Substation operation safety risk management and control system based on multi-view vision

Abstract

The invention discloses a substation operation safety risk management and control system based on multi-view vision, and relates to the technical field of power equipment safety management and control, the management and control system comprises a position estimation module, a position direction description module, an information matching module, a position determination module and an early warning module, the position and direction description module is used for acquiring position and direction description information of a target operator relative to other operators under each view angle; the information matching module is used for matching target operating personnel under the multiple visual angle visual pictures; the position determination module is used for calculating and determining the world coordinate position; the early warning module is used for analyzing whether the target operator enters a dangerous area; according to the safety risk management and control system, the world coordinate position of the target operator can be accurately calculated according to the position of the target operator in the multi-view visual picture, the accuracy of three-dimensional position calculation of the target operator is improved, then real-time management and control of the near-electricity operation safety distance are achieved, and the accuracy of position detection of the operator is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of power equipment safety management and control, and specifically relates to a substation operation safety risk management and control system based on multi-view vision. Background Art

[0002] As a power grid hub, substations play an important role in the power system. The unattended operation of substations is the development trend of today's power system. In order to ensure the normal operation of substations, daily operation and maintenance of substations is inevitable. On the other hand, with the rapid development of the power grid, the construction and expansion of substations are increasing. How to ensure the personal safety of workers and maintenance personnel in substations and the safety of substation equipment has become an increasingly prominent issue.

[0003] At present, the substation auxiliary integrated monitoring system that has been built in China is mainly based on the safety of the substation, and is used for security, fire protection, and environmental monitoring. However, during the monitoring process, only the operating behavior of the operator in the operating area from a single perspective can be recorded, and the exact operating distance between the operator and various equipment cannot be obtained in time, resulting in the inability to prevent safety risks in advance when working near electricity. Summary of the invention

[0004] The purpose of the present invention is to provide a substation operation safety risk management and control system based on multi-perspective vision with a simple structure and reasonable design in order to solve the above problems.

[0005] The present invention achieves the above-mentioned purpose through the following technical solutions:

[0006] A substation operation safety risk management and control system based on multi-view vision, including:

[0007] A position estimation module is used to estimate the position of the target operator at each viewing angle;

[0008] A position and direction description module is used to obtain the position and direction description information of the target operator relative to other operators at each viewing angle;

[0009] An information matching module is used to match the target operator in multiple visual images according to the position and direction description information of the target operator relative to other operators;

[0010] A position determination module is used to calculate and determine the world coordinate position of the target operator based on the two-dimensional position of the target operator in multiple visual perspectives;

[0011] The early warning module is used to analyze in real time whether the target operator has entered the danger zone based on the spatial distance value between the target equipment and the target operator, and to make an intrusion early warning signal based on the analysis results.

[0012] As a further optimization solution of the present invention, the position estimation module estimates the position of the target operator at each viewing angle through a camera imaging model, and the camera imaging model is:

[0013]

[0014] Among them, z c is the distance depth coefficient, u d 、v d is the pixel coordinate point, f cx 、f cy is the focal length of the camera, d x ,d y is the physical length of a pixel, u0 and v0 are the intersection points of the camera optical axis and the image plane, and x w ,y w 、z w is the world coordinate point, R 3*3 , T 3*1 is the camera external parameter, and M is the projection matrix.

[0015] As a further optimization scheme of the present invention, the position direction description information includes direction information and distance information. The direction information is the angular orientation information of the target operator relative to other operators in the same viewing angle picture; the distance information is the distance information between the target operator and other operators in the same viewing angle picture.

[0016] As a further optimization solution of the present invention, the step of obtaining the position and direction description information of the target operator relative to other operators at each viewing angle includes:

[0017] a. Get the three-dimensional position P of the jth person at each perspective vj =(x vj ,y vj , 0),j∈[0,...,N v ], where V is the number of viewing angles, N v is the number of operators at the Vth perspective, v∈[1,...,V];

[0018] b. Calculate the position and direction description information of the target operator relative to other operators at each viewing angle using the following formula:

[0019]

[0020] k i ∈[0,...,N v ],k i ≠j

[0021] Among them, W vjEach line has a three-dimensional position P relative to other workers. vj The direction and distance information of the target operator, W vj The first column is the relative three-dimensional position of other operators P vj The direction information of the target operator, W vj The first column is arranged in the order from 0 to 2π.

