Unit, TPU), artificial intelligence (Artificial Intelligence, AI) type processors and other processing devices) and a memory for storing data. Optionally, the above-mentioned computer terminal may also include a transmission device, an input and output device, and a display device for communication functions. It can be understood by those of ordinary skill in the art that the above-mentioned structural description is only for illustration, and it does not limit the structure of the above-mentioned computer terminal. For example, the computer terminal may also include more or fewer components than the above-mentioned structural description, or have a configuration different from the above-mentioned structural description.

[0033] The memory can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the substation inspection method in the embodiment of the present disclosure. The processor executes various functional applications and data processing by running the computer program stored in the memory, that is, the above-mentioned substation inspection method is realized. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories can be connected to the mobile terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0034] The transmission device is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device can be a radio frequency (Radio Frequency, referred to as RF) module, which is used to communicate with the Internet wirelessly.

[0035] The display device may be, for example, a touch screen type liquid crystal display (LCD) and a touch display (also referred to as a "touch screen" or "touch display screen"). The liquid crystal display may enable a user to interact with a user interface of a mobile terminal. In some embodiments, the mobile terminal has a graphical user interface (GUI), and a user may interact with the GUI by finger contacts and / or gestures on a touch-sensitive surface, wherein the human-computer interaction functions here may optionally include the following interactions: creating web pages, drawing, word processing, making electronic documents, games, video conferencing, instant messaging, sending and receiving emails, call interfaces, playing digital videos, playing digital music and / or web browsing, etc. The executable instructions for executing the above human-computer interaction functions are configured / stored in a computer program product or readable storage medium executable by one or more processors.

[0036] Figure 1 is a flow chart of a substation inspection method according to one embodiment of the present disclosure, such as Figure 1 As shown, the method comprises the following steps:

[0037] Step S11, obtaining current transformer insulator position data;

[0038] Step S12, determining an inspection route according to the current transformer insulator position data;

[0039] Step S13, according to the inspection route, controlling the inspection robot to perform inspection and collect three-dimensional point cloud data of the current transformer insulator;

[0040] Step S14, constructing a three-dimensional model corresponding to the current transformer insulator according to the three-dimensional point cloud data;

[0041] Step S15, comparing the three-dimensional model with a preset standard model to determine the inspection result.

[0042] The above-mentioned current transformer insulator refers to a component used for electrical isolation between the high-voltage side and the low-voltage side or the ground of the current transformer in the power system substation. It is mainly made of insulating materials (such as ceramics, glass or composite materials) and is used to support the conductor part of the current transformer while preventing current from flowing in unexpected paths to ensure the safe operation of power equipment.

[0043] The above-mentioned 3D point cloud data refers to a large number of 3D coordinate points on the surface of the current transformer insulator collected by the 3D laser scanner or other 3D imaging equipment carried by the inspection robot during the intelligent inspection process. Specifically, the 3D point cloud data not only contains the geometric information of the insulator surface contour, but also includes the reflection intensity or color information of each point, so as to fully and meticulously reflect the shape, structure and surface state of the insulator.

[0044] The above-mentioned preset standard model refers to a three-dimensional model of the current transformer insulator in an ideal state or a factory state. The model is usually constructed based on the design drawings or standard specifications of the insulator and includes the ideal geometric shape, size and surface features of the insulator.

[0045] For example, it is assumed that during the intelligent inspection process of the substation, the inspection robot is deployed in a high-voltage area, and the goal is to check the status of multiple groups of current transformer insulators. The inspection robot accurately obtains the three-dimensional coordinate information of the current transformer insulator and determines the inspection route through its high-precision positioning system, such as integrated GPS and visual positioning technology. When the inspection robot follows the planned route and moves to the vicinity of each group of current transformer insulators, the three-dimensional laser scanner equipped inside it starts working, quickly collects the three-dimensional point cloud data of the insulator surface, and sends it to the back-end data processing system. Subsequently, the back-end data processing system constructs a three-dimensional model corresponding to the current transformer insulator based on the three-dimensional point cloud data obtained above. After the three-dimensional model is constructed, the three-dimensional model is compared with the preset standard model of the current transformer insulator, so that details that do not conform to the standard model can be identified, such as cracks, wear, structural deformation or surface contamination, etc., so as to provide instant and accurate inspection feedback to the substation operation and maintenance personnel, thereby ensuring the safe operation of the power equipment.

