Substation surveillance camera layout method based on regional risk level classification
Through the method based on regional risk level division, the camera layout of substations is optimized, and the problems of incomplete monitoring coverage and blind spots are solved, efficient and economical safety monitoring system design is achieved, and the safety risks of substations are reduced.
Patent Information
- Application Number
- CN202211624005.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-12-16
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2042-12-16
AI Technical Summary
The existing technology is difficult to reasonably arrange cameras in substations, resulting in incomplete monitoring coverage and blind spots, and cannot effectively reduce the safety risks of personal and equipment.
Using a method based on regional risk level division, the camera field angle, visibility analysis, and construct a two-dimensional monitoring status table and black and white binary plan, combined with a greedy search algorithm, the camera layout is optimized to maximize the monitoring coverage cost function. Ensure intensive monitoring of high-risk areas.
It has achieved efficiently and economically meeting the monitoring needs of different substations in different areas on limited camera terminals, reducing monitoring blind spots, and reducing safety risks.
Smart Images

Figure CN115996321B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of electrical equipment and electrical engineering technology, and in particular to a method for arranging substation monitoring cameras based on regional risk level division. Background Art
[0002] Video surveillance technology has broad application prospects in substation security monitoring and management. With the rapid development of artificial intelligence technology, "unmanned operation" has become an inevitable trend in the development of substation automation and the ultimate goal of smart substation development. At present, substations with voltage levels of 220kV and below have basically achieved unmanned operation, and 500kV substations are entering a stage of less-manned operation, in which the optimized layout of camera terminals plays an important role. The optimized layout of camera terminals determines the coverage rate of video surveillance systems for safety monitoring of substation personnel and equipment, which is conducive to improving the level of real-time monitoring of substation personnel and equipment, thereby greatly reducing the risk of personal injury and power outages due to equipment failure.
[0003] The essence of the camera layout optimization problem is to find the appropriate position and azimuth so that the monitoring coverage area of the camera array is maximized. It is worth noting that the optimization model of this problem may vary according to the model parameters of the camera (for example, the type of camera, potential position, azimuth, etc.). Therefore, its optimization goal mainly reflects specific security monitoring needs, such as maximizing the monitoring coverage rate, or the overlapping area of multi-view monitoring coverage. In addition, since the various buildings and equipment in the substation are of different sizes and heights, when simplifying the stereogram of the substation into a plane diagram for the deployment of monitoring cameras, it is necessary to consider the importance of the area (or the degree of risk) and the occlusion between the building equipment. When the equipment in a certain area is more important or the mutual occlusion problem of the equipment near the area is more prominent, and the security risk is greater, the higher the monitoring clarity level required in the area, the denser the camera layout. Summary of the invention
[0004] The purpose of the present invention is to provide a substation monitoring camera layout method based on regional risk level division, which can reasonably layout cameras, ensure monitoring coverage, and reduce monitoring blind spots.
[0005] In order to achieve the above object, the technical solution adopted by the present invention is:
[0006] A method for substation monitoring camera layout based on regional risk level classification includes the following steps:
[0007] Step S1: Calculate the field of view of the surveillance camera to be used, and then obtain a surveillance coverage area model of an array composed of a plurality of surveillance cameras;
[0008] Step S2: Based on the field of view of each monitoring camera, the visibility analysis of each point within the monitoring range of the substation is performed using the Bresenham straight line algorithm;
[0009] Step S3: constructing a two-dimensional monitoring state table according to the types, potential positions and azimuth sets of the different monitoring cameras;
[0010] Step S4: constructing a black-and-white binary plane map of the substation according to a set of empty spaces and obstacles within the monitoring range of the substation;
[0011] Step S5: Calculate the shortest Euclidean distance between each point on the black-and-white binary plane map of the substation and the obstacle, so as to divide the regional risk level;
[0012] Step S6: Define the camera monitoring coverage cost function considering the regional risk level, take maximizing the cost function as the optimization goal, use the greedy search algorithm to solve the optimization model, and obtain the substation monitoring camera layout plan.