[0022] As a further optimization solution of the present invention, the matching process of the target operator under the multi-view visual picture includes:

[0023] 1) According to the degree of overlap between the position and direction description of the target operator relative to other operators, the matching scores of the target operator in one picture and the operators in other viewing angles are calculated;

[0024] 2) Setting a matching similarity threshold. If the matching score between the target operator in one picture and the operator in other viewing angle pictures is greater than the matching similarity threshold, the operator in the other picture corresponding to the matching score is taken as the matched target operator;

[0025] 3) If the number of people who match the target operator in one screen with operators in other screens exceeds 2, the operator in other screens corresponding to the largest matching score is used as the matched target operator.

[0026] As a further optimization solution of the present invention, the two-dimensional position of the target operator under multiple viewing angles is (u v ,v v ),v∈[1,...,V], then the world coordinates of the target operator are (x w ,y w ,z w ), where (x w ,y w ,z w ) is obtained by the following calculation equation:

[0027]

[0028] Among them, M v is the camera projection matrix of the vth frame.

[0029] As a further optimization scheme of the present invention, the spatial distance value is the straight-line distance d between the world coordinate position of the target operator and the spatial coordinate position of the target device, and the intrusion distance threshold of the dangerous area of ​​the target device is R. When d≤R, the early warning module issues an early warning warning to the target person based on the analysis result of the target operator entering the dangerous area.

[0030] As a further optimization scheme of the present invention, the safety risk management and control system also includes an evacuation direction module, which is used to send an evacuation signal to the target operator and guide the target operator to evacuate in a direction away from the dangerous area when the target operator enters the dangerous area.

[0031] As a further optimization scheme of the present invention, the evacuation pointing module includes a safety helmet and an early warning vibrator. A plurality of early warning vibrators are arranged circumferentially around the skull wearing area of ​​the safety helmet. The early warning vibrators are electrically connected to the early warning module. The early warning vibrators contact the skull of the target worker. When the target worker wearing the safety helmet enters the danger zone, the early warning vibrator in the direction of the target device pointing to the target worker vibrates. The direction of the vibrating early warning vibrator is the evacuation direction of the target worker.

[0032] As a further optimization solution of the present invention, the security risk management and control system includes the following steps:

[0033] Step 1: Estimate the position of the target operator at each viewing angle;

[0034] Step 2: Calculate and obtain the position and direction description information of the target operator relative to other operators at each viewing angle;

[0035] Step 3: Match the target operator in multiple visual perspectives according to the position and direction description information of the target operator relative to other operators;

[0036] Step 4: Calculate and determine the world coordinate position of the target operator based on the two-dimensional position of the target operator in multiple visual perspectives;

[0037] Step 5: Analyze in real time whether the target operator has entered the danger zone based on the spatial distance value between the target equipment and the target operator, and make an intrusion warning signal based on the analysis results.

[0038] The present invention has at least the following beneficial effects: a substation operation safety risk management and control system based on multi-perspective vision provided by the present invention includes a position estimation module, a position direction description module, an information matching module, a position determination module and an early warning module. By calculating the three-dimensional position of the target operator in real time according to the multi-perspective visual image, the three-dimensional position of the target operator in each perspective is first roughly estimated, and then the operator is matched, and then the world coordinate position of the target operator is accurately calculated according to the position of the target operator in the multi-perspective visual image, thereby improving the accuracy of the three-dimensional position calculation of the target operator, and then realizing real-time management and control of the safe distance of near-electrical operations, thereby getting rid of the constraints of electric field induction measurement, monocular ranging and the like, and improving the accuracy of operator position detection, thereby ensuring real-time supervision of the operating behavior of operators at the operating site, significantly improving the technical level of power grid safety production, and ensuring the safe operation of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is the process of the safety risk management system of the present invention Figure 1 ;