[0046] Based on the above steps S11 to S15, the inspection robot is controlled to perform inspections and collect three-dimensional point cloud data of current transformer insulators, and then a three-dimensional model corresponding to the current transformer insulator is constructed according to the three-dimensional point cloud data. Finally, the three-dimensional model is compared with the preset standard model to determine the inspection result, thereby achieving the purpose of comprehensive and automated monitoring of the status of current transformer insulators in the substation, thereby achieving the technical effect of improving the accuracy and adaptability of substation inspections, and thus solving the technical problems of low accuracy and poor adaptability of the substation inspection method provided in the relevant technology.

[0047] Optionally, in step S14, a three-dimensional model corresponding to the current transformer insulator is constructed according to the three-dimensional point cloud data, including:

[0048] Step S141, using a preset registration algorithm to register the three-dimensional point cloud data to obtain target point cloud data;

[0049] Step S142, using a preset reconstruction algorithm, performing surface reconstruction on the target point cloud data to obtain a three-dimensional model.

[0050] The above-mentioned preset registration algorithm refers to an algorithm used to spatially align three-dimensional point cloud data, and its purpose is to eliminate the relative position and posture differences between different point cloud data sets, ensure that data from different time points or perspectives can be accurately matched, and form consistent target point cloud data. The preset registration algorithm includes but is not limited to the sampling consistency initial registration algorithm, the iterative closest point algorithm, the feature matching algorithm, the rigid body transformation estimation algorithm, and the registration method based on point cloud features. The above algorithms can achieve accurate alignment between data sets through iterative optimization or feature matching.

[0051] The above-mentioned preset reconstruction algorithm refers to an algorithm that converts the aligned target point cloud data into a three-dimensional surface model, including but not limited to a Poisson surface reconstruction algorithm, a radial basis function interpolation method, a triangulation algorithm, and a point cloud-based surface growth algorithm. The above-mentioned algorithm can construct a continuous and smooth three-dimensional model through interpolation, fitting or surface growth, thereby realizing accurate reconstruction of the surface structure of the object.

[0052] For example, after scanning the current transformer insulators in the substation multiple times, the inspection robot collected a large amount of 3D point cloud data obtained from different angles and time points. To ensure that the acquired 3D point cloud data can be seamlessly integrated to form a comprehensive and accurate 3D model of the insulator, the 3D point cloud data is first registered using the iterative closest point algorithm, and then the Poisson surface reconstruction algorithm is used to convert the target point cloud data into a 3D surface model.

[0053] Based on the above optional embodiment, the three-dimensional point cloud data is registered using a preset registration algorithm, which overcomes the position deviation, posture difference and noise interference in the original point cloud data, and ensures that the data acquired at different time points or different viewing angles can be accurately aligned to form highly consistent target point cloud data. At the same time, using the preset reconstruction algorithm, a continuous and smooth three-dimensional surface model can be constructed, accurately reproducing the surface structure and morphological characteristics of the current transformer insulator, thereby improving the accuracy of substation inspection.

[0054] Optionally, in step S141, the preset registration algorithm includes a sampling consistency initial registration algorithm and an iterative closest point algorithm, and the preset registration algorithm is used to register the three-dimensional point cloud data to obtain the target point cloud data, including:

[0055] Step S1411, performing a first registration operation on the three-dimensional point cloud data using a sampling consistency initial registration algorithm to obtain a first registration result;

[0056] Step S1412: using an iterative closest point algorithm, perform a second registration operation on the first registration result to obtain target point cloud data.

[0057] The sampling consistency initial registration algorithm is an algorithm that efficiently processes 3D point cloud data and achieves its initial alignment. The uniqueness of this algorithm lies in combining the rotational invariance characteristics of point cloud data with two key methods - greedy initial alignment and sampling consistency algorithm to achieve robust registration in complex data sets, thereby being able to eliminate incorrect matching point pairs while maintaining computational efficiency.

[0058] The above-mentioned iterative closest point algorithm refers to minimizing the distance between two sets of point clouds in an iterative manner to achieve accurate registration. The basic idea of ​​the iterative closest point is to find the best rigid body transformation (including rotation and translation) between the target point cloud and the reference point cloud so that the target point cloud and the reference point cloud overlap as much as possible.

