Auxiliary control data analysis system and method for intelligent substation

Abstract

The present invention relates to an analysis method for an auxiliary control data analysis system of an intelligent substation, comprising the following steps: S1: an information monitoring module stores all collected video data in a database by monitoring a camera; S2: a position analysis module analyzes whether an object has moved; S3: an object's movement trajectory is analyzed to determine the object entering the monitoring area; S4: a data processing module performs data processing according to different monitored objects; and S5: the processed information is sent to an information display module for display. The present invention determines the category of the monitored object by analyzing the object's movement trajectory and performs classification processing, thereby preventing suspicious persons from entering the substation and avoiding the phenomenon of damage to the substation caused by the intrusion of foreign objects.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to an auxiliary control data analysis system and method for an intelligent substation. Background Art

[0002] With the rapid development of computer technology and network communication technology, automation technology, the latest computer technology, network communication technology, video equipment technology and control technology have been widely used in the security prevention of substations to conduct online monitoring and reliable control of substations, and transmit the information remotely to the monitoring center or dispatching center.

[0003] The intelligent auxiliary control system is centered around the substation video surveillance system. To ensure safe production in the power system, existing technologies primarily monitor the installation locations of key equipment throughout the station and its surroundings around the clock. However, it is unable to analyze the type of monitored objects through monitoring data, and there are loopholes in monitoring non-human external forces such as plastic film and kites.

[0004] Therefore, people need an auxiliary control data analysis system and method for smart substations to solve the above problems. By analyzing the movement trajectory of the object, the category of the monitored object can be judged and classified. While preventing suspicious persons from entering the substation, the phenomenon of damage to the substation caused by the intrusion of foreign objects is avoided. Summary of the Invention

[0005] In view of this, an object of the present invention is to provide an auxiliary control data analysis system and method for a smart substation to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] An auxiliary control data analysis system for an intelligent substation includes an information monitoring module, a database, a position analysis module, a data processing module, and an information display module; the information monitoring module monitors video data collected by a camera; the database stores all collected video data; the position analysis module analyzes the movement trajectory of objects based on the video data and determines the objects entering the monitoring area; the data processing module performs data processing according to different monitored objects; and the information display module displays the processed information.

[0008] Furthermore, the information monitoring module includes a video acquisition unit, a video processing unit, an image fusion unit and an object judgment unit; the video acquisition unit is used to use a camera to collect video monitoring data in real time and send the data to the video processing unit; the video processing unit uses a time wheel algorithm to periodically capture video images; the image fusion unit uses a coincidence algorithm to overlap the captured video images and sends the overlapped image data to the object judgment unit; the object judgment unit determines whether the object has moved.

[0009] Furthermore, the position analysis module includes a position extraction unit, a trajectory analysis unit and an object confirmation unit; the position extraction unit obtains the motion trajectory of the moving object by establishing two-dimensional plane coordinates; the trajectory analysis unit analyzes the position pattern of the object and sends the position analysis data to the object confirmation unit; the object confirmation unit confirms the category of the monitored object based on the position pattern of the object.

[0010] Furthermore, the data processing module includes an alarm unit, an area prediction unit and an object monitoring unit; the alarm unit is used to selectively alarm the confirmed monitoring object category; the area prediction unit is used to extract the object area that last appeared in the camera to predict the position of the monitoring object, and send the position information to the patrol robot; the object monitoring unit is used to use the patrol robot to find the monitoring object and monitor it.

[0011] A method for analyzing auxiliary control data of a smart substation includes the following steps:

[0012] S1: The information monitoring module collects video data through monitoring cameras and stores all collected video data in the database;

[0013] S2: The position analysis module analyzes whether the object has moved;

[0014] S3: Analyze the object's movement trajectory and determine the object entering the monitoring area;

[0015] S4: The data processing module processes data according to different monitoring objects;

[0016] S5: Send the processed information to the information display module for display.

[0017] Furthermore, the S2 is specifically:

[0018] S201: Extract the video surveillance data of camera q, and use the time wheel algorithm to take a screenshot of the video data every t (ms) to form n image data sets L = {l s}, s = 1, 2, ..., n;

[0019] S202: overlapping the captured image data using a coincidence algorithm;

[0020] S203: By traversing the image data set L, the image data l s ={(x i ,y i )}, i = 1, 2, ..., σ and l s+1 ={(x i ,y i )}, i = 1, 2, ..., α, perform image comparison and analyze whether new pixels are generated: if new pixels exist, it means that the object a is detected to move.