[0013] In step S1, the field of view F of the surveillance camera is: Where r is the effective monitoring range of the surveillance camera, w I is the width of the image sensor in the surveillance camera, and f is the focal length of the surveillance camera.
[0014] The calculation formula of the effective monitoring range r of the surveillance camera is: Where: h I is the height h of the image sensor in the surveillance camera I 、h O is the estimated height of the object within the monitoring range of the surveillance camera, and p is the minimum acceptable percentage of the height of the object.
[0015] In step S2, visibility analysis is performed on each point within the monitoring coverage area. If the line connecting the point and the starting point of the monitoring camera is within the field of view and does not intersect with any obstacles, it is marked as "visible", otherwise it is marked as "invisible", thereby obtaining the visibility analysis result.
[0016] In step S3, there are K types of surveillance cameras, each surveillance camera has O azimuth angles, and there are L camera installation positions. Then, the total number of states of the surveillance camera at the lth camera installation position NS=K×O; the two-dimensional surveillance state table CS is defined as: in, is the monitoring state quantity of the monitoring camera at the lth camera installation position in the nth state; then the monitoring state quantity of the monitoring camera at the lth camera installation position in the nth state with model k and azimuth o is Among them, c (i,j) is a binary variable. For the surveillance camera at the lth camera installation position, whose model is k and azimuth is o, if the point (i, j) in the surveillance area is "visible", then c (i,j) =1, otherwise c (i,j) =0.
[0017] In step S4, the three-dimensional substation monitoring coverage area is simplified into a two-dimensional top view, which includes a set of empty sites A = a1 ... a H And the obstacle set B = b1…b U , H and U are the number of elements in the empty space set and the obstacle set respectively, then according to the classification of the empty space set A and the obstacle set B in the figure, the black and white binary plane graph is established.
[0018] In step S5, the calculation of each grid point A on the black and white binary plane image is performed. x,y and the obstacle set B = b1…b U and select the shortest distance.
[0019] By using multiple boundary values to divide m Euclidean distance intervals, the regional risk levels are divided according to the Euclidean distance interval in which the shortest calculated Euclidean distance lies to obtain various risk areas, where m is an integer greater than or equal to 3.
[0020] In step S6, a camera monitoring coverage cost function considering the regional risk level is defined. Among them, C q is the monitoring coverage cost of the qth risk area, w q The weight coefficient of the qth risk area; then the optimization objective is
[0021] C q =max[S l ], where S l is the row vector row l in the two-dimensional monitoring state table CS
[0022] Due to the application of the above-mentioned technical scheme, the present invention has the following advantages compared with the prior art: the present invention provides a substation monitoring camera layout method based on regional risk level division. This method is extremely important for building an efficient and economical substation safety monitoring system layout design, and can meet the monitoring needs of different areas of the substation. On the basis of a limited number of terminal cameras, it can ensure the monitoring coverage of key high-risk areas and realize the differentiated optimization configuration of limited monitoring resources, thereby effectively reducing monitoring blind spots and reducing the risks caused by insufficient monitoring. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Attached Figure 1 The present invention is a flow chart of a method for arranging substation monitoring cameras based on regional risk level division.
[0024] Attached Figure 2 This is a framework diagram of the substation monitoring camera layout method based on regional risk level division of the present invention.
[0025] Attached Figure 3 It is a black and white binary plane image established in the embodiment of the present invention.
[0026] like Figure 4 Shown is a color block diagram representing the risk level division in an implementation case of the present invention. DETAILED DESCRIPTION
[0027] The present invention will be further described below in conjunction with the embodiments shown in the accompanying drawings.
[0028] Embodiment 1: As shown in the attached Figure 1 and attached Figure 2 As shown, a method for substation monitoring camera layout based on regional risk level division includes the following steps:
[0029] Step S1: Calculate the field of view of the surveillance camera to be used, and then obtain a surveillance coverage area model of an array composed of a plurality of surveillance cameras.
