A safety inspection abnormal data warning system

Abstract

The present invention provides a safety inspection abnormal data warning system, which is characterized by including a path storage unit, an inspection unit, a data acquisition unit, an equipment ledger unit, and a data processing and warning unit. The data processing and warning unit includes a difference calculation unit, a processing unit, a standard value unit, an equipment maintenance plan unit, and an alarm unit. The inspection unit conducts inspections according to a preset inspection path. During the inspection process, the data acquisition unit collects the data of the equipment and the surrounding environment on the inspection path, and the collected data is sent to the difference calculation unit for calculation. The difference calculation unit calculates and compares the collected data with the standard value. If the comparison result has a large difference, the processing unit extracts the real-time maintenance plan of each equipment and judges whether there is a new maintenance plan for the corresponding equipment. If there is no new maintenance plan for the corresponding equipment, the processing unit judges that abnormal data is stored and controls the alarm unit to issue an alarm to prompt the staff to process it in time.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid security inspection, and particularly relates to a safety inspection abnormal data warning system. Background Art

[0002] During the operation of the power grid, it is necessary to ensure the normal operation of each device. Therefore, power grid companies often set up corresponding inspection personnel to regularly inspect each device in the power grid to determine whether there are abnormalities in the device and the surrounding environment. However, the judgment of abnormalities by inspection personnel is subjective, and inspection personnel only judge whether there are abnormalities by comparing the currently collected data with the standard data. During the daily operation of the power grid, corresponding adjustments will be made to meet the electricity demand, resulting in changes in the data of the device and its surrounding environment. If inspection personnel still judge based on their own experience at this time, judgment errors will occur. Summary of the Invention

[0003] In view of this, the present invention proposes a safety inspection abnormal data warning system, which can judge whether the changes in the data of the device and its surrounding environment are caused by the maintenance plan. If it is not caused by the maintenance plan, an alarm will be issued to notify the staff of data abnormalities.

[0004] The technical solution of the present invention is realized as follows:

[0005] A safety inspection abnormal data warning system includes:

[0006] A path storage unit for storing the inspection path;

[0007] An inspection unit for moving along the inspection path;

[0008] A data collection unit for collecting data of devices and their surrounding environment on the inspection path;

[0009] A device ledger unit for collecting data of devices and their surrounding environment on the inspection path during the second inspection;

[0010] A data processing and warning unit for judging whether there are abnormalities according to the data of the device and its surrounding environment, and giving an alarm when abnormalities occur;

[0011] The data processing and warning unit includes a difference calculation unit, a processing unit, a standard value unit, a device maintenance plan unit and an alarm unit. The processing unit is respectively connected to the difference calculation unit, the standard value unit, the device maintenance plan unit and the alarm unit for data connection. The difference calculation unit is respectively connected to the data collection unit and the device ledger unit for data connection. The standard value unit stores data standard values, and the device maintenance plan unit stores the real-time maintenance plans of each device; the inspection unit is respectively connected to the path storage unit, the data collection unit and the device ledger unit for data connection.

[0012] Preferably, the system further comprises a packaging unit, which packages the data collected by the data collection unit into data packets.

[0013] Preferably, it further comprises a timing unit, which assigns a collection time to each data packet, and the real-time maintenance plan includes the start and end times of maintenance for each device.

[0014] Preferably, it further comprises a device identification unit, wherein the device identification unit is connected to the device bill metadata.

[0015] Preferably, the device identification unit includes an image acquisition unit, a nameplate extraction unit and a category matching unit, the image acquisition unit, the nameplate extraction unit and the category matching unit are sequentially data-connected, and the category matching unit is connected to the device bill metadata.

[0016] Preferably, the category matching unit has a built-in trained neural network, the input of the neural network is the nameplate image data, and the output is the device category.

[0017] Preferably, the device and surrounding environment data include temperature data, humidity data, magnetic leakage data, current data and voltage data.