[0021] Furthermore, the step S202 is specifically as follows:

[0022] A1: Establish two-dimensional plane coordinates and extract data l of image s in image dataset L s ={(x i ,y i )}, i = 1, 2, ..., σ; according to the formula: Get the texture features of image s;

[0023] Among them, p represents the pixel point (x i ,y i )’s neighborhood, d p Represents the grayscale value of the neighborhood pixel, d c Represents the grayscale value of the center pixel;

[0024] A2: By traversing the image dataset L = {l s}, s = 1, 2, ..., n, the same image texture of each pixel is fused to form a new image data l s+1 ={(x i ,y i )}, i = 1, 2,…, α.

[0025] Furthermore, the S3 is specifically:

[0026] S301: Extracting image data s+1 The new pixel point set in the object moves to form the point A = {(x i ,y i )}, i=1, 2,...,β, where At this time, each moving point corresponds to a time value, forming a time set T = {it}, i = 1, 2, ..., β (ms);

[0027] S302: Connecting the moving point data at different times to determine whether the moving object is a person or an object;

[0028] S303: Determine whether the user is an operation and maintenance personnel or other suspicious persons based on the discreteness of the mobile positions.

[0029] Furthermore, the step S302 is specifically as follows:

[0030] B1: Calculate the angle between each two line segment vectors based on the position of the moving point: Take the difference between two consecutive points to form a line segment vector set: (u x ,u y )={(x i+1 -x i ,y i+1 -y i )}, i = 1, 2, ..., β-1, then the set of angles between every two consecutive line segment vectors is obtained:

[0031] B2: Analyze the regularity of the moving position based on the vector angle to determine whether the moving object is a person or an object: If there is θ i =180°, indicating that there is a straight line area at the moving position point, then it is determined that the moving object is a person, at this time, enter step S303; on the contrary, if there is no θ i =180°, indicating that there is no straight line area at the moving position point, and the moving object is determined to be an object, in which case the process goes to step S401;

[0032] S303: Cluster the discreteness of the mobile location to determine whether the user is an operation and maintenance personnel or other suspicious persons. The steps are as follows:

[0033] C1: Collect object moving point data A={(x i ,y i )}, i = 1, 2, ..., β, obtain all the equipment locations in the substation stored in the database and form the equipment location set C k ={(m j , n j )}, j=1, 2,...k, k≤β;

[0034] C2: cluster the mobile data using the device location as the cluster center: by traversing C k ={(m j , n j )}, j=1, 2, ... k are respectively (m j , n j ) as the center and R as the radius to form a circular area: (xm j ) 2 +(yn j ) 2 =R 2, the circular areas are independent of each other and do not intersect each other; by traversing the moving point data A={(x i ,y i )}, i = 1, 2, ..., β, cluster the moving point data in the circular area to form cluster data: Z = {(x i ,y i )}, i = 1, 2, ..., ∈, where,

[0035] C3: Analyze the number of cluster data to determine whether the user is an operation and maintenance personnel or other suspicious personnel: By traversing the cluster data: Z = {(x i ,y i )}, i = 1, 2, ..., ∈;

[0036] Since suspicious persons may move near multiple devices, while operation and maintenance personnel go directly to the target faulty device; at this time, let the number of ∈>μ be ρ, where μ is the set threshold. If ρ>ω, where ω is the set threshold, then it is determined that the user belongs to other suspicious persons. At this time, enter step S403. Otherwise, if ρ<ω, then it is determined that the user belongs to operation and maintenance personnel. At this time, enter step S402.

[0037] Furthermore, in step S4: data processing is performed according to different monitoring objects. The specific steps are as follows:

[0038] S401: If the monitored object is determined to be an object, the steps are as follows:

[0039] D1: The auxiliary control system processes the alarm and sends the alarm signal to relevant personnel;

[0040] D2: Move point A to the object i ={(x i ,y i )}, i = 1, 2, ..., β for straight line fitting: Assume that the straight line equation is: Y1 = a + bx; where b is the slope and a is the intercept, according to the formula: Get the new equation of the line

[0041] D3: Collect moving point A i The horizontal coordinate set {x i}, i = 1, 2, ..., β, and the time set T = {it}, i = 1, 2, ..., β (ms) form a new data set {(it, x i )}, i = 1, 2, ..., β, perform straight line fitting as in step D2, and obtain the straight line equation

[0042] D4: According to the equation of the straight line and Predict the moving point position of the object at time (β+1)t and send the position to the inspection robot at the substation for search;

[0043] S402: If it is determined that the monitored object is an operation and maintenance personnel, the operation and maintenance personnel are monitored for any misoperation by collecting device data;

[0044] S403: If it is determined that the monitored object is another suspicious person, the auxiliary control system performs an alarm process.

