Robot inspection method, system and equipment based on 500kV transformer substation

Abstract

The invention relates to the technical field of transformer substation inspection, and discloses a robot inspection method, system and equipment based on a 500kV transformer substation. The method comprises the following steps: acquiring a real-time acquisition image during inspection; recognizing substation equipment and the position of the equipment according to the real-time acquired image by an equipment mode recognition technology based on image matching; judging to obtain a state recognition result of the substation equipment according to the real-time acquired image; analyzing and obtaining a sound analysis result based on the operation sound information of the position acquisition equipment and a vector quantization algorithm; according to the state recognition result and the sound analysis result, whether early warning is performed is judged; and correcting the intelligent navigation system based on a Kalman filtering matching positioning tracking technology so as to adjust an inspection path. An image matching identification technology, a sound analysis technology of a vector quantization algorithm and a matching positioning tracking technology of Kalman filtering are adopted, the inspection quality and efficiency of a transformer substation are effectively improved, the service life of equipment is prolonged, the reliability of a power grid is improved, and the manual workload is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of substation inspection, and in particular to a robot inspection method, system and equipment based on a 500kV substation. Background Art

[0002] With the rapid development of the power industry, the scale of the power grid continues to expand, and the number of substations has increased dramatically, which has led to the increasingly prominent contradiction of the relatively insufficient number of inspection personnel. At present, manual inspection is still the inspection method adopted by most substations. Due to the influence of the skill level, business proficiency and working status of the inspection operators, the missed inspection and misjudgment of the equipment operation status often occur, resulting in certain ineffective operations. In serious cases, it may cause power accidents and affect the normal power consumption of power customers.

[0003] The traditional manual inspection and maintenance mechanism is difficult to adapt to the rapid development and construction of substations in terms of inspection efficiency and operation and maintenance quality. Especially after the 500kV ultra-high voltage substation gradually shifted to a less-manned or even unmanned operation, the traditional manual inspection faced many problems such as heavy workload, high safety risks and unstable inspection quality. Summary of the invention

[0004] The main purpose of the present invention is to provide a robot inspection method, system and equipment based on a 500kV substation, aiming to solve at least one of the above-mentioned technical problems.

[0005] To achieve the above object, the present invention provides a robot inspection method based on a 500kV substation, comprising:

[0006] Based on intelligent inspection robots, real-time images are acquired during substation inspections;

[0007] The device pattern recognition technology based on image matching identifies the substation equipment and the location of the substation equipment according to the real-time collected image;

[0008] Determine whether the state of the substation equipment is abnormal according to the real-time collected image, and obtain a state recognition result;

[0009] Collecting operation sound information of the substation equipment based on the location of the substation equipment, and analyzing the operation sound information based on a vector quantization algorithm to obtain a sound analysis result;

[0010] Determining whether to issue an early warning based on the state recognition result and the sound analysis result;

[0011] The intelligent navigation system of the inspection robot is corrected based on the matching positioning and tracking technology of Kalman filtering to adjust the inspection path of the inspection robot.

[0012] In some embodiments, the device pattern recognition technology based on image matching identifies substation equipment and the location of the substation equipment according to the real-time collected image, including:

[0013] Obtain pictures of substation equipment to be inspected based on preset season, weather and time conditions;

[0014] Classify and store the photographed pictures according to the stop points, seasons and weather factors to establish a standard picture library;

[0015] The equipment pattern recognition technology based on image matching matches the real-time collected image with the captured image of the standard image library to identify the substation equipment and the location of the substation equipment.

[0016] In some embodiments, the matching formula of the device pattern recognition technology of image matching is:

[0017]

[0018] Among them, I r To collect images in real time; s The pictures are taken from the standard picture library; S(I r ,I s ) is the similarity between the real-time collected image and the captured image in the standard image library.

[0019] In some embodiments, judging whether the state of the substation equipment is abnormal according to the real-time collected image to obtain a state recognition result includes:

[0020] Obtaining a visible light image and an infrared image captured simultaneously according to the real-time collected image;

[0021] Registering the visible light image with the infrared image to obtain temperature information of the substation equipment;

[0022] Determine whether the state of the substation equipment is abnormal according to a preset threshold and the temperature information, and obtain a state identification result;

[0023] When the status identification result is that the status of the substation equipment is abnormal, an alarm signal is sent.

[0024] In some embodiments, the collecting the operation sound information of the substation equipment based on the location of the substation equipment, and analyzing the operation sound information based on a vector quantization algorithm to obtain a sound analysis result includes:

[0025] collecting operation sound information of the substation equipment based on the location of the substation equipment;

[0026] Extracting features of the running sound information based on a vector quantization algorithm to obtain sound signal features;

[0027] Extracting the sound features of the substation equipment during normal operation, and training to form a sound feature library based on the sound features;

[0028] The collected sound signal features are compared with the features of the sound feature library to obtain the sound analysis results.

[0029] In some embodiments, a Mel-frequency cepstral coefficient calculation formula is used to extract features of the operating sound information, and the Mel-frequency cepstral coefficient calculation formula is:

[0030]

[0031] Among them, C k is the Mel filter coefficient; S(m) is the short-time Fourier transform of the signal; H(m,n,k) is the Mel filter.