[0059] Specifically, after acquiring the 3D point cloud data, the sampling consistency initial registration algorithm is first used to perform an initial registration operation on the 3D point cloud data to eliminate incorrect matches. Furthermore, based on the iterative closest point algorithm, the 3D point cloud data after the initial registration is accurately registered to ensure the accuracy of the registration result.

[0060] Based on the above optional embodiments, the three-dimensional point cloud data is preliminarily and accurately registered using the sampling consistency initial registration algorithm and the iterative closest point algorithm, which can not only speed up data processing and reduce the consumption of computing resources, but also improve the accuracy and reliability of the model.

[0061] Optionally, in step S1411, a first registration operation is performed on the three-dimensional point cloud data using a sampling consistency initial registration algorithm to obtain a first registration result, including:

[0062] Step S21, selecting first three-dimensional point cloud data and second three-dimensional point cloud data from the three-dimensional point cloud data, wherein the first three-dimensional point cloud data is data collected by a first acquisition frame, the second three-dimensional point cloud data is data collected by a second acquisition frame, and the first acquisition frame and the second acquisition frame are continuous frames;

[0063] Step S22, determining a plurality of sample points from the first three-dimensional data;

[0064] Step S23, determining a plurality of similar points from the second three-dimensional point cloud data according to the plurality of sample points and a preset similarity condition;

[0065] Step S24, determining an initial point cloud transformation matrix according to the plurality of sample points and the plurality of similar points;

[0066] Step S25: performing a first registration operation on the three-dimensional point cloud data according to the initial point cloud transformation matrix to obtain a first registration result.

[0067] The first three-dimensional point cloud data refers to the three-dimensional point cloud data of the current transformer insulator captured by the inspection robot or other three-dimensional scanning equipment at the first acquisition frame. Specifically, the first three-dimensional point cloud data contains information such as the shape, position and surface details of the insulator at that time point, and is the reference point cloud in the subsequent registration process.

[0068] The second three-dimensional point cloud data refers to the three-dimensional point cloud data of the current transformer insulator captured by the inspection robot or other three-dimensional scanning equipment in the second acquisition frame. Specifically, the second three-dimensional point cloud data reflects the state change of the insulator after a short time interval.

[0069] Exemplarily, e sample points are selected from the first three-dimensional data, and the minimum distance d of the sample point pairing distance greater than a preset threshold is calculated. min For the above e sample points, all point clouds that meet the similarity conditions are found in the second three-dimensional point cloud data, and the correspondence between the sample points of some point clouds is randomly selected. Furthermore, according to the correspondence between the first three-dimensional data and the second three-dimensional data, the initial point cloud transformation matrix is ​​calculated, and the metric error is calculated to check the accuracy of the point cloud registration result.

[0070] For example, the metric error usually refers to the average distance error of corresponding point pairs between the target point cloud and the reference point cloud in the registered point cloud data, which can be calculated by formula (1):

[0071]

[0072] Among them, t e is the preset threshold; L k (e i ) is the measurement error of the kth sample point.

[0073] For example, by repeatedly iterating the registration process and calculating the metric error after each iteration, the rigid body transformation parameters can be continuously optimized until the metric error reaches an acceptable threshold or converges to a minimum value. When the optimal metric error is reached, the sampling consistency initial registration algorithm can be used to remove the mismatched points to obtain the initial point cloud transformation matrix R as shown in formula (2): SAC-IA:

[0074]

[0075] Based on the above optional embodiment, the initial point cloud transformation matrix is ​​determined based on multiple sample points in the first three-dimensional data and multiple similar points in the second three-dimensional data, which can reduce the amount and complexity of calculation while ensuring the alignment quality, thereby improving the substation inspection efficiency.

[0076] Optionally, in step S1412, a second registration operation is performed on the first registration result using an iterative closest point algorithm to obtain target point cloud data, including:

[0077] Step S31, determining target point cloud data and reference point cloud data according to the three-dimensional point cloud data, wherein the target point cloud data is the point cloud data corresponding to the first acquisition frame, the reference point cloud data is the point cloud data corresponding to the second acquisition frame, and the first acquisition frame and the second acquisition frame are continuous frames;

[0078] Step S32, determining a rotation matrix and a translation vector according to the target point cloud data and the reference point cloud data;

[0079] Step S33, determining matching point pairs in the target point cloud data and the reference point cloud data according to the rotation matrix and the translation vector; determining the target point cloud transformation matrix according to the matching point pairs;

[0080] Step S34: performing a second registration operation on the first registration result according to the target point cloud transformation matrix to obtain target point cloud data.

