Intelligent inspection method for power distribution network

Abstract

The invention relates to the technical field of power distribution networks, in particular to an intelligent inspection method for a power distribution network, which comprises the following steps: S1, constructing an intelligent inspection system for the power distribution network; s2, partitioning the area covered by the power distribution network; s3, obtaining power distribution network inspection related information; s4, analyzing the obtained power distribution network inspection related information; s5, manually verifying the analysis result; and S6, performing corresponding slave processing on the condition according to the specific condition. The area covered by the power distribution network is divided into a common area, a hidden danger area and a dynamic area, resources needed by inspection are reasonably distributed according to the importance degree of the areas, inspection related information is obtained through two methods, an analysis module can automatically analyze the inspection information, the accuracy is improved through manual verification, and the inspection efficiency is improved. The inspection work efficiency can be effectively improved, the inspection period is shortened, the inspection work quality can be greatly improved, and the safety operation risk of a power grid is effectively reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of power distribution networks, and in particular to an intelligent inspection method for power distribution networks. Background Art

[0002] In recent years, with the rapid development of my country's economy, the demand for electricity supply has increased year by year, and the grid structure has been continuously improved. On the one hand, the scale of transmission lines has continued to expand, and the distribution points of high-voltage transmission lines are multifaceted and wide. The national economic development is closely related to the operation status of transmission lines; on the other hand, the terrain through which the transmission lines pass is complex and diverse, so the daily operation and maintenance of transmission lines is important and arduous. It is difficult to inspect transmission lines in mountainous areas, and it is difficult for power grid inspection staff to carry out refined inspections. With the upgrading of urban construction, the operating environment of urban transmission lines is harsh, construction hazards are increasing, and the requirements for power supply reliability are increasing. At the same time, the number of operation and maintenance personnel is zero or even negative. All of the above bring huge challenges to the inspection of transmission lines.

[0003] The existing inspection methods are mostly manual inspections, which are highly dependent on the inspectors' sense of responsibility and technical literacy. The lack of responsibility of the personnel often leads to problems such as inadequate inspections, untimely discovery of defects, and incomplete investigation of hidden dangers. The gap in the technical level of personnel directly affects the judgment of defects and hidden dangers, especially in complex environments. Manual inspections require more manpower, financial and material resources, and are also less efficient. Therefore, it is necessary to improve the existing inspection methods. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent inspection method for a distribution network to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: a smart inspection method for a distribution network, comprising the following steps: S1: Build an intelligent inspection system for distribution network; S2: Divide the area covered by the distribution network into different zones; S3: Obtaining distribution network inspection related information; S4: Analyze the obtained distribution network inspection related information; S5: Manual verification of the analysis results; S6: Handle the situation accordingly based on the specific circumstances.

[0006] Preferably, the distribution network intelligent inspection system includes a central processing unit module, a partition module, a storage module, a data acquisition module, an analysis module, a verification module and a post-processing module. The central processing unit module is bidirectionally connected to the partition module, the storage module and the data acquisition module. The data acquisition module is connected to the analysis module, the analysis module is connected to the verification module, and the verification module is connected to the post-processing module. The partitioning module is used to partition the area covered by the distribution network; The data acquisition module is used to acquire relevant information data of the distribution network inspection; The analysis module is used to analyze the inspection related information data obtained by the data acquisition module; The verification module is used to manually verify the analysis results obtained by the analysis module to determine whether the analysis results are correct; The post-processing module is used to perform processing operations according to the results obtained by the analysis module and the verification module.

[0007] Preferably, the area covered by the distribution network is divided into a general area, a potential hazard area and a dynamic area, and the inspection intensity of the general area is less than that of the potential hazard area.

[0008] Preferably, the hidden danger areas include areas with high objects, large span areas, important cross-span areas, areas near densely populated areas or lines of flammable and explosive places, areas with external force collision, and areas for wind and flood prevention; The high object area is described with poles and tall objects as objects. By performing three-dimensional laser modeling on the transmission line, the distance between the wire and the top of the object is calculated, the hidden danger section between the transmission line and the tall object is analyzed, and different hidden danger levels are determined according to the regulations of the line management unit; According to the laser point cloud tower data, the long span section is defined as a tension section with a spacing greater than or equal to 1000 meters or a tower height of 100 meters or more; The important crossing areas are the crossing sections where the transmission lines cross railways, expressways, first-class highways, first-class and second-class navigable rivers, special pipelines, tramways and cableways, and the crossing sections where 500kV overhead transmission lines cross 500kV and above AC and DC overhead transmission lines; The line area near densely populated areas or flammable and explosive places refers to schools, markets, gas stations, etc. The power transmission lines within 100 meters around them are near densely populated areas or flammable and explosive areas; The anti-external force collision zone is where the tower is located within 10 meters of the roadside, within 50 meters of the riverside, or in an area where vehicles and ships are easily accessible and where external force collision may occur; The wind and flood protection means that the tower foundation is located within 10 meters of the riverside and hillside, which is a wind and flood protection area.