[0032] In some embodiments, the Kalman filter-based matching positioning and tracking technology corrects the intelligent navigation system of the inspection robot to adjust the inspection path of the inspection robot, including:

[0033] The matching positioning and tracking technology based on Kalman filtering is used to linearize the intelligent navigation system of the inspection robot;

[0034] Kalman filtering is used to optimally estimate the state of the intelligent navigation system after linearization processing to obtain fused navigation data;

[0035] The inspection target point of the inspection robot is located according to the fused navigation data to adjust the inspection path of the inspection robot.

[0036] In some embodiments, the Kalman filter is used to optimally estimate the state of the intelligent navigation system after linearization processing to obtain fused navigation data, including:

[0037] Construct discrete equations to calculate the Kalman gains of the GPS navigation system and DR navigation system after linearization;

[0038] The output states of the GPS navigation system and the DR navigation system after linearization are estimated by using a Kalman filter in combination with the Kalman gain to obtain an estimated value;

[0039] The navigation data of the inspection robot is corrected according to the estimated value, so that the output data of the GPS navigation system and the DR navigation system are fused to obtain fused navigation data.

[0040] In addition, to achieve the above purpose, the present invention also proposes a robot inspection system based on a 500kV substation, comprising:

[0041] An image acquisition module is used to obtain real-time images when the intelligent inspection robot inspects the substation;

[0042] An identification and detection module, used for identifying substation equipment and the location of the substation equipment according to the real-time collected image based on the equipment pattern recognition technology of image matching;

[0043] A state judgment module, used to judge whether the state of the substation equipment is abnormal according to the real-time collected image, and obtain a state recognition result;

[0044] A sound analysis module, used for collecting operation sound information of the substation equipment based on the location of the substation equipment, and analyzing the operation sound information based on a vector quantization algorithm to obtain a sound analysis result;

[0045] An abnormal warning module, used to determine whether to issue a warning based on the state recognition result and the sound analysis result;

[0046] The positioning and tracking module is used to correct the intelligent navigation system of the inspection robot based on the matching positioning and tracking technology of Kalman filtering to adjust the inspection path of the inspection robot.

[0047] In addition, to achieve the above-mentioned purpose, the present invention also proposes an electronic device, which includes: a memory, a processor, and a robot inspection program based on a 500kV substation stored in the memory and run on the processor, and the robot inspection program based on a 500kV substation is configured to implement the robot inspection method based on a 500kV substation as described above.

[0048] The present invention provides a robot inspection method based on a 500kV substation, comprising: obtaining a real-time acquisition image based on an intelligent inspection robot during substation inspection; identifying substation equipment and the position of the substation equipment according to the real-time acquisition image based on an equipment pattern recognition technology based on image matching; judging whether the state of the substation equipment is abnormal according to the real-time acquisition image to obtain a state recognition result; collecting the operation sound information of the substation equipment based on the position of the substation equipment, and analyzing the operation sound information based on a vector quantization algorithm to obtain a sound analysis result; judging whether to warn according to the state recognition result and the sound analysis result; and correcting the intelligent navigation system of the inspection robot based on the matching positioning tracking technology of Kalman filtering to adjust the inspection path of the inspection robot. In the present invention, the use of equipment pattern recognition technology based on image matching, substation equipment sound collection and analysis technology based on vector quantization algorithm, and matching positioning tracking technology based on Kalman filtering can effectively improve the quality and efficiency of robot inspection of substations. Compared with traditional manual inspection, the present invention has significant improvements in detection accuracy and work efficiency, thereby extending the life of substation equipment and improving power grid reliability. At the same time, the manual workload was greatly reduced and the allocation of human resources was optimized. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A schematic diagram of the structure of an electronic device in a hardware operating environment involved in an embodiment of the present invention;

[0050] Figure 2 It is a flow chart of an embodiment of a robot inspection method based on a 500kV substation according to the present invention;

[0051] Figure 3 This is a flow chart of device mode recognition involved in an embodiment of the present invention;

[0052] Figure 4 It is a schematic diagram of an infrared image and a visible light image involved in an embodiment of the present invention;

[0053] Figure 5 A schematic diagram of a main transformer inspection line involved in an embodiment of the present invention;

[0054] Figure 6 A schematic diagram showing a comparison of detection accuracy rates involved in embodiments of the present invention;

[0055] Figure 7 The structure diagram of an embodiment of a robot inspection system based on a 500kV substation of the present invention.

[0056] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0059] In addition, the descriptions of "first", "second", etc. in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the ability of ordinary technicians in the field to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by the present invention. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0060] Reference Figure 1 , Figure 1 The figure is a schematic diagram of the structure of an electronic device of the hardware operating environment involved in the embodiment of the present invention.

[0061] like Figure 1As shown, the electronic device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, Wi-Fi) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM memory) or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk memory. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0062] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the electronic device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0063] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and a robot inspection program based on a 500 kV substation.

[0064] exist Figure 1 In the electronic device shown, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the electronic device of the present invention can be set in the electronic device, and the electronic device calls the robot inspection program based on the 500kV substation stored in the memory 1005 through the processor 1001, and executes the robot inspection method based on the 500kV substation provided by an embodiment of the present invention.

[0065] The present invention provides a robot inspection method, system and equipment based on a 500kV substation.