[0018] Preferably, the nameplate extraction unit includes a virtual dividing line unit, an area of interest calculation unit, a dividing line translation unit and a target extraction unit. The virtual dividing line unit divides the image data collected by the image acquisition unit into several rectangular blocks. The area of interest calculation unit calculates the area of the area of interest containing the nameplate image in the rectangular block. The dividing line translation unit translates the virtual dividing line according to the calculated areas of multiple areas of interest so that the nameplate image is located in a single rectangular block. The target extraction unit extracts the nameplate image data in the rectangular block and transmits it to the category matching unit.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] The present invention provides a safety inspection abnormal data early warning system, in which the inspection unit conducts inspections according to a preset inspection route. During the inspection process, the data acquisition unit collects data of the equipment and surrounding environment on the inspection route, records the data, and transmits the collected data to the difference calculation unit, which calculates the difference between the collected data and the standard value. If the difference result is large, the processing unit reads the real-time maintenance plan of each device stored in the equipment maintenance plan unit, and determines whether the equipment with large data differences has a new maintenance plan to avoid false alarms. If there is no new maintenance plan for the corresponding equipment, the processing unit can control the alarm unit to issue an alarm to prompt the staff to handle it in time. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only the preferred embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0022] Figure 1 It is the schematic diagram of a safety inspection abnormal data warning system of the present invention;

[0023] In the figure, 1 is a path storage unit, 2 is an inspection unit, 3 is a data acquisition unit, 4 is an equipment ledger unit, 5 is a data processing and warning unit, 6 is a difference calculation unit, 7 is a processing unit, 8 is a standard value unit, 9 is an equipment maintenance plan unit, 10 is an alarm unit, 11 is a packaging unit, 12 is a timing unit, 13 is an equipment identification unit, 14 is an image acquisition unit, 15 is a nameplate extraction unit, 16 is a type matching unit, 17 is a virtual dividing line unit, 18 is an interested area calculation unit, 19 is a dividing line translation unit, and 20 is a target extraction unit. Detailed implementation manners

[0024] To better understand the technical content of the present invention, a specific embodiment is provided below, and the present invention will be further described in conjunction with the drawings.

[0025] See Figure 1 , a safety inspection abnormal data warning system provided by the present invention includes:

[0026] The path storage unit 1 is used to store the inspection path;

[0027] The inspection unit 2 is used to move along the inspection path;

[0028] The data acquisition unit 3 is used to acquire data of equipment and the surrounding environment on the inspection path;

[0029] The equipment ledger unit 4 is used to call equipment ledger data;

[0030] The data processing and warning unit 5 is used to judge whether an abnormality occurs according to the equipment and the surrounding environment data, and give an alarm when an abnormality occurs;

[0031] The data processing and warning unit 5 includes a difference calculation unit 6, a processing unit 7, a standard value unit 8, an equipment maintenance plan unit 9, and an alarm unit 10. The processing unit 7 is respectively connected to the difference calculation unit 6, the equipment maintenance plan unit 9, and the alarm unit 10 for data connection. The difference calculation unit 6 is respectively connected to the data acquisition unit 3 and the standard value unit 8 for data connection. The standard value unit 8 stores data standard values, and the equipment maintenance plan unit 9 stores real-time maintenance plans for each device. The inspection unit 2 is respectively connected to the path storage unit 1, the data acquisition unit 3, and the equipment ledger unit 4 for data connection.

[0032] In a safety inspection abnormal data warning system of the present invention, a pre-set inspection path is stored in the path storage unit 1. The inspection unit 2 can perform a safety inspection on the power grid according to the inspection path. At the same time, the inspection unit 2 can drive the data acquisition unit 3, the equipment ledger unit 4, and the data processing and warning unit 5 to move along the inspection path. During the movement, the data acquisition unit 3 performs data acquisition, collects data on the equipment and the surrounding environment on the inspection path, and the data on the equipment and the surrounding environment will be transmitted to the data processing and warning unit 5. If there is an abnormal data situation, an alarm can be given to prompt the staff to perform maintenance in time.