[0030] The field of view F of the surveillance camera is:
[0031]
[0032] In the formula, w I is the width of the image sensor in the surveillance camera, f is the focal length of the surveillance camera, and r is the effective monitoring range of the surveillance camera, which can be determined by the height h of the image sensor in the surveillance camera. I , the estimated height h of the object within the monitoring range of the surveillance camera O And the minimum acceptable percentage p of the height of the photographed object is calculated. The specific calculation formula is:
[0033]
[0034] Substituting it into formula (2), we finally get the expressions of the camera field of view angle F and focal length f:
[0035]
[0036] Step S2: Based on the field of view of each monitoring camera, the Bresenham line algorithm is used to perform visibility analysis on each point within the monitoring range of the substation.
[0037] This step is specifically as follows: relying on the field of view F of each surveillance camera obtained in step S1, the Bresenham straight line algorithm is used to perform visibility analysis on each point within the surveillance coverage. If the line connecting the point and the starting point of the surveillance camera is within the field of view and does not intersect with any obstacles, it is marked as "visible", otherwise it is marked as "invisible", thereby obtaining the visibility analysis result.
[0038] Step S3: construct a two-dimensional monitoring status table according to the types, potential positions and azimuth sets of different monitoring cameras.
[0039] The specific step is: there are K types of surveillance cameras, each surveillance camera has O azimuth angles, and there are L camera installation positions, then the total number of surveillance camera states at the l-th camera installation position NS = K × O. According to the types K, potential positions L and azimuth angle sets O of different surveillance cameras, a two-dimensional surveillance state table CS is constructed, where the surveillance state set S at each position l = {s1, s2, …, s NS The monitoring state quantity of the surveillance camera at the lth camera installation position in the nth state with model k and azimuth o is:
[0040]
[0041] Among them, c (i,j) is a binary variable. For the surveillance camera at the lth camera installation position, whose model is k and azimuth is o, if the point (i, j) in the surveillance area is "visible", then c (i,j) =1, otherwise c (i,j) = 0. All the camera installation positions l (l = 1, 2, ..., L) are recorded as L decision stages, and the monitoring state set of each stage is recorded as S l (l=1,2,…,L), contains NS state quantities (s1,s2,…,s NS ), then the two-dimensional monitoring state table CS is defined as:
[0042]
[0043] in, is the monitoring state quantity of the monitoring camera at the lth camera installation position in the nsth state, obtained using formula (4).
[0044] Step S4: construct a black-and-white binary plan view of the substation based on the set of empty spaces and obstacles within the monitoring range of the substation.
[0045] The specific step is: simplify the three-dimensional substation monitoring coverage area into a two-dimensional top view, which contains the empty site set A = a1…a H And the obstacle set B = b1…b U , H and U are the number of elements in the empty space set and obstacle set respectively. According to the classification of the empty space set A and the obstacle set B in the figure, a black-and-white binary plane graph G is established.
[0046] Step S5: Calculate the shortest Euclidean distance between each point on the black-and-white binary plane map of the substation and the obstacle, so as to divide the regional risk level.
[0047] The specific step is: calculate each grid point A on the black and white binary plane graph G x,y and the obstacle set B = b1…b U The Euclidean distances of the points are calculated, and the shortest distance is selected as d. The Euclidean distance intervals are divided into m using multiple boundary values (distance thresholds). Then, the regional risk levels are divided according to the Euclidean distance interval in which the shortest Euclidean distance is calculated to obtain each risk area, where m is an integer greater than or equal to 3. For example, the shortest distance d obtained in step S5 is x,y Assign the corresponding risk level R(A x ,d):
[0048]
[0049] Where: m = 3, which means that three levels of risk areas are divided, v1, v2, v3 and t1, t2, t3 represent the risk level (v1: high risk; v2: medium risk; v3: low risk) and the predefined distance threshold range (t1 <t2<t3)
[0050] Step S6: Define the camera monitoring coverage cost function considering the regional risk level, take maximizing the cost function as the optimization goal, use the greedy search algorithm to solve the optimization model, and obtain the substation monitoring camera layout plan.