[0045] Furthermore, in step S5, the information after data processing is displayed. The specific steps are as follows:

[0046] S501: The inspection robot is equipped with a camera device, which records search information in real time and connects to the auxiliary control system for display;

[0047] S502: Using a camera, monitor and display in real time the operation and maintenance personnel repairing the faulty equipment.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] The present invention uses a time wheel algorithm to perform screenshot processing on video data at regular intervals, which can efficiently utilize thread resources for batch scheduling and facilitate subsequent data overlap; by using a overlap algorithm to overlap the captured image data, further analyzing whether the object has moved based on whether new pixel points are generated, thereby improving the accuracy of the data and facilitating subsequent analysis of the category of the object; by analyzing the angle of the object's motion trajectory to determine whether there is a straight line in the object's motion trajectory, it is determined whether the monitored object is a person or an object, which is conducive to enabling the system to perform monitoring and processing according to different categories and improve the efficiency of data processing; by analyzing whether the user is a suspicious person or an operation and maintenance personnel based on the degree of discreteness of the position, it is conducive to subsequent data processing and improves the security capability of the substation; by performing multiple straight line fitting on the object's motion trajectory, the area where the object has fallen is predicted, which facilitates the search of the inspection robot in the substation and improves safety; by analyzing the object's motion trajectory to determine the category of the monitored object and perform classification processing, while preventing suspicious persons from entering the substation, the phenomenon of harm to the substation caused by the intrusion of foreign objects is avoided. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a system structure diagram of the present invention;

[0052] Figure 2 It is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0053] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0054] Please refer to Figure 1-2 The present invention provides an auxiliary control data analysis system for an intelligent substation, comprising an information monitoring module, a database, a position analysis module, a data processing module and an information display module; the information monitoring module monitors video data collected by a camera; the database stores all collected video data; the position analysis module analyzes the movement trajectory of an object based on the video data and determines the object entering the monitoring area; the data processing module performs data processing according to different monitoring objects; and the information display module displays the information after data processing.

[0055] In this embodiment, preferably, the information monitoring module includes a video acquisition unit, a video processing unit, an image fusion unit and an object judgment unit; the video acquisition unit is used to use a camera to collect video monitoring data in real time, and send the data to the video processing unit; the video processing unit uses a time wheel algorithm to regularly intercept video images; the image fusion unit uses a coincidence algorithm to overlap the intercepted video images, and sends the overlapped image data to the object judgment unit; the object judgment unit determines whether the object moves.

[0056] In this embodiment, preferably, the position analysis module includes a position extraction unit, a trajectory analysis unit and an object confirmation unit; the position extraction unit obtains the motion trajectory of the moving object by establishing two-dimensional plane coordinates; the trajectory analysis unit analyzes the position pattern of the object and sends the position analysis data to the object confirmation unit; the object confirmation unit confirms the category of the monitored object according to the position pattern of the object.

[0057] In this embodiment, preferably, the data processing module includes an alarm unit, an area prediction unit and an object monitoring unit; the alarm unit is used to selectively alarm the confirmed monitoring object category; the area prediction unit is used to extract the object area that last appeared in the camera to predict the position of the monitoring object, and send the position information to the patrol robot; the object monitoring unit is used to use the patrol robot to find the monitoring object and monitor it.

[0058] In this embodiment, a method for analyzing auxiliary control data of a smart substation is also provided, comprising the following steps:

[0059] S1: m cameras are set up in the substation to collect video surveillance data V in real time. m ={v q}, q = 1, 2, ..., m, and store all collected video data in the database;

[0060] S2: The position analysis module analyzes whether the object has moved;

[0061] Preferably, in this embodiment, step S2 is specifically as follows:

[0062] S201: Extract the video surveillance data of camera q, and use the time wheel algorithm to take a screenshot of the video data every t (ms) to form n image data sets L = {l s}, s = 1, 2, ..., n;

[0063] S202: overlapping the captured image data using a coincidence algorithm;

[0064] S203: By traversing the image data set L, the image data l s ={(x i ,y i )}, i = 1, 2, ..., σ and l s+1 ={(x i ,y i )}, i = 1, 2, ..., α, perform image comparison and analyze whether new pixels are generated: if new pixels exist, it means that the object a is detected to move.