[0066] The embodiment of the present invention provides a robot inspection method based on a 500kV substation, referring to Figure 2 , Figure 2 The figure is a flow chart of an embodiment of a robot inspection method based on a 500kV substation according to the present invention.

[0067] like Figure 2 As shown, the robot inspection method based on the 500kV substation includes:

[0068] Step S100: acquiring real-time images when inspecting the substation based on the intelligent inspection robot;

[0069] Step S200: using an image matching-based device pattern recognition technology to identify substation equipment and the location of the substation equipment according to the real-time collected image;

[0070] Step S300: judging whether the state of the substation equipment is abnormal according to the real-time collected image, and obtaining a state recognition result;

[0071] Step S400: collecting operation sound information of the substation equipment based on the location of the substation equipment, and analyzing the operation sound information based on a vector quantization algorithm to obtain a sound analysis result;

[0072] Step S500: determining whether to issue an early warning according to the state recognition result and the sound analysis result;

[0073] Step S600: Correcting the intelligent navigation system of the inspection robot based on the matching positioning and tracking technology of Kalman filtering to adjust the inspection path of the inspection robot.

[0074] It should be noted that the execution subject in this embodiment may be an electronic device, which may be a computer device with data processing functions, or other devices that can achieve the same or similar functions. This embodiment does not limit this. In this embodiment, a computer device is used as an example for explanation.

[0075] It is understandable that this embodiment uses the intelligent inspection robot system applied in a 500kV substation to realize the intelligent inspection function of the substation as an example for explanation. The main functions of the intelligent inspection robot system include: using a high-definition camera to inspect the appearance of the equipment, and using image recognition technology to perform real-time matching and analysis; using an infrared thermal imager for non-contact temperature measurement to promptly detect equipment heating defects; automatically reading various instrument data to monitor the operating status of the equipment; using sound recognition technology to detect abnormal noise of the equipment; having an automatic charging function to ensure long-term operation; supporting remote control to improve inspection flexibility. The comprehensive application of these functions has greatly improved the quality and efficiency of inspections, reduced the intensity of manual labor, and provided a strong guarantee for the safe and stable operation of unmanned substations. Among them, the intelligent inspection robot system includes but is not limited to inspection and signal acquisition devices such as robot dogs, robots, drones, sensors, and cameras.

[0076] In one embodiment, the device pattern recognition technology based on image matching identifies the substation equipment and the location of the substation equipment based on the real-time collected image, including: obtaining photographs of the substation equipment to be detected based on preset season, weather and time conditions; classifying and storing the photographs according to stop points, seasons and weather factors to establish a standard image library; the device pattern recognition technology based on image matching matches the real-time collected image with the photographs in the standard image library to identify the substation equipment and the location of the substation equipment.

[0077] In one embodiment, the matching formula of the device pattern recognition technology of image matching is:

[0078]

[0079] Among them, I r To collect images in real time; s The pictures are taken from the standard picture library; S(I r ,I s ) is the similarity between the real-time collected image and the captured image in the standard image library.

[0080] Specifically, Figure 3 As shown, the device pattern recognition technology based on image matching: establish a standard image library containing various environmental factors (such as season, weather, time, etc.). During inspection, the real-time acquired image is matched with the standard image (shot image) in the standard image library to identify the specific substation equipment type and its location. The device pattern recognition technology based on image matching used in this embodiment can effectively overcome the influence of factors such as shooting angle deviation and realize all-weather equipment recognition. The device pattern recognition technology based on image matching is one of the core technologies of the intelligent inspection robot system, which can automatically identify various equipment in the substation and detect its working status.

[0081] For example, a standard image library of equipment is established: photos of all equipment to be inspected in the substation are taken under different seasons, weather and time conditions to establish a comprehensive standard image library. These images in the standard image library can be classified and stored according to factors such as stop points, seasons, and weather.

[0082] Exemplarily, real-time matching and detection and identification: During the matching process, the real-time collected images can be uploaded to the base station system by the intelligent inspection robot system (such as a robot) and matched in real time with the standard images in the standard image library. In one example, when the robot performs an inspection task, it will upload the captured images (real-time collected images) to the base station system in real time, and the base station system will match the real-time collected images with the corresponding standard images in the standard image library in real time. The process of real-time matching and detection and identification can be expressed by the following formula:

[0083]

[0084] Among them, I r To collect images in real time; s The pictures are taken from the standard picture library; S(I r ,I s ) is the similarity between the real-time collected image and the captured image in the standard image library.

[0085] For example, the location and type of each substation equipment in the real-time image can be determined through the matching algorithm. At the same time, the visible light image is registered with the infrared image taken at the same time to obtain the temperature information of the substation equipment. Here, the image registration transformation formula is as follows:

[0086] I aligned =T(I r )

[0087] Among them, I aligned is the image after registration processing; T is used for real-time acquisition of image I r The transformation function of .

[0088] In one embodiment, judging whether the state of the substation equipment is abnormal based on the real-time acquired image to obtain a state recognition result includes: obtaining a visible light image and an infrared image taken simultaneously based on the real-time acquired image; aligning the visible light image with the infrared image to obtain temperature information of the substation equipment; judging whether the state of the substation equipment is abnormal based on a preset threshold and the temperature information to obtain a state recognition result; and sending an alarm signal when the state recognition result is that the state of the substation equipment is abnormal.