[0051] This step is specifically as follows: define a camera monitoring coverage cost function that takes into account the regional risk level, and its value is the weighted sum of the regional coverage costs under each risk level:
[0052]
[0053] Among them, C q The monitoring coverage cost for the qth risk area is:
[0054] C q =max[S l ],q=1,2,…,m (8)
[0055] S l is the row vector row of the lth row in the two-dimensional monitoring state table CS w q The weight coefficient of the qth risk area.
[0056] The optimization goal is to maximize the monitoring coverage cost function at all locations:
[0057]
[0058] The greedy search algorithm is used to search for the optimal camera model and azimuth parameters at each location l, and finally the optimal substation monitoring system layout plan is obtained.
[0059] A specific example is as follows:
[0060] Taking a specific 220kV substation in a certain area as an example, the specific steps are as follows:
[0061] Step S1: Three types of surveillance cameras are selected, and their model parameters are shown in Table 1. According to equations (1)-(3), their field of view angles are calculated as follows: F1=110.9°, F2=76.4°, and F3=36.5°.
[0062] Table 1 Camera model configuration parameters
[0063]
[0064] Step S2: Based on the field of view F of each camera obtained in step S1, the Bresenham straight line algorithm is used to perform visibility analysis on each point within the monitoring coverage. The specific method is: if the line connecting the point and the camera starting point is within the field of view and does not intersect with obstacles, it is marked as "visible" (binary image is assigned a value of 1), otherwise it is marked as "invisible" (binary image is assigned a value of 0).
[0065] Step S3: Construct a monitoring state table CS based on the set of different camera types K = 3 (k = 1, 2, 3), potential positions L = 10 (l = 1, 2, 3, ..., 10) and azimuths O = 5 (o = 30). The monitoring state set S = {s1, s2, ..., s NS}, NS is the total number of state quantities, NS=K×O=3×5=15.
[0066]
[0067] In the formula, Here M = 400, N = 300 is the number of pixels in the two-dimensional monitoring plane; c (i,j) is a binary (0-1) variable. For a camera at position l, its model is k and its azimuth is o. If a point (i, j) on the monitoring area is "visible", then c (i,j) =1, otherwise c (i,j) =0.
[0068] Step S4: Simplify the three-dimensional substation monitoring coverage area into a two-dimensional top-down plane map, which includes a set of empty sites and a set of obstacles, and establish a black-and-white binary plane map G, such as Figure 3 shown.
[0069] Step S5: Calculate each grid point A on the plane graph G x,y and the obstacle set B = b1…b U The Euclidean distance is , U is the number of obstacles (including buildings, substation equipment, etc.) within the monitoring range, U = 8. The shortest distance is selected and recorded as d.
[0070] Step S6: Based on the shortest distance d obtained in step S5, for each point A on the graph G x,y Assign corresponding risk levels. In this case, the monitoring area is divided into three risk levels, namely:
[0071]
[0072] Where: v1, v2, v3 and t1, t2, t3 represent the risk level (v1: high risk; v2: medium risk; v3: low risk) and the predefined distance threshold range respectively. Among them, t1 = 10, t2 = 20, t3 = 50. Different risk levels are represented by different color blocks in the plane map (v1: red; v2: yellow; v3: green), such as Figure 4 shown.
[0073] Step S7: Define a camera monitoring coverage cost function that takes into account the regional risk level, whose value is the weighted sum of the regional coverage costs under each risk level:
[0074]
[0075] in,
[0076] C q =max[S l ]
[0077] Where: C1, C2, C3 are the regional monitoring coverage costs, corresponding to risk levels v1, v2, and v3 respectively; w1, w2, and w3 are the weight coefficients of different regional risk levels. w1 = 0.77, w2 = 0.23, w3 = 0.
[0078] The optimization goal is to maximize the monitoring coverage cost function at all positions (L = 10):
[0079]
[0080] Use the greedy search algorithm to search for the optimal camera model k at each position l * and azimuth parameter o * , and finally the optimal substation monitoring system layout plan is obtained.