[0081] Exemplarily, the first three-dimensional point cloud data, that is, the target point cloud data, is marked as X={x1, x2, ..., x m}; The second three-dimensional point cloud data, that is, the reference point cloud data is marked as Y = {y1, y2, ..., y n}. The number of point clouds in X is m, the number of point clouds in Y is n, and m≤n.

[0082] Furthermore, the four-element method is used to set the initial value, and the K-dimensional tree algorithm is used to accelerate the search for the nearest point cloud. Assume that the rotation transformation vector is a unit quaternion q R =[q x ,q y ,q z ,q w ], where q x ≥0, and q x +q y +q z +q w =1, translation transformation vector q T =[t x ,t y ,tz ]. The point cloud centroid W of the first three-dimensional point cloud data x As shown in formula (3), the point cloud centroid W of the second three-dimensional point cloud data is y As shown in formula (4):

[0083]

[0084]

[0085] The rotation matrix R(q R ) is shown in formula (5):

[0086]

[0087] Translation vector q T As shown in formula (6):

[0088] q T =W y -R(q R )W x Formula (6)

[0089] Assuming that the normal vectors of the paired point clouds are n1 and n2 respectively, their weight w can be calculated by formula (7):

[0090] w=n1·n2 Formula (7)

[0091] Usually, the first three-dimensional point cloud data and the second three-dimensional point cloud data do not completely overlap, and the wrong matching relationship points including the edge need to be eliminated. Furthermore, after eliminating the wrong matching relationship points, the coordinate transformation matrix is ​​optimized by the iterative method based on the least squares method to minimize the error function. The error function is shown in formula (8):

[0092]

[0093] Where E(R,t) is the error function; x i is the i-th point cloud in the first three-dimensional point cloud data; y i is the second 3D point cloud data with x i The corresponding point cloud; R(x i ) is x i The rotation matrix of i The translation vector of .

[0094] Furthermore, the above steps are repeated to minimize the error, and the rotation matrix R(q R ) and the translation vector q T Perform registration operation on the first 3D point cloud data and the second 3D point cloud data to obtain the target point cloud transformation matrix RICP As shown in formula (9):

[0095]

[0096] In formula (9), the 9 elements in the upper left corner are regarded as matrix T1, the first 3 elements in the fourth column are regarded as matrix T2, the first 3 elements in the fourth row are regarded as matrix T3, and the 1 element in the lower right corner is regarded as matrix T4, and four matrices are obtained, namely the target point cloud data. Among them, T1 is used to rotate, symmetric and other changes to the point cloud data, as shown in formula (10); T2 is used to translate the point cloud data, as shown in formula (11); T3 is used to perform perspective projection transformation on the point cloud data, as shown in formula (12); T4 is used to perform overall scale transformation on the point cloud data, as shown in formula (13).

[0097]

[0098] T2=[-0.0760790.05601560.54358] Formula (11)

[0099]

[0100] T4=[1]Formula (13)

[0101] Figure 2 is a schematic diagram of a point cloud data registration result according to one embodiment of the present disclosure, such as Figure 2 As shown, the left side is the reference point cloud data, and the right side is the target point cloud data. The number of point clouds of the reference point cloud data, the target point cloud data, and the new point cloud after registration is shown in Table 1:

[0102] Table 1 Point cloud quantity calculation results

[0103]

[0104] As shown in Table 2, there are 5176 reference point cloud data, 4413 target point cloud data, and 9589 new point cloud data after registration.

[0105] Based on the above optional embodiment, performing a second registration operation on the first registration result by iterating the closest point algorithm can effectively reduce the error of point cloud registration and improve the registration accuracy.

[0106] Optionally, in step S142, a preset reconstruction algorithm is used to perform surface reconstruction on the target point cloud data to obtain a three-dimensional model, including:

[0107] Step S1421, constructing a vector field according to the normal vector of the target point cloud data;

[0108] Step S1422, determining an isosurface of the indicator function according to the vector field and the preset indicator function;

[0109] Step S143, post-processing the isosurface according to a preset post-processing strategy to obtain a three-dimensional model.

