Equipment remote inspection system based on edge computing

Abstract

The invention discloses an equipment remote inspection system based on edge computing, relates to the technical field of equipment monitoring, and solves the problem that equipment inspection data is generated slowly due to uneven distribution of computing power resources when a plurality of pieces of equipment execute an acquisition process synchronously. The corresponding inspection period is determined based on the confirmed data period characteristics, then the associated timeline of the corresponding equipment is determined based on the related time point of the inspection period, and the corresponding inspection time period is calibrated in the associated timeline, so that a better data characteristic confirmation effect can be effectively achieved, the subsequent inspection data processing process is facilitated, and the inspection efficiency is improved. On the basis of the computing power resources held by the corresponding inspection system, when the inspection cycle arrives, the demand resources of the corresponding associated equipment in the coincidence time period are reasonably allocated, normal generation of the inspection data in the coincidence time period can be guaranteed, and a good inspection data generation effect is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment monitoring, and in particular to an equipment remote inspection system based on edge computing. Background Art

[0002] Regarding the inspection process of the internal equipment of the power station, the main transformer, primary equipment site, switch room, capacitor room, relay protection room, and main control room of the booster station are generally inspected remotely to check for abnormal conditions of the equipment and operating environment. The opening of the wind turbine tower door in the wind farm site, the sinking and cracking around the tower base, the start and stop of the wind turbine, and the operating status of the box transformer are inspected remotely; the internal environment of the wind turbine cabin, control cabinet, hydraulic station, etc. are inspected remotely.

[0003] The patent application with the publication number CN115224803A discloses a remote patrol system and method for substation automation equipment, including a master station client, an output device, a substation server and an in-station collection unit; the master station client runs a patrol strategy configuration software for generating patrol strategy instructions; the in-station collection unit is used to patrol the substation equipment according to the requirements of the patrol strategy instructions and generate a patrol result file; the substation server is respectively connected to the master station client and the in-station collection unit to forward the patrol strategy instructions to the in-station collection unit, generate a status information message according to the patrol result file and send it to the master station client; the output device is connected to the master station client to output the information corresponding to the status information message. It improves the patrol efficiency.

[0004] In the actual equipment inspection process, the feature-based master control system generally controls the associated inspection equipment to perform data inspection on the designated equipment. However, in the actual inspection process, each device needs to execute different collection protocols during the inspection. When multiple devices execute the collection process synchronously, the uneven distribution of computing resources will cause slow generation of equipment inspection data, poor inspection effect, and a long time period. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides a remote inspection system for equipment based on edge computing, which solves the problem of slow generation of equipment inspection data due to uneven distribution of computing resources when multiple devices are synchronously executing the collection process.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a remote inspection system for devices based on edge computing, comprising: The device cycle feature collection end confirms the data cycle features of different monitored remote devices, where the data cycle features are preset features and include an interval period and a duration period; The inspection data confirmation end confirms the inspection period belonging to the remote device based on the confirmed data cycle characteristics of the different remote devices, and transmits the confirmed inspection period belonging to the different remote devices to the associated feature processing end, specifically in the following manner: Use the current moment as the calibration moment to confirm the subsequent associated timeline; Based on the interval period T of the data cycle characteristics of the corresponding remote device i And the duration Z i , calibrate the inspection period in the associated timeline, where i represents different remote devices, confirm the data generated by the remote device before the calibration time, and calibrate the end time of the data generation as J1; Based on the end time J1 and the interval period T i , determine the first moment on the associated timeline, where the first moment is J1 extended by T i The confirmed corresponding time, and then according to the confirmed first time and the duration period Zi, confirm the first group of inspection time periods, at the end time of the first group of inspection time periods, based on T i and Z i , and then confirm the second group of inspection time periods, and so on, confirm each inspection time period one by one on the associated time line of the corresponding remote device, and transmit the associated time line confirmed by the corresponding remote device to the associated feature processing end; The associated feature processing end, based on the associated timeline of different remote devices, identifies whether there are overlapping inspection periods of subsequent multiple remote devices, and based on the identified overlapping period, confirms the associated remote device and marks it as an associated device, and confirms the required computing power range of the associated device corresponding to the overlapping period from the historical associated data. The specific method is as follows: Based on the inspection periods of different remote device associated time lines, the inspection periods with overlapping time are marked as overlapping periods, and several remote devices associated with the overlapping periods are marked as associated devices; From the historical associated data, identify the computing power resource data generated by the corresponding associated device within a continuous period, and sort the confirmed several groups of computing power resource data, determine the computing power data sequence in an ascending order, and determine the clustering sequence from the computing power data sequence: identify the computing power difference Cz between adjacent computing power data from the computing power data sequence, and Cz≥0, calibrate the computing power difference Cz≤Y1 as the standard deviation value, calibrate the computing power data associated with the standard deviation value as similar data, and do not calibrate the computing power difference Cz>Y1. Based on the similar data confirmed in sequence, lock the similar data columns that appear successively, where Y1 is a preset value, and from the confirmed several groups of similar data columns, lock the similar data column with the largest number of data, and calibrate this similar data column as the selected data column, and confirm the required computing power range belonging to this associated device based on the maximum and minimum values ​​of the selected data column; The data execution processing end reasonably allocates the computing power resources associated with this patrol system based on the required computing power range confirmed by the corresponding associated equipment during the overlap period, determines the computing power resources associated with the corresponding associated equipment, and executes the computing power resources when the overlap period arrives. The specific method is as follows: Based on the different required computing power intervals confirmed by different associated devices during the overlapping period, the middle value of the corresponding required computing power interval is locked, where the middle value is the average of the two groups of endpoint values ​​of the computing power interval, and the middle value confirmed by the corresponding associated device is calibrated as Z k , where k represents different associated devices; The intermediate value Z confirmed by several groups of associated devices k Perform ratio processing to determine the ratio sequence of several groups of associated devices, and propose k=1, 2, ..., n, where n represents the total number of associated devices in this overlap period. The confirmed ratio sequence is expressed as: Z 1 :Z 2 :……:Z n ; Based on the computing power resources associated with this patrol system and the confirmed ratio sequence, the computing power resources allocated to the corresponding associated equipment are confirmed, and based on the confirmed overlap period, when the overlap period arrives, the confirmed computing power resources are promptly allocated to collect patrol data of the associated equipment within a continuous period.