[0033] Since the equipment ledger unit 4 is provided in the present invention, the ledger data of the corresponding equipment can be obtained. During the inspection, according to the ledger data of the equipment on one side of the inspection, the data acquisition unit 3 collects data on the equipment and the surrounding environment, and transmits the collected data to the difference calculation unit 6. The difference calculation unit 6 simultaneously reads the standard value stored in the standard value unit 8, calculates the collected data with the standard value, and obtains the corresponding difference result. The difference result is transmitted to the processing unit 7. The processing unit determines whether the difference result is large. When the difference is large, the processing unit 7 reads the real-time maintenance plans of each device stored in the equipment maintenance plan unit 9, finds out the real-time maintenance plan of the device with a large difference result, and determines whether there is a new maintenance plan for the device that causes the collected data to fluctuate greatly. If there is no new maintenance plan, the processing unit 7 controls the alarm unit 10 to give an alarm to prompt the staff to perform repair or maintenance. If there is a new maintenance plan for the device, the staff can consider whether the change in the maintenance plan causes the change in the collected data to avoid false alarms.

[0034] For the data on the equipment and the surrounding environment collected, which includes various data of multiple devices, the difference calculation unit 6 compares the collected value of the same data of the same device with the standard value to determine whether there is data abnormality for the device.

[0035] Preferably, it further includes a packaging unit 11, and the packaging unit 11 packages the data collected by the data acquisition unit 3 into data packets.

[0036] After the set packaging unit 11 packs the collected data into data packets, it can facilitate data transmission and improve privacy.

[0037] Preferably, it further includes a timing unit 12. The timing unit 12 assigns the acquisition time to each data packet, and the real-time maintenance plan includes the start and end times of the maintenance of each device.

[0038] Since the maintenance plan will be updated in real time and the real-time maintenance plan will include the specific start and end times of the maintenance of each device, in order to ensure that the detected device data changes can be compared with the real-time maintenance plan, the present invention assigns a time to each data packet, and the assigned time is specific to microseconds. If the time when the device collects data changes is not within the time period of the updated real-time maintenance plan of the device, it can be determined that the device has data anomalies.

[0039] Preferably, it further includes a device identification unit 13, and the device identification unit 13 is connected to the device ledger unit 4 in terms of data.

[0040] There are certain differences in the data that different devices need to collect. Therefore, the present invention sets up a device identification unit 13. Before collecting the data of the device and its surrounding environment, the type of the device is first identified. After obtaining the device type, the device ledger unit 4 extracts the device ledger information, so that different data can be selectively collected according to the device ledger information.

[0041] Preferably, the device identification unit 13 includes an image acquisition unit 14, a nameplate extraction unit 15, and a type matching unit 16. The image acquisition unit 14, the nameplate extraction unit 15, and the type matching unit 16 are connected in series in terms of data. The type matching unit 16 is connected to the device ledger unit 4 in terms of data. The type matching unit 16 is built-in with a trained neural network, and the input of the neural network is the nameplate image data, and its output is the device type.

[0042] When performing device identification, first, the image acquisition unit 14 acquires device image data, and the image data contains the nameplate features of the device. Then, the nameplate extraction unit 15 extracts the nameplate image data from the image data and transmits the nameplate image data to the type matching unit 16. The type matching unit 16 is built-in with a trained neural network, which can process the nameplate image data and finally obtain the device type. Thus, the device ledger unit 4 extracts the device ledger corresponding to the device type transmitted by the type matching unit 16, so as to facilitate the data acquisition unit 3 to collect the corresponding data.

[0043] Preferably, the device and its surrounding environment data include temperature data, humidity data, magnetic leakage data, current data, and voltage data.

[0044] The data of the device itself includes current data, voltage data, etc., while the surrounding environment data includes humidity data, temperature data, leakage magnetic data, etc. Different devices require different data to be collected.

[0045] Preferably, the nameplate extraction unit 15 includes a virtual dividing line unit 17, a region of interest calculation unit 18, a dividing line translation unit 19, and a target extraction unit 20. The virtual dividing line unit 17 divides the image data collected by the image acquisition unit 14 into several rectangular blocks. The region of interest calculation unit 18 calculates the area of the region of interest containing the nameplate image within the rectangular block. The dividing line translation unit 19 translates the virtual dividing line according to the calculated areas of multiple regions of interest, so that the nameplate image is located in a single rectangular block. The target extraction unit 20 extracts the nameplate image data in the rectangular block and transmits it to the type matching unit 16.