[0089] Exemplarily, according to a preset threshold, it is determined whether the state of the substation equipment is abnormal, and a state identification result is obtained. If the substation equipment is found to be abnormal, such as the temperature of the substation equipment is too high or the meter reading is abnormal, an alarm can be issued immediately.

[0090] In this embodiment, Figure 4 As shown in the figure, the appearance of the equipment can be inspected by a high-definition camera, and image recognition technology can be used for real-time matching and analysis; infrared thermal imagers can be used for non-contact temperature measurement to promptly detect heating defects of the equipment, etc. This equipment pattern recognition technology based on image matching can adapt to all-weather working conditions, overcome the influence of factors such as shooting angle and light changes, and greatly improve the accuracy and reliability of equipment identification and detection.

[0091] In one embodiment, operating sound information of the substation equipment is collected based on the location of the substation equipment, and the operating sound information is analyzed based on a vector quantization algorithm to obtain a sound analysis result, including: collecting the operating sound information of the substation equipment based on the location of the substation equipment; extracting features of the operating sound information based on a vector quantization algorithm to obtain sound signal features; extracting sound features of the substation equipment during normal operation, and forming a sound feature library based on the sound features; comparing the collected sound signal features with the features of the sound feature library to obtain a sound analysis result.

[0092] In one embodiment, a Mel-frequency cepstral coefficient calculation formula is used to extract features of the operation sound information, and the Mel-frequency cepstral coefficient calculation formula is:

[0093]

[0094] Among them, C k is the Mel filter coefficient; S(m) is the short-time Fourier transform of the signal; H(m,n,k) is the Mel filter.

[0095] Specifically, the sound collection and analysis technology of substation equipment based on the vector quantization algorithm: The sound collection and analysis technology is an important part of the intelligent inspection robot system. It determines whether the equipment is in a normal state by analyzing the sound when the equipment is running. This technology is mainly based on the vector quantization (VQ) algorithm. Its basic principle is to extract the sound characteristics of the substation equipment during normal operation and form a feature library of normal sounds. The substation equipment sound collection and analysis technology based on the vector quantization algorithm used in this embodiment can effectively improve the reliability and stability of the inspection and provide important support for the intelligent inspection robot system.

[0096] It should be noted that the technology of judging whether the equipment is in a normal state by analyzing the sound when the equipment is running is mainly based on the vector quantization (VQ) algorithm. Its basic principle is to extract the sound characteristics of the substation equipment during normal operation and form a feature library of normal sound. The process of the vector quantization (VQ) algorithm (the first step to the third step) is as follows:

[0097] The first step is feature extraction: In the process of processing sound signals, the first step is to extract features from the collected audio signals. Common features include short-time energy, zero-crossing rate, MFCC (Mel-frequency cepstral coefficients), etc. Here, in this embodiment, the use of MFCC (Mel-frequency cepstral coefficients) to extract sound signal features is used as an example for explanation, and the calculation formula is as follows:

[0098]

[0099] Among them, C kis the Mel filter coefficient; S(m) is the short-time Fourier transform of the signal; H(m,n,k) is the Mel filter.

[0100] The second step is to train the feature library: the sound features collected during normal operation of the substation equipment are centrally trained to form a feature library of normal sounds. During the actual inspection process, the sound features collected in real time are compared with the feature library of normal sounds to obtain the sound analysis results. The comparison process can be expressed by the following formula:

[0101]

[0102] Where D is the distance between two sets of features; x i and i They are respectively the sound features collected in real time and the feature values ​​in the feature library.

[0103] The third step is abnormality judgment: according to the comparison results of the second step (sound analysis results), when the characteristics of the sound to be identified (sound characteristics collected in real time) do not match the characteristics in the feature library of normal sound, it is determined that there may be abnormality in the substation equipment.

[0104] It is understood that vector quantization (VQ) is a signal compression technology that reduces the storage requirements or transmission bandwidth of data by mapping a continuous data space to a limited number of representative values ​​(called codewords). In sound analysis, vector quantization can be used for tasks such as sound signal compression, feature extraction, and pattern recognition.

[0105] Exemplarily, the general steps of using a vector quantization algorithm for sound analysis include: preprocessing sound data: converting analog sound signals into digital signals through an analog-to-digital converter (ADC); segmenting continuous sound signals into short, overlapping frames, usually using a window function (such as a Hamming window or a Hanning window) to reduce edge effects at frame boundaries; extracting features from each frame of sound data to obtain a set of feature vectors, such as commonly used features including Mel-frequency cepstral coefficients (MFCCs), linear prediction coefficients (LPCs) or spectral features. Designing a codebook: selecting a representative training set from a large amount of sound data; clustering the sound feature vectors using the training set to generate a codebook, and the clustering algorithm can be K-means, hierarchical clustering or Gaussian mixture model (GMM), etc.; optimizing the generated codebook to improve quantization accuracy and compression efficiency. Quantize sound data: For each frame of sound feature vector extracted, find the nearest codeword (most similar vector) in the codebook; use the index of the found codeword instead of the original feature vector for storage or transmission, thereby achieving compression; if the sound signal needs to be reconstructed, the corresponding codeword in the codebook can be used to approximate the original feature vector, and then the sound is restored through the decoding process. Sound analysis application: Use VQ for sound pattern recognition, and identify different sounds or speakers by comparing the similarity between the sound feature vector and the codeword in the codebook; in digital audio transmission or storage, use VQ to reduce the amount of data and improve efficiency; through the design and selection of the codebook, background noise can be eliminated to a certain extent and the sound quality can be improved. Through the above steps, the VQ algorithm used in this embodiment can be effectively used for sound analysis, reducing the data size while maintaining acceptable sound quality.