[0007] Preferably, the interval period is to generate a set of data in each corresponding time period, and the duration period is the duration of the corresponding data generation.

[0008] Preferably, it also includes: an optimization adjustment end, based on the computing power resources allocated to the corresponding associated devices in the overlap period, monitoring the acquisition rate of the corresponding associated devices in the overlap period, and adjusting the computing power resources in real time based on the monitored acquisition rate, so as to further improve the inspection data acquisition rate in the overlap period; the specific method is: Based on the allocated computing resources, control the associated devices to collect inspection data within a continuous period, reduce the computing resources of each group of associated devices, and simultaneously monitor whether their collection rate is reduced: If the acquisition rate is not reduced, the associated device is calibrated as an adjustable device; Based on the confirmed adjustable devices, the computing power resources are adjusted down until the acquisition rate of the adjustable devices decreases, and the reduction of the computing power resources is stopped, and the reduced computing power resources are reasonably allocated to other related devices.

[0009] Preferably, when the computing power resources of the associated device are adjusted downward, if there is a reduction in the acquisition rate, no calibration is performed.

[0010] The present invention provides a remote inspection system for equipment based on edge computing. Compared with the prior art, it has the following beneficial effects: The present invention confirms the data cycle characteristics of each group of equipment, determines the corresponding inspection cycle based on the confirmed data cycle characteristics, and then determines the associated time line of the corresponding equipment based on the relevant time points of the inspection cycle. The associated time line is marked with the corresponding inspection period, which can effectively achieve a better data feature confirmation effect and facilitate the subsequent data processing process of the inspection. Based on the computing power resources held by the corresponding inspection system, when the inspection cycle arrives, the demand resources of the corresponding related equipment in the overlapping period are reasonably allocated, which can ensure the normal generation of inspection data in the overlapping period and achieve better inspection data generation effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a schematic diagram of the principle framework of the present invention; Figure 2 It is a schematic diagram of determining the adjustable device of the present invention. DETAILED DESCRIPTION

[0012] 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.