[0040] Figure 2 This is the process of the safety risk management system of the present invention Figure 2 ;

[0041] Figure 3 It is a flow chart of the steps of the risk management method applied by the safety risk management system of the present invention. DETAILED DESCRIPTION

[0042] The present application is further described in detail below in conjunction with the accompanying drawings. It is necessary to point out here that the following specific implementation methods are only used to further illustrate the present application and cannot be understood as limiting the scope of protection of the present application. Technical personnel in this field can make some non-essential improvements and adjustments to the present application based on the above application content.

[0043] like Figure 1 As shown, a substation operation safety risk management and control system based on multi-view vision includes:

[0044] A position estimation module is used to estimate the position of the target operator at each viewing angle;

[0045] A position and direction description module is used to obtain the position and direction description information of the target operator relative to other operators at each viewing angle;

[0046] An information matching module is used to match the target operator in multiple visual images according to the position and direction description information of the target operator relative to other operators;

[0047] A position determination module is used to calculate and determine the world coordinate position of the target operator based on the two-dimensional position of the target operator in multiple visual perspectives;

[0048] The early warning module is used to analyze in real time whether the target operator has entered the danger zone based on the spatial distance value between the target equipment and the target operator, and to make an intrusion early warning signal based on the analysis results.

[0049] In the above embodiment, the three-dimensional position of the target worker is calculated in real time based on the multi-perspective visual image, the three-dimensional position of the target worker in each perspective is first roughly estimated, and then the worker is matched, and then the world coordinate position of the target worker is accurately calculated based on the position of the target worker in the multi-perspective visual image, thereby improving the accuracy of the three-dimensional position calculation of the target worker, and then realizing real-time control of the safe distance of near-electrical operations, thereby getting rid of the constraints of electric field induction measurement, monocular ranging and other methods, and improving the accuracy of worker position detection.

[0050] Exemplarily, the position estimation module estimates the position of the target operator at each viewing angle through a camera imaging model, and the camera imaging model is:

[0051]

[0052] Among them, z c is the distance depth coefficient, u d 、v d is the pixel coordinate point, f cx 、f cy is the focal length of the camera, d x ,d y is the physical length of a pixel, u0 and v0 are the intersection points of the camera optical axis and the image plane, and x w ,y w 、z w is the world coordinate point, R 3*3 、T 3*1 is the camera external parameter, and M is the projection matrix.

[0053] It should be noted that the camera imaging model is a mathematical model that describes how a camera converts a three-dimensional scene in the real world into a two-dimensional image. This model is based on physical optics and geometry, and uses mathematical formulas to characterize the entire imaging process, that is, the geometric transformation relationship between the spatial point of the photographed object and the imaging point of the photo.

[0054] Assuming that the target operator is on the ground, the coordinate of the target operator on the Z axis in the world coordinate is z w = 0, according to the pixel position u of the target operator on the ground in the picture d 、v d, substitute into the above camera imaging model, and solve the X and Y axis coordinates x of the target operator in the world coordinates w ,y w , thereby roughly estimating the three-dimensional position of the target operator at each viewing angle.

[0055] It should be noted that since the target operator is located on the ground, the position of the target operator is roughly estimated at this time, that is, the world coordinate is (x w ,y w , 0).

[0056] Exemplarily, the position and direction description information includes direction information and distance information. The direction information is the angular position information of the target operator relative to other operators in the same viewing angle; the distance information is the distance information between the target operator and other operators in the same viewing angle. Through the direction information and distance information, the position of the target operator in each viewing angle relative to other operators in the viewing angle can be accurately determined.