[0009] Preferably, there are two methods for obtaining distribution network inspection-related information in S3, one is by obtaining relevant picture information of the distribution network, and the other is by detecting electricity consumption-related data of the distribution network through sensors; the methods for obtaining distribution network-related picture information are manual photography, drone photography, and monitoring probes equipped on distribution lines. Sensors are used to detect line voltage, current, temperature and humidity of equipment, and vibration.

[0010] Preferably, the analysis module includes a picture analysis unit and a circuit data analysis unit; The image analysis unit automatically pre-processes the image using a scale scaling algorithm and stores the image in a target detection image folder, analyzes and detects the image using a Faster-RCNN model trained to improve detection performance, and obtains a detection result after the image to be detected is detected by the model; The circuit data analysis unit sets a threshold value for all required distribution network data information. The circuit data analysis unit compares and analyzes the data detected by the sensor and the corresponding data. The detected data is judged to be normal when it is within the threshold value, and it is judged to be abnormal when the detected data is outside the threshold value.

[0011] Preferably, the analysis results of the analysis module are transmitted to the verification module and displayed to the staff, who judge and analyze the analysis results. If an abnormal situation occurs, the situation is processed accordingly through the post-processing module.

[0012] Compared with the prior art, the present invention has the following beneficial effects: The present invention proposes an intelligent inspection method for distribution network, which divides the area covered by the distribution network into ordinary area, potential danger area and dynamic area, reasonably allocates the resources required for inspection according to the importance of the area, obtains inspection-related information through two methods, and the analysis module can analyze the inspection information by itself, and then increase the accuracy through manual verification, which can effectively improve the inspection efficiency, shorten the inspection cycle, greatly improve the quality of inspection work, and effectively reduce the safety operation risk of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a schematic diagram of the process structure of the present invention.

[0014] Figure 2 It is a schematic diagram of the structure of the intelligent inspection system for distribution network of the present invention.

[0015] Figure 3 This is a schematic diagram of the structure of the analysis module of the present invention. DETAILED DESCRIPTION

[0016] 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.

[0017] See also Figures 1 to 3The present invention provides a technical solution: an intelligent inspection method for a distribution network, comprising the following steps: S1: Build an intelligent inspection system for distribution network; S2: Divide the area covered by the distribution network into different zones; S3: Obtaining distribution network inspection related information; S4: Analyze the obtained distribution network inspection related information; S5: Manual verification of the analysis results; S6: Handle the situation accordingly based on the specific circumstances.

[0018] The distribution network intelligent inspection system includes a central processing unit module, a partition module, a storage module, a data acquisition module, an analysis module, a verification module and a post-processing module. The central processing unit module is bidirectionally connected with the partition module, the storage module and the data acquisition module. The data acquisition module is connected with the analysis module, the analysis module is connected with the verification module, and the verification module is connected with the post-processing module. The partition module is used to partition the area covered by the distribution network; The data acquisition module is used to obtain relevant information data of distribution network inspection; The analysis module is used to analyze the inspection-related information data obtained by the data acquisition module; The verification module is used to manually verify the analysis results obtained by the analysis module to determine whether the analysis results are correct; The post-processing module is used to perform processing operations based on the results obtained by the analysis module and the verification module.

[0019] The zoning of the area covered by the distribution network is to divide the area covered by the distribution network into ordinary areas, hidden danger areas and dynamic areas. The inspection intensity in ordinary areas is lower than that in hidden danger areas. The inspection intensity in dynamic areas needs to be determined according to specific circumstances. For example, there is a temporary construction site in a certain area, and the construction site exists for six months. During the six months of construction, the failure rate of the distribution network in the area may be greater than at other times; or in an area where migratory birds live, when migratory birds are present, the frequency of bird damage to the distribution network will definitely be greater than when the migratory birds are not there. Therefore, the inspection intensity in the dynamic area needs to be determined according to the specific situation in order to improve resource utilization efficiency.