[0013] First embodiment See also Figure 1, the present application provides an edge computing-based equipment remote inspection system, including an equipment periodic feature collection end, an inspection data confirmation end, an associated feature processing end, a data execution processing end, and an optimization adjustment end, wherein the equipment periodic feature collection end, the inspection data confirmation end, the associated feature processing end, and the data execution processing end are electrically connected from an output node to an input node, and the optimization adjustment end and the data execution processing end are bidirectionally connected; Among them, the equipment cycle feature collection end confirms the data cycle features of different monitored remote devices. The data cycle features are preset features, which are prepared in advance by relevant operators based on the data generation situation of the equipment. The data cycle features include interval period and duration period. The interval period is a set of data generated in each interval corresponding period (that is, a set of data is generated every 5 minutes, and 5 minutes belongs to the interval period of this equipment), and the duration period is the duration of the corresponding data generation; Among them, the inspection data confirmation end confirms the inspection period belonging to the remote device based on the confirmed data cycle characteristics of different remote devices, and transmits the confirmed inspection period belonging to different remote devices to the associated feature processing end. The specific method of determining the inspection period is: The current time is used as the calibration time to confirm the subsequent associated timeline. The length of the associated timeline is the set value, which is generally 72h or 48h. Based on the interval period T of the data cycle characteristics of the corresponding remote device i And the duration Z i , calibrate the inspection period in the associated timeline, where i represents different remote devices, confirm the data generated by the remote device before the calibration time, and calibrate the end time of the generated data as J1 (that is, the corresponding relevant data generated by the remote device at the current time, the corresponding relevant data has the corresponding time characteristics, and the end time can be determined); Based on the end time J1 and the interval period T i , determine the first moment on the associated timeline, where the first moment is J1 extended by T i The confirmed corresponding time, and then according to the confirmed first time and the duration period Zi, confirm the first group of inspection time periods, at the end time of the first group of inspection time periods, based on T i and Z i , and then confirm the second group of inspection time periods, and so on, confirm each inspection time period one by one on the associated time line of the corresponding remote device, and transmit the associated time line confirmed by the corresponding remote device to the associated feature processing end; Specifically, the current time is 14:03, and the proposed setting value is 6h, so the time length of the associated timeline is: 14:03-20:03, the interval period of the corresponding equipment is 1h, and the continuous period is 3min. From the historical relevant data, the end time of this equipment is identified, and its end time is 14:01. Then, a group of interval periods of 1h are determined in the future, and the first time can be determined from the associated timeline. The first time is 15:01. Because the continuous period is 3min, the first group of inspection time periods determined is: 15:01-15:04, and this time period is marked in this associated timeline. Similarly, other inspection time periods in this associated timeline are confirmed one by one, and relevant calibration is performed.

[0014] Second embodiment In the specific implementation process of this embodiment, compared with the above embodiment, the main execution end of this embodiment is the associated feature processing end; Its associated feature processing end, based on the associated timeline of different remote devices, identifies whether there are overlapping inspection periods of subsequent multiple remote devices, and based on the identified overlapping period, confirms the associated remote device and marks it as an associated device, and confirms the required computing power interval of the associated device corresponding to the overlapping period from the historical associated data. Specifically, after the associated timeline is confirmed, the overlapping period of several remote devices can be confirmed. When there is a corresponding overlapping period, it means that during the overlapping period, there is inspection data communication between multiple remote devices. Different inspection data are associated with different inspection devices, so the related inspection protocols are also different, which will result in different computing power data, and it is necessary to confirm the required computing power interval. Among them, the specific method of confirming whether there is an overlapping period is: Based on the inspection periods of different remote device associated time lines, the inspection periods with overlapping time are marked as overlapping periods, and several remote devices associated with the overlapping periods are marked as associated devices; From the historical associated data, identify the computing power resource data generated by the corresponding associated device in a continuous period. The corresponding associated device will perform data inspection and collection in the continuous period, and adjust the collection protocol in real time to complete the data collection process, and sort the confirmed groups of computing power resource data, and determine the computing power data sequence in an ascending order, and determine the clustering sequence from the computing power data sequence: identify the computing power difference Cz between adjacent computing power data from the computing power data sequence, and Cz≥0, calibrate the computing power difference Cz≤Y1 as the standard deviation value, and calibrate the computing power data associated with the standard deviation value as the same type of data, and for Cz>Y1 The computing power difference is not calibrated. Based on the confirmed similar data in sequence, the data columns of the same type that appear successively are locked (that is, starting from the initial computing power data, the computing power data sequence that is always in Cz≤Y1 is used to determine the data columns of the same type. When Cz>Y1, the subsequent data columns of the same type that appear are confirmed), where Y1 is a preset value, and its specific value is determined by the operator based on experience. From the confirmed groups of similar data columns, the similar data column with the largest number of data is locked, and this similar data column is calibrated as the selected data column. Based on the maximum and minimum values ​​of the selected data column, the required computing power range belonging to this associated device is confirmed; Specifically, the proposed confirmed computing power data sequence is: {1, 2, 3, 4, 7, 8, 9, ...}, where the determined Y1 takes the value of 1. Starting from the first group of data in this sequence, the relevant confirmation of the data sequence is performed. Therefore, the first group of similar data sequences confirmed is: {1, 2, 3, 4}, and the second group of similar data sequences confirmed is: {7, 8, 9}. Therefore, this method can effectively confirm each subsequent data sequence in turn.