[0110] Figure 3 is a schematic diagram of a surface reconstruction method according to one embodiment of the present disclosure, such as Figure 3 As shown, Figure 3 (a) in the figure is a directed point cloud. Figure 3 (b) in the figure is a vector field. Figure 3 (c) in is the preset indicator function, Figure 3 (d) in the figure is the reconstructed surface.

[0111] The above-mentioned preset indicator function refers to a binary function used to describe the internal and external areas of the point cloud data. In the internal area of ​​the point cloud, the preset indicator function value is 1, indicating that the area is occupied by the point cloud. In the external area of ​​the point cloud, the function value is 0, indicating that the area is a space that is not affected by the point cloud data.

[0112] The above-mentioned isosurface refers to a surface composed of a set of specific values ​​determined by a preset indicator function. In the Poisson surface reconstruction algorithm, the isosurface is usually defined as a set of points with an indicator function value equal to 0.5, which is used to approximate the surface boundary of the point cloud data.

[0113] The above-mentioned preset post-processing strategy refers to a series of optimization measures taken after the initial construction of the 3D model is completed in order to improve the quality and applicability of the model, including but not limited to hole repair, texture mapping and model smoothing.

[0114] For example, the surface S is shown in formula (14):

[0115] S={x:χ(x)=0} Formula (14)

[0116] If the point is outside the surface S, then χ=0; if the point is outside the surface S, then χ=1.

[0117] Exemplarily, the registered point cloud data is reconstructed using a Poisson surface reconstruction algorithm. Specifically, the Poisson equation is used as an implicit function to calculate the best isosurface of the insulator point cloud data accurately registered by the iterative closest point algorithm, and interpolation is performed to approximate the original insulator surface.

[0118] For example, assuming that the point cloud data sample is Where any point cloud x k ∈R 3 , whose corresponding normal is n k ∈R 2, so that the vector field can be constructed according to the normal vector of the target point cloud data, and then the isosurface of the indicator function can be determined according to the above constructed vector field and the preset indicator function χ. In , the internal normal of each point cloud is the gradient of the indicator function, that is, Given a continuous vector field n, insert it into the point cloud data, that is, n(x k. )=n k Finally, the gradient of the indicator function χ is determined to best match the vector field n, so that the complete reconstructed point cloud surface can be obtained.

[0119] Figure 4 is a schematic diagram of a three-dimensional model according to one embodiment of the present disclosure, such as Figure 4 As shown, the three-dimensional reconstructed model of the current transformer insulator is displayed.

[0120] Exemplarily, first, the first and second three-dimensional point cloud data from two frames of current transformer insulators are aligned and integrated to generate new point cloud data; secondly, Poisson surface reconstruction is performed on the generated new point cloud data to convert the point cloud data into a continuous surface model; finally, in order to improve the integrity and visual effect of the model, the reconstructed surface is post-processed to obtain a three-dimensional reconstructed model of the current transformer insulator.

[0121] Based on the above optional embodiments, the isosurface of the indicator function is determined according to the vector field and the preset indicator function, and then the isosurface is post-processed according to the preset post-processing strategy. This can not only obtain a three-dimensional model that accurately reflects the current transformer insulator in terms of geometric form, but also ensure that the obtained model is highly consistent with the actual equipment in terms of details, visual effects and physical properties, thereby enhancing the practical value of the model.

[0122] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present disclosure, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present disclosure.

[0123] In the embodiments of the present disclosure, a substation inspection device is also provided, which is used to implement the above embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0124] Figure 5 is a structural block diagram of a substation inspection device according to one embodiment of the present disclosure, such as Figure 5 As shown, the device comprises:

[0125] An acquisition module 501 is used to acquire the position data of the current transformer insulator;

[0126] A determination module 502, for determining an inspection route according to the current transformer insulator position data;

[0127] The acquisition module 503 is used to control the inspection robot to perform inspections and collect three-dimensional point cloud data of the current transformer insulators according to the inspection route;

[0128] A construction module 504 is used to construct a three-dimensional model corresponding to the current transformer insulator according to the three-dimensional point cloud data;

[0129] The comparison module 505 is used to compare the three-dimensional model with a preset standard model to determine the inspection result.

[0130] Optionally, the construction module 504 is also used to: use a preset registration algorithm to align the three-dimensional point cloud data to obtain target point cloud data; use a preset reconstruction algorithm to perform surface reconstruction on the target point cloud data to obtain a three-dimensional model.