[0081] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable people familiar with the technology to understand the content of the present invention and implement it accordingly, and they cannot be used to limit the protection scope of the present invention. Any equivalent changes or modifications made according to the spirit of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for substation surveillance camera layout based on regional risk level classification, Features: It includes the following steps: Step S1: Calculate the field of view of the surveillance camera to be used, and then obtain a surveillance coverage area model of an array composed of a plurality of surveillance cameras; Step S2: Based on the field of view of each monitoring camera, the visibility analysis of each point within the monitoring range of the substation is performed using the Bresenham straight line algorithm; Step S3: constructing a two-dimensional monitoring state table according to the types, potential positions and azimuth sets of the different monitoring cameras; Step S4: constructing a black-and-white binary plane map of the substation according to a set of empty spaces and obstacles within the monitoring range of the substation; Step S5: Calculate the shortest Euclidean distance between each point on the black-and-white binary plane map of the substation and the obstacle, so as to divide the regional risk level; Step S6: define a camera monitoring coverage cost function that takes into account the regional risk level, take maximizing the cost function as the optimization goal, use a greedy search algorithm to solve the optimization model, and obtain a substation monitoring camera layout plan; In step S4, the three-dimensional substation monitoring coverage area is simplified into a two-dimensional top view, which includes a set of empty sites A=a 1 …a H And the obstacle set B = b 1 …b U , H and U are the number of elements in the empty space set and the obstacle set respectively, then according to the classification of the empty space set A and the obstacle set B in the figure, the black and white binary plane graph is established; In step S5, the calculation of each grid point A on the black and white binary plane image is x,y and obstacle set B = b 1 …b U The Euclidean distances of the two regions are calculated, and the shortest distance is selected. The m Euclidean distance intervals are divided by multiple boundary values. Then, the regional risk levels are divided according to the Euclidean distance interval in which the shortest Euclidean distance is calculated, and each risk region is obtained. m is an integer greater than or equal to 3. In step S6, a camera monitoring coverage cost function considering the regional risk level is defined. Among them, C q is the monitoring coverage cost of the qth risk area, w q The weight coefficient of the qth risk area; then the optimization objective is 2. According to the method for substation monitoring camera layout based on regional risk level classification according to claim 1, Features: In step S1, the field of view F of the surveillance camera is: Where r is the effective monitoring range of the surveillance camera, w I is the width of the image sensor in the surveillance camera, and f is the focal length of the surveillance camera.
3. According to claim 2, the substation monitoring camera layout method based on regional risk level classification, Features: The calculation formula of the effective monitoring range r of the surveillance camera is: Where: h I is the height of the image sensor in the surveillance camera, h O is the estimated height of the object within the monitoring range of the surveillance camera, and p is the minimum acceptable percentage of the height of the object.
4. According to claim 1, the substation monitoring camera layout method based on regional risk level classification, Features: In step S2, visibility analysis is performed on each point within the monitoring coverage area. If the line connecting the point and the starting point of the monitoring camera is within the field of view and does not intersect with any obstacles, it is marked as "visible", otherwise it is marked as "invisible", thereby obtaining a visibility analysis result.
5. According to claim 2, the substation monitoring camera layout method based on regional risk level classification, Features: In step S3, there are K types of surveillance cameras, each surveillance camera has O azimuth angles, and there are L camera installation positions. Then, the total number of states of the surveillance camera at the lth camera installation position NS=K×O; the two-dimensional surveillance state table CS is defined as: in, is the monitoring state quantity of the monitoring camera at the lth camera installation position in the nth state; then The monitoring state quantity of the monitoring camera at the lth camera installation position in the nth state with model k and azimuth o Among them, c (i,j) is a binary variable. For the surveillance camera at the lth camera installation position, whose model is k and azimuth is o, if the point (i, j) in the surveillance area is "visible", then c (i,j) =1, otherwise c (i,j) =0.
6. According to claim 1, the substation monitoring camera layout method based on regional risk level classification, Features: C q =max[S l ], where S l is the row vector row l in the two-dimensional monitoring state table CS