[0015] Third embodiment Among them, the data execution processing end, based on the required computing power range confirmed by the corresponding associated equipment during the overlap period, reasonably allocates the computing power resources associated with this patrol system, determines the computing power resources associated with the corresponding associated equipment, and executes the computing power resources when the overlap period arrives. The specific method of reasonable allocation is: Based on the different required computing power intervals confirmed by different associated devices during the overlapping period, the middle value of the corresponding required computing power interval is locked, where the middle value is the average of the two groups of endpoint values ​​of the computing power interval, and the middle value confirmed by the corresponding associated device is calibrated as Z k , where k represents different associated devices; The intermediate value Z confirmed by several groups of associated devices k Perform ratio processing to determine the ratio sequence of several groups of associated devices, and propose k=1, 2, ..., n, where n represents the total number of associated devices in this overlap period. The confirmed ratio sequence is expressed as: Z 1 :Z 2 :……:Zn ; Based on the computing power resources associated with this patrol system and the confirmed ratio sequence, the computing power resources allocated to the corresponding associated equipment are confirmed, and based on the confirmed overlap period, when the overlap period arrives, the confirmed computing power resources are promptly allocated to collect patrol data of the associated equipment within a continuous period.

[0016] Specifically, when there is an overlapping period, different related devices require different computing resources. If the computing resources are not allocated reasonably, some devices will generate normal inspection data during inspection, while some devices will generate delayed inspection data during inspection. This will cause the overall inspection data in the inspection process to be delayed, resulting in incomplete inspection data and the inability to guarantee the overall inspection effect of different related devices. Because the inspection equipment includes macro cameras, card cameras, ball cameras, visible light gimbals, dual-spectrum gimbals, AR Eagle Eyes, full-color tube cameras, smart monitoring, AI microphones and other related equipment, it is necessary to allocate computing resources reasonably.

[0017] Its optimization adjustment end, combined with Figure 2 Based on the computing resources allocated to the corresponding associated devices during the overlap period, the acquisition rate of the corresponding associated devices during the overlap period is monitored, and the computing resources are adjusted in real time based on the monitored acquisition rate to further improve the inspection data acquisition rate during the overlap period. The specific method of real-time adjustment is: Based on the allocated computing resources, control the associated devices to collect inspection data within a continuous period, reduce the computing resources of each group of associated devices, and simultaneously monitor whether their collection rate is reduced: If the acquisition rate is not reduced, the associated device is calibrated as an adjustable device; If the acquisition rate is reduced, no calibration is performed; Based on the confirmed adjustable devices, the computing power resources are reduced until the collection rate of the adjustable devices decreases, and the reduction is stopped. The reduced computing power resources are reasonably allocated to other related devices, so as to achieve better inspection data collection effects.

[0018] Fourth embodiment The specific implementation process of this embodiment includes all the implementation processes of the above three groups of embodiments.

[0019] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0020] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. The equipment remote inspection system based on edge computing is characterized by: include: The device cycle feature collection end confirms the data cycle features of different monitored remote devices, where the data cycle features are preset features and include an interval period and a duration period; The inspection data confirmation end confirms the inspection time period belonging to the remote device based on the confirmed data cycle characteristics of the different remote devices, and transmits the confirmed inspection time period belonging to the different remote devices to the associated characteristic processing end; The associated feature processing end identifies whether there are overlapping inspection periods of subsequent multiple remote devices based on the associated timelines of different remote devices, and based on the identified overlapping periods, confirms the associated remote devices and marks them as associated devices, and confirms the required computing power interval of the associated devices corresponding to the overlapping periods from the historical associated data; The data execution and processing end reasonably allocates the computing power resources associated with this patrol system based on the required computing power range confirmed by the corresponding associated equipment during the overlapping period, determines the computing power resources associated with the corresponding associated equipment, and executes the allocation of computing power resources when the overlapping period arrives.