[0020] Hazard areas include areas with high objects, large spans, important cross-over areas, areas near densely populated areas or flammable and explosive places, external force collision areas, wind and flood prevention areas, and other specific hazard areas in certain areas, such as wildfire areas, lightning strike areas, etc. The high-object area is explained with poles and towers and higher objects as objects. The higher objects are high-pole plants, etc. Through 3D laser modeling of the transmission line, the distance between the wire and the top of the object is calculated, and the hidden danger section between the transmission line and the higher object is analyzed. According to the regulations of the line management unit, different hidden danger levels are determined. Generally, the closer the distance between the wire and the top of the object, the higher the hidden danger level, especially high-pole plants. Such objects that can have a certain growth rate are more likely to affect the line, and inspection efforts need to be strengthened; According to the laser point cloud tower data, the tension section with a fixed span of 1000 meters or more or a tower height of 100 meters or more is a long span section; Important crossing areas are those where transmission lines cross railways, expressways, first-class highways, first-class and second-class navigable rivers, special pipelines, tramways and cableways, and where 500kV overhead transmission lines cross 500kV and above AC and DC overhead transmission lines; Transmission lines within 100 meters of schools, markets, gas stations, etc. are located near densely populated areas or flammable and explosive places; The anti-external force collision zone is where the tower is located within 10 meters of the roadside, within 50 meters of the riverside, or in an area where vehicles and ships can easily reach and collide with external forces; For wind and flood protection, the tower foundation is located within 10 meters of the river or hillside, and the inspection priority is external force damage impact > important intersections / near densely populated areas or flammable and explosive places > prevention of external force collision > others.

[0021] There are two methods for obtaining distribution network inspection-related information in S3. One is to obtain relevant image information of the distribution network, and the other is to detect the power consumption-related data of the distribution network through sensors. The methods for obtaining distribution network-related image information are manual photography, drone photography, and monitoring probes equipped on distribution lines. Sensors are used to detect line voltage, current, equipment temperature, humidity, and vibration. Other required data about the line can also be obtained, such as line load.

[0022] The analysis module includes a picture analysis unit and a circuit data analysis unit; The image analysis unit automatically preprocesses the image using a scale scaling algorithm (adjusting it to a uniform scale) and stores it in the target detection image folder. The image is analyzed and detected by the Faster-RCNN model that has been trained to improve the detection performance. The detection result is obtained after the image to be detected is detected by the model. The Faster-RCNN model combines the region proposal network and the convolutional neural network. The convolutional neural network uses a convolution kernel of a certain scale to perform convolution calculations on the image data, which can well extract image edges, textures and other features. It has a very good effect in the field of image analysis and target detection. Its basic structure includes input layer, convolution layer, pooling layer, fully connected layer and output layer. In the convolution layer, the parameters in the convolution kernel are used as weights to perform weighted summation with the local input and add the bias value. The output result is activated by the activation function to obtain nonlinearity, and then the local maximum value or average value is selected through the pooling layer to reduce the data dimension to reduce the degree of overfitting. The convolution layer and the pooling layer are set alternately, and the convolution and pooling operations are performed multiple times to fully extract image features and train a more robust neural network. Increasing the depth (number of layers of the neural network) or width (number of convolution kernels or neurons in each layer) of the model is the safest way to obtain a high-quality model. However, this method results in too many parameters, which increases the computational complexity. It is easy to overfit when the training data set is limited. In addition, when the number of network layers is large, the gradient diffusion problem is prone to occur, making the model difficult to optimize. The circuit data analysis unit sets a threshold value for all required distribution network data information. The circuit data analysis unit compares and analyzes the data detected by the sensor and the corresponding data. The detected data is judged to be normal when it is within the threshold value, and it is judged to be abnormal when the detected data is outside the threshold value.

[0023] The analysis results of the analysis module are transmitted to the verification module and displayed to the staff. The staff judge and analyze the analysis results. If an abnormal situation occurs, the post-processing module will handle the situation accordingly. For example, a picture shows a person flying a kite near the line. The wires and the kite in the picture overlap. The analysis module incorrectly judges that a foreign object is hanging on the wire. The staff can judge the analysis result as an analysis error and do not need to handle the corresponding abnormal situation. If a sensor somewhere detects that a transformer in a certain place is out of power, the staff will verify the data and believe that there is indeed an abnormality. The post-processing module will issue an abnormal information report to facilitate maintenance.