[0131] Optionally, the construction module 504 is also used to: use a sampling consistency initial registration algorithm to perform a first registration operation on the three-dimensional point cloud data to obtain a first registration result; use an iterative closest point algorithm to perform a second registration operation on the first registration result to obtain target point cloud data.

[0132] Optionally, the construction module 504 is also used to: select first three-dimensional point cloud data and second three-dimensional point cloud data from the three-dimensional point cloud data, wherein the first three-dimensional point cloud data is data collected by a first acquisition frame, the second three-dimensional point cloud data is data collected by a second acquisition frame, and the first acquisition frame and the second acquisition frame are continuous frames; determine multiple sample points from the first three-dimensional data; determine multiple similar points from the second three-dimensional point cloud data based on the multiple sample points and a preset similarity condition; determine an initial point cloud transformation matrix based on the multiple sample points and the multiple similar points; and perform a first registration operation on the three-dimensional point cloud data based on the initial point cloud transformation matrix to obtain a first registration result.

[0133] Optionally, the construction module 504 is also used to: determine target point cloud data and reference point cloud data based on the three-dimensional point cloud data, wherein the target point cloud data is the point cloud data corresponding to the first acquisition frame, the reference point cloud data is the point cloud data corresponding to the second acquisition frame, and the first acquisition frame and the second acquisition frame are continuous frames; determine the rotation matrix and the translation vector based on the target point cloud data and the reference point cloud data; determine the matching point pairs in the target point cloud data and the reference point cloud data based on the rotation matrix and the translation vector; determine the target point cloud transformation matrix based on the matching point pairs; and perform a second registration operation on the first registration result based on the target point cloud transformation matrix to obtain the target point cloud data.

[0134] Optionally, the construction module 504 is also used to: construct a vector field based on the normal vector of the target point cloud data; determine the isosurface of the indicator function based on the vector field and the preset indicator function; and post-process the isosurface according to a preset post-processing strategy to obtain a three-dimensional model.

[0135] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0136] According to another aspect of the embodiments of the present disclosure, an electronic device is further provided, including: a memory storing an executable program; and a processor for running the program, wherein the method in each embodiment of the present disclosure is executed when the program is running.

[0137] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0138] S1, obtaining the current transformer insulator position data;

[0139] S2, determining the inspection route according to the current transformer insulator position data;

[0140] S3, according to the inspection route, control the inspection robot to perform inspection and collect three-dimensional point cloud data of the current transformer insulator;

[0141] S4, constructing a three-dimensional model corresponding to the current transformer insulator according to the three-dimensional point cloud data;

[0142] S5, comparing the three-dimensional model with a preset standard model to determine the inspection result.

[0143] According to another aspect of an embodiment of the present disclosure, a computer-readable storage medium is further provided. The computer-readable storage medium includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in various embodiments of the present disclosure.

[0144] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0145] S1, obtaining the current transformer insulator position data;

[0146] S2, determining the inspection route according to the current transformer insulator position data;

[0147] S3, according to the inspection route, control the inspection robot to perform inspection and collect three-dimensional point cloud data of the current transformer insulator;

[0148] S4, constructing a three-dimensional model corresponding to the current transformer insulator according to the three-dimensional point cloud data;

[0149] S5, comparing the three-dimensional model with a preset standard model to determine the inspection result.

[0150] Optionally, in this embodiment, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store computer programs.

[0151] According to another aspect of the embodiments of the present disclosure, a computer program product is also provided, including a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present disclosure is implemented.

[0152] Optionally, in this embodiment, the computer program product may be configured as a computer program for executing the following steps:

[0153] S1, obtaining the current transformer insulator position data;

[0154] S2, determining the inspection route according to the current transformer insulator position data;

[0155] S3, according to the inspection route, control the inspection robot to perform inspection and collect three-dimensional point cloud data of the current transformer insulator;

[0156] S4, constructing a three-dimensional model corresponding to the current transformer insulator according to the three-dimensional point cloud data;

[0157] S5, comparing the three-dimensional model with a preset standard model to determine the inspection result.

[0158] In the above embodiments of the present disclosure, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0159] In the several embodiments provided in the present disclosure, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0160] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0161] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0162] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present disclosure is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present disclosure. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0163] The above is only a preferred embodiment of the present disclosure. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present disclosure. These improvements and modifications should also be regarded as the scope of protection of the present disclosure.