[0046] When performing nameplate extraction, first, the virtual dividing line unit 17 divides the collected image data into several rectangular blocks through virtual lines. Then, the region of interest calculation unit 18 calculates the area of the region of interest containing the nameplate image in the rectangular block. If the nameplate image is not located in a single rectangular block, the region of interest calculation unit 18 will obtain multiple rectangular blocks containing regions of interest, and the areas of the regions of interest contained in the rectangular blocks are different. According to the area difference, the dividing line translation unit 19 translates the dividing line defined by the virtual dividing line unit 17 so that the nameplate image is located in one of the single rectangular blocks. Finally, the target extraction unit 20 extracts the rectangular block containing the nameplate image to form nameplate image data, and then sends it to the type matching unit 16 for type matching.

[0047] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A safety inspection abnormal data warning system, characterized in that Including: A path storage unit for storing inspection paths; An inspection unit for moving along the inspection path; A data acquisition unit for acquiring data of equipment and the surrounding environment on the inspection path; An equipment ledger unit for calling equipment ledger data; A data processing and warning unit for judging whether an abnormality occurs according to the equipment and surrounding environment data and giving an alarm when an abnormality occurs; The data processing and warning unit includes a difference calculation unit, a processing unit, a standard value unit, an equipment maintenance plan unit and an alarm unit. The processing unit is respectively connected to the difference calculation unit, the equipment maintenance plan unit and the alarm unit for data connection. The difference calculation unit is respectively connected to the data acquisition unit and the standard value unit for data connection. The standard value unit stores data standard values, and the equipment maintenance plan unit stores the real-time maintenance plans of each equipment. The inspection unit is respectively connected to the path storage unit, the data acquisition unit and the equipment ledger unit for data connection; The inspection unit conducts inspections according to a preset inspection path. During the inspection process, the data acquisition unit acquires data of equipment and the surrounding environment on the inspection path, records the data, and transmits the acquired data to the difference calculation unit. The difference calculation unit calculates the difference between the acquired data and the standard value. When the difference result is large, the processing unit reads the real-time maintenance plans of each equipment stored in the equipment maintenance plan unit, judges whether the equipment with large data differences has a new maintenance plan to avoid false alarms. When the corresponding equipment has no new maintenance plan, the processing unit can control the alarm unit to give an alarm to prompt the staff to process it in time.

2. The safety inspection abnormal data warning system according to claim 1, characterized in that, It further includes a packaging unit that packages the data acquired by the data acquisition unit into data packets.

3. The safety inspection abnormal data warning system according to claim 2, characterized in that, It further includes a timing unit that assigns the acquisition time to each data packet. The real-time maintenance plan includes the start and end times of the maintenance of each equipment.

4. The safety inspection abnormal data warning system according to claim 1, wherein It further includes an equipment identification unit that is connected to the equipment ledger unit for data connection.

5. The safety inspection abnormal data warning system according to claim 4, wherein, The equipment identification unit includes an image acquisition unit, a nameplate extraction unit and a type matching unit. The image acquisition unit, the nameplate extraction unit and the type matching unit are sequentially connected for data connection. The type matching unit is connected to the equipment ledger unit for data connection.

6. The safety inspection abnormal data warning system according to claim 5, characterized in that, The type matching unit is built-in with a trained neural network. The input of the neural network is nameplate image data, and its output is the equipment type.

7. A safety inspection abnormal data warning system according to claim 1 or 6, characterized in that The equipment and surrounding environment data includes temperature data, humidity data, magnetic leakage data, current data and voltage data.

8. A safety inspection abnormal data warning system according to claim 5, characterized in that The nameplate extraction unit includes a virtual dividing line unit, a region of interest calculation unit, a dividing line translation unit, and a target extraction unit. The virtual dividing line unit divides the image data collected by the image acquisition unit into several rectangular blocks. The region of interest calculation unit calculates the area of the region of interest containing the nameplate image within the rectangular block. The dividing line translation unit translates the virtual dividing line based on the calculated areas of multiple regions of interest, so that the nameplate image is located in a single rectangular block. The target extraction unit extracts the nameplate image data in the rectangular block and sends it to the category matching unit.

Images