[0024] The present invention divides the area covered by the distribution network into ordinary areas, potential danger areas and dynamic areas, reasonably allocates the resources required for inspection according to the importance of the zones, obtains inspection-related information through two methods, and the analysis module can analyze the inspection information by itself, and then increase the accuracy through manual verification, which can effectively improve the inspection efficiency, shorten the inspection cycle, greatly improve the quality of inspection work, and effectively reduce the safety operation risk of the power grid.

[0025] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent inspection method for a distribution network, characterized in that: The following steps are included: S1: Build an intelligent inspection system for distribution network; S2: Divide the area covered by the distribution network into different zones; S3: Obtaining distribution network inspection related information; S4: Analyze the obtained distribution network inspection related information; S5: Manual verification of the analysis results; S6: Handle the situation accordingly based on the specific circumstances.

2. The intelligent inspection method for a distribution network according to claim 1, characterized in that: The distribution network intelligent inspection system includes a central processing unit module, a partition module, a storage module, a data acquisition module, an analysis module, a verification module and a post-processing module. The central processing unit module is bidirectionally connected to the partition module, the storage module and the data acquisition module. The data acquisition module is connected to the analysis module, the analysis module is connected to the verification module, and the verification module is connected to the post-processing module. The partitioning module is used to partition the area covered by the distribution network; The data acquisition module is used to acquire relevant information data of the distribution network inspection; The analysis module is used to analyze the inspection related information data obtained by the data acquisition module; The verification module is used to manually verify the analysis results obtained by the analysis module to determine whether the analysis results are correct; The post-processing module is used to perform processing operations according to the results obtained by the analysis module and the verification module.

3. The intelligent inspection method for distribution network according to claim 1, characterized in that: The zoning of the area covered by the distribution network is to divide the area covered by the distribution network into ordinary areas, hidden danger areas and dynamic areas. The inspection intensity in ordinary areas is lower than that in hidden danger areas.

4. The intelligent inspection method for distribution network according to claim 3 is characterized in that: The risk areas include areas with high objects, large spans, important cross-span areas, areas near densely populated areas or flammable and explosive places, areas with external force collisions, and areas for wind and flood prevention; The high object area is described with poles and tall objects as objects. By performing three-dimensional laser modeling on the transmission line, the distance between the wire and the top of the object is calculated, the hidden danger section between the transmission line and the tall object is analyzed, and different hidden danger levels are determined according to the regulations of the line management unit; According to the laser point cloud tower data, the long span section is defined as a tension section with a spacing greater than or equal to 1000 meters or a tower height of 100 meters or more; The important crossing areas are the crossing sections where the transmission lines cross railways, expressways, first-class highways, first-class and second-class navigable rivers, special pipelines, tramways and cableways, and the crossing sections where 500kV overhead transmission lines cross 500kV and above AC and DC overhead transmission lines; The line area near densely populated areas or flammable and explosive places refers to schools, markets, gas stations, etc. The power transmission lines within 100 meters around them are near densely populated areas or flammable and explosive areas; The anti-external force collision zone is where the tower is located within 10 meters of the roadside, within 50 meters of the riverside, or in an area where vehicles and ships are easily accessible and where external force collision may occur; The wind and flood protection means that the tower foundation is located within 10 meters of the riverside and hillside, which is a wind and flood protection area.

5. The intelligent inspection method for distribution network according to claim 1, characterized in that: There are two methods for obtaining distribution network inspection-related information in S3. One is to obtain relevant image information of the distribution network, and the other is to detect the power consumption-related data of the distribution network through sensors. The methods for obtaining distribution network-related image information are manual photography, drone photography, and monitoring probes equipped on distribution lines. Sensors are used to detect line voltage, current, equipment temperature and humidity, and vibration.

6. The intelligent inspection method for distribution network according to claim 2, characterized in that: The analysis module includes a picture analysis unit and a circuit data analysis unit; The image analysis unit automatically pre-processes the image using a scale scaling algorithm and stores the image in a target detection image folder, analyzes and detects the image using a Faster-RCNN model trained to improve detection performance, and obtains a detection result after the image to be detected is detected by the model; The circuit data analysis unit sets a threshold value for all required distribution network data information. The circuit data analysis unit compares and analyzes the data detected by the sensor and the corresponding data. The detected data is judged to be normal when it is within the threshold value, and it is judged to be abnormal when the detected data is outside the threshold value.

7. The intelligent inspection method for distribution network according to claim 2, characterized in that: The analysis results of the analysis module are transmitted to the verification module and displayed to the staff, who judge and analyze the analysis results. If an abnormal situation occurs, the situation is handled accordingly through the post-processing module.

Images