[0106] It should be noted that the present embodiment uses MFCC Mel-frequency cepstral coefficients for feature extraction of speech recognition and sound analysis. The exemplary steps of extracting features of sound using MFCC include: pre-emphasis: sound signals usually have higher energy in the low-frequency part. In order to enhance the high-frequency part and reduce this unevenness, a pre-emphasis filter can be applied, which is usually a first-order high-pass filter; framing: the pre-emphasized sound signal is divided into short-time frames, usually with a length of 20 to 30 milliseconds per frame, and there is a 50% overlap between frames, which can be achieved by moving a fixed-length window; windowing: in order to reduce the edge effect between frames, a window function is applied to each frame, and the window function includes a Hamming window. Window) or Hanning Window; Fourier transform: perform fast Fourier transform (FFT) on each windowed frame to obtain a spectrum representation; Mel filter bank: pass the output of FFT through a Mel filter bank, which consists of a series of bandpass filters covering the Mel scale. The Mel scale is a logarithmic scale that matches the auditory characteristics of the human ear; Logarithmic operation: take the logarithm of the output of each Mel filter (that is, the energy of each frequency band). This step can enhance the nonlinear characteristics of the signal and make the features more stable; Discrete cosine transform: perform discrete cosine transform (DCT) on the Mel spectrum after logarithmic operation to obtain the final MFCC features. Usually only the first few coefficients of DCT (usually 12 to 20) are taken because they contain most of the information. Through the above steps, MFCC features are extracted from the sound signal for speech recognition and sound analysis tasks.

[0107] In one embodiment, the intelligent navigation system of the inspection robot is corrected based on the matching positioning and tracking technology of Kalman filtering to adjust the inspection path of the inspection robot, including: linearizing the intelligent navigation system of the inspection robot based on the matching positioning and tracking technology of Kalman filtering; using Kalman filtering to optimally estimate the state of the intelligent navigation system after linearization to obtain fused navigation data; and locating the inspection target point of the inspection robot according to the fused navigation data to adjust the inspection path of the inspection robot.

[0108] In one embodiment, a Kalman filter is used to optimally estimate the state of the intelligent navigation system after linearization processing to obtain fused navigation data, including: constructing a discrete equation to calculate the Kalman gain of the GPS navigation system and the DR navigation system after linearization processing; estimating the output state of the GPS navigation system and the DR navigation system after linearization processing through a Kalman filter in combination with the Kalman gain to obtain an estimated value; and correcting the navigation data of the inspection robot according to the estimated value to fuse the output data of the GPS navigation system and the DR navigation system to obtain fused navigation data.

[0109] Specifically, the matching positioning and tracking technology based on Kalman filtering: The essence of the Kalman filter tracking algorithm is a linear system state method. Through the system input observation data, the input and observation signals with noise are processed, and then the system state is optimally estimated, that is, filtered. And using GPS / DR navigation technology, the output state of the intelligent navigation system is estimated through the Kalman filter. On this basis, the estimated value is used to continuously correct the navigation data, thereby completing the organic fusion of the output data of the two navigation systems (GPS navigation and DR navigation), such as Figure 5 The main transformer inspection route shown in the figure, the intelligent substation positioning and tracking technology based on the Kalman filter tracking algorithm adopted in this embodiment has strong advantages, which can enable the inspection robot to complete the work efficiently and accurately and maintain the stability of power equipment and supply.

[0110] Exemplarily, the matching positioning and tracking technology based on Kalman filtering, the specific steps (step 1 to step 3) are as follows:

[0111] Step 1: The essence of the Kalman filter tracking algorithm is a linear system state method. Through the system input observation data, the input and observation signals with noise are processed to optimally estimate the state of the system, that is, filtering. The Kalman filter tracking process can be expressed in two forms: one is to predict the subsequent state of the target in a linear space based on the known motion state equation; the other is a random process with a certain distribution and a mean of zero.

[0112] In the basic equation of Kalman filtering, the estimated state at time (k+1) can be expressed as x(k+1). Similarly, the state at time k is expressed as x(k). The state equation is:

[0113] x(k+1)=F(k+1,k)x(k)+V(k)

[0114] Among them, V(k) is the system noise sequence; F(k+1,k) is the one-step transfer matrix at time k.

[0115] Based on this, the observation matrix of Kalman filter is as follows:

[0116] Z(k)=C(k)x(k)+W(k)

[0117] Among them, C(k) is the observation matrix; W(k) is the observation noise sequence existing in the tracking process.

[0118] V(k) and W(k) generally satisfy the following equation:

[0119]

[0120] Where Q(k) is the noise variance matrix in the system.

[0121]

[0122] Where R(k) is the observation noise variance matrix. E[V(k)W(j) T ]=0.