[0057] Specifically, the steps of obtaining the position and direction description information of the target operator relative to other operators at each viewing angle include:

[0058] a. Get the three-dimensional position P of the jth person at each perspective vj =(x vj ,y vj , 0),j∈[0,...,N v ], where V is the number of viewing angles, N v is the number of operators at the Vth perspective, v∈[1,...,V];

[0059] It should be noted that the three-dimensional position P of the jth person vj =(x vj ,y vj , 0) is the world coordinate position of the target operator obtained by the above rough estimation, and the number of V viewing angles is consistent with the number of real-time monitoring cameras arranged in the working environment of the target equipment;

[0060] b. Calculate the position and direction description information of the target operator relative to other operators at each viewing angle using the following formula:

[0061]

[0062] k i ∈[0,...,N v ],k i ≠j

[0063] Among them, W vjEach line has a three-dimensional position P relative to other workers. vj The direction and distance information of the target operator, W vj The first column is the relative three-dimensional position of other operators P vj The direction information of the target operator, W vj The first column is arranged in the order from 0 to 2π.

[0064] It should be noted that in the above position and direction description information, for example, at the Vth viewing angle, there are N v workers, among which the jth person is regarded as the target worker, and his 3D coordinate position is P vj =(x vj ,y vj , 0), then the coordinate positions of other operators are Therefore, As the angular position information of the target operator relative to other operators, The distance information between the target worker and other workers.

[0065] It should be further explained that W vj The first column is arranged in the order of 0 to 2π, which means that the coordinate position of the target operator is used as the origin, and the other operators around the target operator are arranged in the order of 0 to 2π, and then the angle position information and distance information of the target operator relative to the other operators are calculated in turn, so as to finally calculate the position and position description information W vj .

[0066] Exemplarily, the matching process of the target operator in the multi-view visual picture includes:

[0067] 1) According to the degree of overlap between the position and direction description of the target operator relative to other operators, the matching scores of the target operator in one picture and the operators in other viewing angles are calculated;

[0068] 2) Setting a matching similarity threshold. If the matching score between the target operator in one picture and the operator in other viewing angle pictures is greater than the matching similarity threshold, the operator in the other picture corresponding to the matching score is taken as the matched target operator;

[0069] 3) If the number of people who match the target operator in one screen with operators in other screens exceeds 2, the operator in other screens corresponding to the largest matching score is used as the matched target operator.

[0070] It should be noted that the degree of overlap refers to the consistency of the relative position and direction of the target operator and other operators observed from different perspectives. In other words, it describes how the relative position and direction of the target operator from different perspectives match the relative position and direction of other operators. It is quantified by the matching score, which intuitively represents the visual consistency of the target operator from multiple perspectives. This score can help evaluate the degree of match between the target operator observed from different perspectives and the actual position and direction.

[0071] Through the above process, the target operators in multiple viewing angles can be accurately matched to achieve real-time monitoring of the target operators under multiple viewing angles.

[0072] Exemplarily, the two-dimensional position of the target operator in multiple viewing angles is (u v ,v v ),v∈[1,...,V], then the world coordinates of the target operator are (x w ,y w ,z w ), where (x w ,y w ,z w ) is obtained by the following calculation equation:

[0073]

[0074] Among them, M v is the camera projection matrix of the vth frame.

[0075] In the above embodiment, the world coordinate position of the target worker is obtained by accurately calculating the position of the target worker in the multi-perspective visual image, thereby improving the accuracy of the three-dimensional position calculation of the target worker, and then realizing real-time control of the safe distance of near-electrical work, thereby getting rid of the constraints of electric field induction measurement, monocular ranging and other methods, and improving the accuracy of worker position detection.

[0076] Exemplarily, the spatial distance value is the straight-line distance d between the world coordinate position of the target operator and the spatial coordinate position of the target device, and the intrusion distance threshold of the dangerous area of ​​the target device is R. When d≤R, the early warning module issues an early warning to the target operator based on the analysis result of the target operator entering the dangerous area. By timely obtaining the accurate operating distance between the target operator and the target device, the target operator can be warned and evacuated in a timely manner, effectively preventing the operator from entering the dangerous area, and improving the accuracy and timeliness of the warning.