Claims

1. A substation inspection method, characterized in that: include: Obtaining current transformer insulator position data; Determining an inspection route according to the current transformer insulator position data; According to the inspection route, controlling the inspection robot to perform inspection and collect three-dimensional point cloud data of the current transformer insulator; Constructing a three-dimensional model corresponding to the current transformer insulator according to the three-dimensional point cloud data; The three-dimensional model is compared with a preset standard model to determine the inspection result.

2. The substation inspection method according to claim 1, characterized in that: The step of constructing a three-dimensional model corresponding to the current transformer insulator according to the three-dimensional point cloud data includes: Using a preset registration algorithm, the three-dimensional point cloud data is registered to obtain target point cloud data; The target point cloud data is reconstructed using a preset reconstruction algorithm to obtain the three-dimensional model.

3. The substation inspection method according to claim 2, characterized in that: The preset registration algorithm includes a sampling consistency initial registration algorithm and an iterative closest point algorithm. The preset registration algorithm is used to register the three-dimensional point cloud data to obtain target point cloud data, including: Performing a first registration operation on the three-dimensional point cloud data using the sampling consistency initial registration algorithm to obtain a first registration result; The iterative closest point algorithm is used to perform a second registration operation on the first registration result to obtain the target point cloud data.

4. The substation inspection method according to claim 3, characterized in that: The step of performing a first registration operation on the three-dimensional point cloud data using the sampling consistency initial registration algorithm to obtain a first registration result includes: Selecting first three-dimensional point cloud data and second three-dimensional point cloud data from the three-dimensional point cloud data, wherein the first three-dimensional point cloud data is data collected by a first acquisition frame, the second three-dimensional point cloud data is data collected by a second acquisition frame, and the first acquisition frame and the second acquisition frame are continuous frames; determining a plurality of sample points from the first three-dimensional data; Determining a plurality of similar points from the second three-dimensional point cloud data according to the plurality of sample points and a preset similarity condition; Determine an initial point cloud transformation matrix according to the plurality of sample points and the plurality of similar points; According to the initial point cloud transformation matrix, a first registration operation is performed on the three-dimensional point cloud data to obtain the first registration result.

5. The substation inspection method according to claim 3, characterized in that: The step of performing a second registration operation on the first registration result by using the iterative closest point algorithm to obtain the target point cloud data includes: Determine target point cloud data and reference point cloud data according to the three-dimensional point cloud data, wherein the target point cloud data is point cloud data corresponding to a first acquisition frame, the reference point cloud data is point cloud data corresponding to a second acquisition frame, and the first acquisition frame and the second acquisition frame are continuous frames; Determine a rotation matrix and a translation vector according to the target point cloud data and the reference point cloud data; Determine matching point pairs in the target point cloud data and the reference point cloud data according to the rotation matrix and the translation vector; Determine a target point cloud transformation matrix according to the matching point pairs; According to the target point cloud transformation matrix, a second registration operation is performed on the first registration result to obtain the target point cloud data.

6. The substation inspection method according to claim 2, characterized in that: The method of performing surface reconstruction on the target point cloud data by using a preset reconstruction algorithm to obtain the three-dimensional model includes: Constructing a vector field according to the normal vector of the target point cloud data; Determining an isosurface of the indicator function according to the vector field and a preset indicator function; According to a preset post-processing strategy, the isosurface is post-processed to obtain the three-dimensional model.

7. A substation inspection device, characterized in that: include: An acquisition module, used for acquiring the position data of the insulator of the current transformer; A determination module, used to determine an inspection route according to the current transformer insulator position data; A collection module, used to control the inspection robot to perform inspection and collect three-dimensional point cloud data of the current transformer insulator according to the inspection route; A construction module, used to construct a three-dimensional model corresponding to the current transformer insulator according to the three-dimensional point cloud data; The comparison module is used to compare the three-dimensional model with a preset standard model to determine the inspection result.

8. An electronic device, characterized in that: include: A memory storing an executable program; A processor, configured to run the program, wherein the program executes the method according to any one of claims 1 to 6 when running.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored executable program, wherein when the executable program is executed, the device where the storage medium is located is controlled to execute the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The invention comprises a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images