[0123] Step 2: The intelligent inspection robot needs to use the combined navigation and positioning technology of Global Positioning System (GPS) / Dead Reckoning (DR) during the inspection process. Among them, the principle of GPS navigation is a satellite-based distance correction. When the user needs to locate, the position of the satellite is used as a known reference position point through GPS, and then the distance between the user and the reference satellite is measured to calculate the user's exact position. However, it is inevitable that during the positioning process, the receiver has a certain probability of causing errors in positioning accuracy, and even interference such as delay errors may occur when calculating the distance, which ultimately leads to inaccurate positioning results. On the other hand, DR refers to a navigation and positioning technology based on two-dimensional space, the essence of which is to lock the position of the target inspection robot through digital analysis and calculation.

[0124] In this embodiment, the inspection robot in the intelligent substation uses GPS / DR navigation technology to estimate the output state of the intelligent navigation system through the Kalman filter. On this basis, the estimated value is used to continuously correct the navigation data, thereby completing the organic fusion of the output data of the two navigation systems.

[0125] Step 3: When Kalman filtering is applied in a nonlinear system, the nonlinear system needs to be approximately linearized, and then the Kalman filter is used to optimally estimate the state of the system. In the estimation process, a discrete model needs to be constructed first, as shown in the following formula:

[0126]

[0127] Among them, f is a nonlinear equation; W is the noise during the operation of the system.

[0128] P is the covariance matrix of the system, which needs to be updated using the following formula:

[0129]

[0130] Among them, F is the matrix form of the system; Q represents the measurement noise.

[0131] The Kalman gain is calculated as follows:

[0132]

[0133] After calculating the Kalman gain, the optimal estimate of the current state can be calculated based on this data, and then:

[0134]

[0135] On this basis, we can get the relevant measured values ​​and estimated values ​​of the system after the extended Kalman filter. By comparing and analyzing them with the true values, we can calculate the error and then judge the accuracy of the extended Kalman filter. The system equation of the inspection robot is:

[0136]

[0137] The system equation of the inspection robot represents the inspection robot's ability to cope with obstacles during its movement. In the radar coordinate system corresponding to this system, x and y are the horizontal and vertical coordinates respectively; the corresponding v x With v y is the relative speed of the obstacle to the inspection robot in the x and y directions; a x with a y is the relative acceleration of the inspection robot in two directions; and are the resistances to travel in different directions respectively; during the movement of the inspection robot, and are white noises in different directions respectively.

[0138] Let the observed quantity be z = [ra] T , δ r , δ a is the error of white noise, then the observation equation is:

[0139]

[0140] The system matrix F and measurement matrix H are expressed as:

[0141]

[0142]

[0143] It should be noted that the example application and analysis of the robot inspection method based on 500kV substation: the application of the robot inspection method based on 500kV substation in this embodiment has achieved remarkable results in a certain 500kV substation, and the main transformer inspection is used as an example for analysis and explanation. Compared with the conventional inspection method, the robot inspection method based on 500kV substation in this embodiment based on the intelligent inspection robot can improve the inspection quality, reduce the workload and improve the economic benefits.

[0144] For example, the inspection quality analysis, workload analysis and economic benefit analysis were performed to verify the application effect of this embodiment. Through the inspection quality analysis, it was found that after being put into use, the accuracy and timeliness of the inspection were significantly improved, and equipment abnormalities were successfully detected and alarmed in time many times, effectively avoiding the occurrence of major accidents. Figure 6 As shown, for example, in the detection of equipment heating defects, it was found that the temperature of the main transformer capacitor group CT was abnormal, with a temperature difference of 15°C, exceeding the preset threshold of 10°C. During a special inspection of the 35kV capacitor group, it was detected that the temperature at the knife switch arm reached 85°C, exceeding the set upper limit of 75°C. It was also found that the temperature of the 202-1A phase knife switch was as high as 100.29°C, much higher than the average temperature of 60°C for the same type of equipment B and C. In terms of abnormal sound recognition, 5 abnormal transformer sounds and 3 abnormal noises of high-resistance equipment were successfully identified during the three-month trial operation, all of which were confirmed to be early signs of equipment failure. In addition, through image recognition technology, 2 porcelain bottle cracks and 1 tiny leak in the oil filling equipment were successfully detected. These subtle abnormalities are easily overlooked by manual inspections.

[0145] For example, through workload analysis, it is found that the application of the intelligent patrol robot system has greatly reduced the workload of manual patrol and improved work efficiency. The 500kV substation has switched to intelligent patrol to replace the traditional manual patrol mode. The intelligent patrol system conducts routine patrols once every 3 days and infrared temperature measurement once every 15 days. Manual patrols only need to complete comprehensive patrols once every 15 times. Intelligent patrol replacement not only does not reduce the quality of patrols, but also improves the overall patrol effect due to the efficiency and accuracy of the intelligent system. In terms of working time saving, after using the intelligent patrol system, a total of 47.5 hours are saved in daily visible light patrols and infrared temperature measurement each month, accounting for 67.9% of the original working time. Daily visible light patrols save 40 hours per month, and infrared temperature measurement saves 7.5 hours per month. This greatly reduces the workload of operation and maintenance personnel, allowing them to devote more energy to other important work. In terms of human resource optimization, the operation and maintenance personnel were reduced from 9 people / shift to 7 people / shift. They were responsible for the daily work of two 500kV substations and realized the unmanned operation of one of the 500kV substations.