[0077] Continue reading Figure 2The safety risk management and control system also includes an evacuation pointing module, which is used to send an evacuation signal to the target operator and guide the target operator to evacuate in a direction away from the dangerous area when the target operator enters the dangerous area.

[0078] Exemplarily, the evacuation direction module includes a safety helmet and an early warning vibrator. A plurality of early warning vibrators are arranged circumferentially around the skull wearing area of ​​the safety helmet. The early warning vibrators are electrically connected to the early warning module. The early warning vibrators contact the skull of the target worker. When the target worker wearing the safety helmet enters the danger zone, the early warning vibrator in the direction of the target device pointing to the target worker vibrates. The direction of the vibrating early warning vibrator is the evacuation direction of the target worker.

[0079] In the above embodiment, by arranging a plurality of early warning vibrators circumferentially around the top of the skull wearing area of ​​the safety helmet, when the target worker wears the safety helmet correctly and enters the danger zone, the early warning vibrator along the direction of the target device pointing to the target worker receives the early warning signal from the early warning module and vibrates, so that the target worker receives the instruction of the early warning vibrator in the first time and turns around to evacuate along the circumferential direction of the early warning vibrator relative to the body.

[0080] It should be noted that when the target workers turn around to evacuate, multiple early warning vibrators and safety helmets rotate synchronously with the target workers' turn, so that the early warning vibrators in the direction of the target equipment pointing to the target workers are replaced. Therefore, in the process of evacuating the dangerous area, the target workers are always guided by the vibration direction of the early warning vibrators in the evacuation direction, thereby ensuring a fast and accurate close-range evacuation of the dangerous area and avoiding panic and confusion when the target workers receive the signal to evacuate the dangerous area.

[0081] In one embodiment, the present invention provides a multi-view vision-based substation operation safety risk management system, which includes the following steps:

[0082] Step 1: Estimate the position of the target operator at each viewing angle;

[0083] Step 2: Calculate and obtain the position and direction description information of the target operator relative to other operators at each viewing angle;

[0084] Step 3: Match the target operator in multiple visual images according to the position and direction description information of the target operator relative to other operators;

[0085] Step 4: Calculate and determine the world coordinate position of the target operator based on the two-dimensional position of the target operator in multiple visual perspectives;

[0086] Step 5: Analyze in real time whether the target operator has entered the danger zone based on the spatial distance value between the target equipment and the target operator, and make an intrusion warning signal based on the analysis results.

[0087] Through the above method, the world coordinate position of the target operator is accurately calculated to improve the accuracy of the three-dimensional position calculation of the target operator, and then realize the real-time control of the safe distance of near-electrical operations, thereby getting rid of the constraints of electric field induction measurement, monocular ranging and other methods, and improving the accuracy of operator position detection.

[0088] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A substation operation safety risk management and control system based on multi-view vision, characterized in that: include: A position estimation module is used to estimate the position of the target operator at each viewing angle; A position and direction description module is used to obtain the position and direction description information of the target operator relative to other operators at each viewing angle; An information matching module is used to match the target operator in multiple visual images according to the position and direction description information of the target operator relative to other operators; A position determination module is used to calculate and determine the world coordinate position of the target operator based on the two-dimensional position of the target operator in multiple visual perspectives; The early warning module is used to analyze in real time whether the target operator has entered the danger zone based on the spatial distance value between the target equipment and the target operator, and to issue an intrusion early warning signal based on the analysis results.

2. According to the multi-view vision-based substation operation safety risk management and control system of claim 1, it is characterized in that: The position estimation module estimates the position of the target operator at each viewing angle through a camera imaging model, and the camera imaging model is: Among them, z c is the distance depth coefficient, u d 、v d is the pixel coordinate point, f cx 、f cy is the focal length of the camera, d x ,d y is the physical length of a pixel, u0 and v0 are the intersection points of the camera optical axis and the image plane, and x w ,y w 、z w is the world coordinate point, R 3*3 、T 3*1 is the camera external parameter, and M is the projection matrix.