2. The device remote inspection system based on edge computing according to claim 1 is characterized in that: The interval period is to generate a set of data in each corresponding time period, and the duration period is the duration of the corresponding data generation.

3. The device remote inspection system based on edge computing according to claim 1 is characterized in that: The specific method of determining the inspection cycle at the inspection data confirmation end is as follows: Use the current moment as the calibration moment to confirm the subsequent associated timeline; Based on the interval period T of the data cycle characteristics of the corresponding remote device i And the duration Z i , calibrate the inspection period in the associated timeline, where i represents different remote devices, confirm the data generated by the remote device before the calibration time, and calibrate the end time of the data generation as J1; Based on the end time J1 and the interval period T i , determine the first moment on the associated timeline, where the first moment is J1 extended by T i The confirmed corresponding time, and then according to the confirmed first time and the duration period Zi, confirm the first group of inspection time periods, at the end time of the first group of inspection time periods, based on T i and Z i , and then confirm the second group of inspection time periods, and so on, confirm each inspection time period one by one on the associated time line of the corresponding remote device, and transmit the associated time line confirmed by the corresponding remote device to the associated feature processing end.

4. The device remote inspection system based on edge computing according to claim 1 is characterized in that: The specific method of the associated feature processing end to confirm the required computing power range of the associated device is: Based on the inspection periods of different remote device associated time lines, the inspection periods with overlapping time are marked as overlapping periods, and several remote devices associated with the overlapping periods are marked as associated devices; From the historical associated data, identify the computing power resource data generated by the corresponding associated device within a continuous period, and sort the confirmed several groups of computing power resource data, determine the computing power data sequence in an ascending order, and determine the clustering sequence from the computing power data sequence: identify the computing power difference Cz between adjacent computing power data from the computing power data sequence, and Cz≥0, calibrate the computing power difference Cz≤Y1 as the standard deviation value, calibrate the computing power data associated with the standard deviation value as similar data, and do not perform any calibration on the computing power difference Cz>Y1. Based on the similar data confirmed in sequence, lock the similar data columns that appear successively, where Y1 is a preset value, and from the confirmed several groups of similar data columns, lock the similar data column with the largest number of data, and calibrate this similar data column as the selected data column, and confirm the required computing power range belonging to this associated device based on the maximum and minimum values ​​of the selected data column.

5. The device remote inspection system based on edge computing according to claim 4 is characterized in that: The specific method of allocating computing resources to the data execution processing end is as follows: Based on the different required computing power intervals confirmed by different associated devices during the overlapping period, the middle value of the corresponding required computing power interval is locked, where the middle value is the average of the two groups of endpoint values ​​of the computing power interval, and the middle value confirmed by the corresponding associated device is calibrated as Z k , where k represents different associated devices; The intermediate value Z confirmed by several groups of associated devices k Perform ratio processing to determine the ratio sequence of several groups of associated devices, and propose k=1, 2, ..., n, where n represents the total number of associated devices in this overlap period. The confirmed ratio sequence is expressed as: Z1: Z2: ...: Z n ; Based on the computing power resources associated with this patrol system and the confirmed ratio sequence, the computing power resources allocated to the corresponding associated equipment are confirmed, and based on the confirmed overlap period, when the overlap period arrives, the confirmed computing power resources are promptly allocated to collect patrol data of the associated equipment within a continuous period.

6. The device remote inspection system based on edge computing according to claim 5 is characterized in that: Also includes: The optimization and adjustment end monitors the collection rate of the corresponding associated devices during the overlapping period based on the computing power resources allocated to the corresponding associated devices during the overlapping period, and adjusts the computing power resources in real time based on the monitored collection rate, so as to further improve the inspection data collection rate during the overlapping period.

7. The device remote inspection system based on edge computing according to claim 6 is characterized in that: The specific method of the optimization and adjustment end to adjust the computing power resources in real time is: Based on the allocated computing resources, control the associated devices to collect inspection data within a continuous period, reduce the computing resources of each group of associated devices, and simultaneously monitor whether their collection rate is reduced: If the acquisition rate is not reduced, the associated device is calibrated as an adjustable device; Based on the confirmed adjustable devices, the computing power resources are adjusted down until the acquisition rate of the adjustable devices decreases, and the reduction of the computing power resources is stopped, and the reduced computing power resources are reasonably allocated to other related devices.

8. The device remote inspection system based on edge computing according to claim 7 is characterized in that: When the computing power resources of the associated device are lowered, there is a decrease in the acquisition rate, and no calibration is performed.

Images