[0146] For example, through economic benefit analysis, it is found that the application of the intelligent inspection robot system and the robot inspection method based on the 500kV substation in this embodiment not only improves the inspection quality and reduces the equipment failure rate, but also brings significant economic benefits. The direct economic benefits mainly come from three aspects: first, avoiding major accident losses, timely discovering and handling equipment defects, avoiding at least 2 major accidents that may occur, and conservatively estimating that the direct economic losses caused by each major accident are more than 5 million yuan, so at least 10 million yuan of potential losses are saved each year; second, labor cost savings, 47.5 hours of working time saved per month, calculated based on an average wage of 100 yuan / hour, can save 570,000 yuan in labor costs annually; third, the benefits brought by the extension of equipment life, by timely discovering and handling minor defects in equipment, it is expected that the service life of major equipment can be extended by 5% to 10%. Taking the main transformer as an example, assuming that the original service life is 30 years, it is extended by 2 years, with a value of about 20 million yuan, and an average annual increase of about 660,000 yuan in asset value.

[0147] It can be understood that in this embodiment, the device pattern recognition technology based on image matching can effectively overcome the influence of factors such as shooting angle deviation and realize all-weather equipment recognition. Equipment pattern recognition technology is one of the core technologies of the intelligent inspection robot system. It can automatically identify various equipment in the substation and detect its working status. The sound collection and analysis technology of substation equipment based on the vector quantization algorithm can effectively improve the reliability and stability of the inspection and provide important support for the intelligent inspection robot system. The matching positioning and tracking technology based on Kalman filtering has strong operability. This technology can help the inspection robot identify and avoid obstacles, accurately locate the inspection target point, plan the inspection work path, and complete the inspection task efficiently and intelligently.

[0148] This embodiment provides a robot inspection method based on a 500kV substation, including: obtaining real-time acquisition images based on an intelligent inspection robot during substation inspection; using device pattern recognition technology based on image matching to identify substation equipment and the location of the substation equipment according to the real-time acquisition images; judging whether the state of the substation equipment is abnormal according to the real-time acquisition images to obtain a state recognition result; collecting the operation sound information of the substation equipment based on the location of the substation equipment, and analyzing the operation sound information based on a vector quantization algorithm to obtain a sound analysis result; judging whether to warn according to the state recognition result and the sound analysis result; and correcting the intelligent navigation system of the inspection robot based on the matching positioning and tracking technology of Kalman filtering to adjust the inspection path of the inspection robot. In this embodiment, the use of device pattern recognition technology based on image matching, substation equipment sound collection and analysis technology based on vector quantization algorithm, and matching positioning and tracking technology based on Kalman filtering can effectively improve the quality and efficiency of robot inspection of substations. Compared with traditional manual inspection, this embodiment has significant improvements in detection accuracy and work efficiency, thereby extending the life of substation equipment and improving power grid reliability. At the same time, the manual workload was greatly reduced and the allocation of human resources was optimized.

[0149] In addition, an embodiment of the present invention also proposes a storage medium, on which a robot inspection program based on a 500kV substation is stored. When the robot inspection program based on a 500kV substation is executed by a processor, the steps of the robot inspection method based on a 500kV substation as described above are implemented.

[0150] Reference Figure 7 , Figure 7 The structure diagram of an embodiment of a robot inspection system based on a 500kV substation of the present invention.

[0151] like Figure 7 As shown, the robot inspection system based on the 500kV substation includes:

[0152] An image acquisition module 10 is used to acquire real-time images when the intelligent inspection robot inspects the substation;

[0153] An identification and detection module 20, configured to identify substation equipment and the location of the substation equipment according to the real-time collected image using a device pattern recognition technology based on image matching;

[0154] A state judgment module 30 is used to judge whether the state of the substation equipment is abnormal according to the real-time collected image, and obtain a state recognition result;

[0155] A sound analysis module 40, configured to collect operation sound information of the substation equipment based on the location of the substation equipment, and analyze the operation sound information based on a vector quantization algorithm to obtain a sound analysis result;

[0156] An abnormal warning module 50, used to determine whether to issue a warning based on the state recognition result and the sound analysis result;

[0157] The positioning and tracking module 60 is used to correct the intelligent navigation system of the inspection robot based on the matching positioning and tracking technology of Kalman filtering to adjust the inspection path of the inspection robot.

[0158] This embodiment proposes a robot inspection system based on a 500kV substation. The system uses equipment pattern recognition technology based on image matching, substation equipment sound collection and analysis technology based on vector quantization algorithm, and matching positioning and tracking technology based on Kalman filtering to effectively improve the quality and efficiency of robot inspections of substations. Compared with traditional manual inspections, this embodiment has significantly improved detection accuracy and work efficiency, thereby extending the life of substation equipment and improving power grid reliability. At the same time, it greatly reduces manual workload and optimizes human resource allocation.

[0159] It should be noted that the technical details that are not described in detail in the embodiment of the robot inspection system based on the 500kV substation can be referred to the robot inspection method based on the 500kV substation as described above provided in any embodiment of the present invention, and will not be repeated here.

[0160] It should be understood that the above is only an example and does not constitute any limitation on the technical solution of the present invention. In specific applications, technicians in this field can make settings as needed, and the present invention does not limit this.

[0161] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of the present invention. In practical applications, technicians in this field can select part or all of them according to actual needs to achieve the purpose of the present embodiment, and no limitation is made here.