3. According to claim 2, a multi-view vision-based substation operation safety risk management and control system is characterized in that: The position direction description information includes direction information and distance information. The direction information is the angular orientation information of the target operator relative to other operators in the same viewing angle picture; the distance information is the distance information between the target operator and other operators in the same viewing angle picture.

4. According to the multi-view vision-based substation operation safety risk management and control system of claim 3, it is characterized in that: The steps of obtaining the position and direction description information of the target operator relative to other operators at each viewing angle include: a. Get the three-dimensional position P of the jth person at each perspective vj =(x vj ,y vj , 0),j∈[0,...,N v ], where V is the number of viewing angles, N v is the number of operators at the Vth perspective, v∈[1,...,V]; b. Calculate the position and direction description information of the target operator relative to other operators at each viewing angle using the following formula: Among them, W vj Each line has a three-dimensional position P relative to other workers. vj The direction and distance information of the target operator, W vj The first column is the relative three-dimensional position of other operators P vj The direction information of the target operator, W vj The first column is arranged in the order from 0 to 2π.

5. A multi-view vision-based substation operation safety risk management and control system according to claim 4, characterized in that: The matching process of the target operator under the multi-view visual screen includes: 1) According to the degree of overlap between the position and direction description of the target operator relative to other operators, the matching scores of the target operator in one picture and the operators in other perspective pictures are counted; 2) Setting a matching similarity threshold. If the matching score between the target operator in one picture and the operator in other viewing angle pictures is greater than the matching similarity threshold, the operator in the other picture corresponding to the matching score is taken as the matched target operator; 3) If the number of people who match the target operator in one screen with operators in other screens exceeds 2, the operator in other screens corresponding to the largest matching score is used as the matched target operator.

6. A multi-view vision-based substation operation safety risk management and control system according to claim 5, characterized in that: The two-dimensional position of the target operator in multiple viewing angles is (u v ,v v ),v∈[1,...,V], then the world coordinates of the target operator are (x w ,y w ,z w ), where (x w ,y w ,z w ) is obtained by the following calculation equation: ... Among them, M v is the camera projection matrix of the vth frame.

7. A multi-view vision-based substation operation safety risk management and control system according to claim 6, characterized in that: The spatial distance value is the straight-line distance d between the world coordinate position of the target operator and the spatial coordinate position of the target device. The intrusion distance threshold of the dangerous area of ​​the target device is R. When d≤R, the early warning module issues an early warning warning to the target person based on the analysis result of the target operator entering the dangerous area.

8. The multi-view vision-based substation operation safety risk management and control system according to claim 7 is characterized in that: The safety risk management and control system also includes an evacuation pointing module, which is used to send an evacuation signal to the target operator and guide the target operator to evacuate in a direction away from the dangerous area when the target operator enters the dangerous area.

9. A multi-view vision-based substation operation safety risk management and control system according to claim 8, characterized in that: The evacuation pointing module includes a safety helmet and an early warning vibrator. A plurality of early warning vibrators are arranged circumferentially on the top wearing area of ​​the safety helmet. The early warning vibrators are electrically connected to the early warning module. The early warning vibrators contact the head of the target worker. When the target worker wearing the safety helmet enters the danger zone, the early warning vibrator in the direction of the target worker pointed by the target device vibrates. The direction of the vibrating early warning vibrator is the evacuation direction of the target worker.

10. A multi-view vision-based substation operation safety risk management and control system according to claim 9, characterized in that: The implementation of the safety risk management system includes the following steps: Step 1: Estimate the position of the target operator at each viewing angle; Step 2: Calculate and obtain the position and direction description information of the target operator relative to other operators at each viewing angle; Step 3: Match the target operator in multiple visual perspectives according to the position and direction description information of the target operator relative to other operators; Step 4: Calculate and determine the world coordinate position of the target operator based on the two-dimensional position of the target operator in multiple visual perspectives; Step 5: Analyze in real time whether the target operator has entered the danger zone based on the spatial distance value between the target equipment and the target operator, and make an intrusion warning signal based on the analysis results.

Images