[0162] In addition, it should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0163] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0164] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, 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 a read-only memory (ROM) / RAM, a magnetic 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 invention.

[0165] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A robot inspection method based on a 500kV substation, characterized in that: include: Based on intelligent inspection robots, real-time images are acquired during substation inspections; The device pattern recognition technology based on image matching identifies the substation equipment and the location of the substation equipment according to the real-time collected image; Determine whether the state of the substation equipment is abnormal according to the real-time collected image, and obtain a state recognition result; Collecting operation sound information of the substation equipment based on the location of the substation equipment, and analyzing the operation sound information based on a vector quantization algorithm to obtain a sound analysis result; Determining whether to issue an early warning based on the state recognition result and the sound analysis result; The intelligent navigation system of the inspection robot is corrected based on the matching positioning and tracking technology of Kalman filtering to adjust the inspection path of the inspection robot.

2. The method according to claim 1, characterized in that The device pattern recognition technology based on image matching identifies substation equipment and the location of the substation equipment according to the real-time collected image, including: Obtain pictures of substation equipment to be inspected based on preset season, weather and time conditions; Classify and store the photographed pictures according to the stop points, seasons and weather factors to establish a standard picture library; The equipment pattern recognition technology based on image matching matches the real-time collected image with the captured image of the standard image library to identify the substation equipment and the location of the substation equipment.

3. The method according to claim 2, characterized in that The matching formula of the device pattern recognition technology of image matching is: Among them, I r To collect images in real time; s The pictures are taken from the standard picture library; S(I r ,I s ) is the similarity between the real-time collected image and the captured image in the standard image library.

4. The method according to claim 1, characterized in that The determining whether the state of the substation equipment is abnormal according to the real-time collected image to obtain a state recognition result includes: Obtaining a visible light image and an infrared image captured simultaneously according to the real-time collected image; Registering the visible light image with the infrared image to obtain temperature information of the substation equipment; Determine whether the state of the substation equipment is abnormal according to a preset threshold and the temperature information, and obtain a state identification result; When the status identification result is that the status of the substation equipment is abnormal, an alarm signal is sent.

5. The method according to claim 1, characterized in that The collecting the operation sound information of the substation equipment based on the position of the substation equipment, and analyzing the operation sound information based on a vector quantization algorithm to obtain a sound analysis result includes: Collecting operation sound information of the substation equipment based on the location of the substation equipment; Extracting features of the running sound information based on a vector quantization algorithm to obtain sound signal features; Extracting the sound features of the substation equipment during normal operation, and training to form a sound feature library based on the sound features; The collected sound signal features are compared with the features of the sound feature library to obtain the sound analysis results.

6. The method according to claim 5, characterized in that The Mel frequency cepstral coefficient calculation formula is used to extract the features of the running sound information. The Mel frequency cepstral coefficient calculation formula is: Among them, C k is the Mel filter coefficient; S(m) is the short-time Fourier transform of the signal; H(m,n,k) is the Mel filter.

7. The method according to any one of claims 1 to 6, characterized in that The matching positioning and tracking technology based on Kalman filtering corrects the intelligent navigation system of the inspection robot to adjust the inspection path of the inspection robot, including: The matching positioning and tracking technology based on Kalman filtering is used to linearize the intelligent navigation system of the inspection robot; Kalman filtering is used to optimally estimate the state of the intelligent navigation system after linearization processing to obtain fused navigation data; The inspection target point of the inspection robot is located according to the fused navigation data to adjust the inspection path of the inspection robot.

8. The method according to claim 7, characterized in that The Kalman filter is used to optimally estimate the state of the intelligent navigation system after linearization processing to obtain fused navigation data, including: Construct discrete equations to calculate the Kalman gains of the GPS navigation system and DR navigation system after linearization; The output states of the GPS navigation system and the DR navigation system after linearization are estimated by using a Kalman filter in combination with the Kalman gain to obtain an estimated value; The navigation data of the inspection robot is corrected according to the estimated value, so that the output data of the GPS navigation system and the DR navigation system are fused to obtain fused navigation data.

9. A robot inspection system based on a 500kV substation, characterized in that: include: An image acquisition module is used to obtain real-time images when the intelligent inspection robot inspects the substation; An identification and detection module, used for identifying substation equipment and the location of the substation equipment according to the real-time collected image based on the equipment pattern recognition technology of image matching; A state judgment module, used to judge whether the state of the substation equipment is abnormal according to the real-time collected image, and obtain a state recognition result; A sound analysis module, used for collecting operation sound information of the substation equipment based on the location of the substation equipment, and analyzing the operation sound information based on a vector quantization algorithm to obtain a sound analysis result; An abnormal warning module, used to determine whether to issue a warning based on the state recognition result and the sound analysis result; The positioning and tracking module is used to correct the intelligent navigation system of the inspection robot based on the matching positioning and tracking technology of Kalman filtering to adjust the inspection path of the inspection robot.

10. An electronic device, characterized in that: The electronic device includes: a memory, a processor, and a robot inspection program based on a 500kV substation stored in the memory and executable on the processor, wherein the robot inspection program based on a 500kV substation is configured to implement the robot inspection method based on a 500kV substation as described in any one of claims 1 to 8.

Images