[0065] S3: Analyze the object's movement trajectory and determine the object entering the monitoring area;

[0066] In this embodiment, preferably, S3 is specifically:

[0067] S301: Extracting image data s+1 The new pixel point set in the object moves to form the point A = {(x i ,y i )}, i=1, 2,...,β, where At this time, each moving point corresponds to a time value, forming a time set T = {it}, i = 1, 2, ..., β (ms);

[0068] S302: Connecting the moving point data at different times to determine whether the moving object is a person or an object;

[0069] S303: Determine whether the user is an operation and maintenance personnel or other suspicious persons based on the discreteness of the mobile positions.

[0070] S4: The data processing module processes data according to different monitoring objects;

[0071] In this embodiment, preferably, S4 is specifically:

[0072] S401: If the monitored object is determined to be an object, the steps are as follows:

[0073] D1: The auxiliary control system processes the alarm and sends the alarm signal to relevant personnel;

[0074] D2: Move point A to the object i ={(x i ,y i )}, i = 1, 2, ..., β for straight line fitting: Assume that the straight line equation is: Y1 = a + bx; where b is the slope and a is the intercept, according to the formula: Get the new equation of the line

[0075] D3: Collect moving point A i The horizontal coordinate set {x i}, i = 1, 2, ..., β, and the time set T = {it}, i = 1, 2, ..., β (ms) form a new data set {(it, x i )}, i = 1, 2, ..., β, perform straight line fitting as in step D2, and obtain the straight line equation

[0076] D4: According to the equation of the straight line and Predict the moving point position of the object at time (β+1)t and send the position to the inspection robot at the substation for search;

[0077] S402: If it is determined that the monitored object is an operation and maintenance personnel, the operation and maintenance personnel are monitored for any misoperation by collecting device data;

[0078] S403: If it is determined that the monitored object is another suspicious person, the auxiliary control system performs an alarm process.

[0079] S5: Send the processed information to the information display module for display.

[0080] Preferably, in this embodiment, step S202 is specifically:

[0081] A1: Establish two-dimensional plane coordinates and extract data l of image s in image dataset L s ={(x i ,y i )}, i = 1, 2, ..., σ; according to the formula: Get the texture features of image s;

[0082] Among them, p represents the pixel point (x i ,y i )’s neighborhood, d p Represents the grayscale value of the neighborhood pixel, d c Represents the grayscale value of the center pixel;

[0083] A2: By traversing the image dataset L = {l s}, s = 1, 2, ..., n, the same image texture of each pixel is fused to form a new image data l s+1 ={(x i ,y i )}, i = 1, 2,…, α.

[0084] In this embodiment, preferably, S302 is specifically:

[0085] B1: Calculate the angle between each two line segment vectors based on the position of the moving point: Take the difference between two consecutive points to form a line segment vector set: (u x ,u y )={(x i+1 -x i ,y i+1 -y i )}, i = 1, 2, ..., β-1, then the set of angles between every two consecutive line segment vectors is obtained:

[0086] B2: Analyze the regularity of the moving position based on the vector angle to determine whether the moving object is a person or an object: If there is θ i =180°, indicating that there is a straight line area at the moving position point, then it is determined that the moving object is a person, at this time, enter step S303; on the contrary, if there is no θ i =180°, indicating that there is no straight line area at the moving position point, and the moving object is determined to be an object, and the process goes to step S4.

[0087] The above description is only a preferred embodiment of the present invention. All equivalent changes and modifications made according to the scope of the patent application of the present invention should fall within the scope of the present invention.

Claims

1. A method for analyzing auxiliary control data of a smart substation, characterized in that: The following steps are involved: S1: The information monitoring module collects video data through monitoring cameras and stores all collected video data in the database; S2: The position analysis module analyzes whether the object has moved; S3: Analyze the movement trajectory of the object and determine the object entering the monitoring area; S4: The data processing module processes data according to different monitoring objects; S5: Send the processed information to the information display module for display; The S3 is specifically: S301: Extracting image data s+1 The new pixel point set in the object moves to form the point A = {(x i ,y i )},i=1,2,…,β, where At this time, each moving point corresponds to a time value, forming a time set T = {it}, i = 1, 2, …, β / ms; S302: Connecting the moving point data at different times to determine whether the moving object is a person or an object; S303: Determine whether the user is an operation and maintenance personnel or other suspicious person based on the discreteness of the mobile location; C1: Collect object moving point data A={(x i ,y i )}, i=1,2,…,β, obtain all the equipment locations in the substation stored in the database and form the equipment location set C k ={(m j ,n j )},j=1,2,…k,k≤β; C2: cluster the mobile data using the device location as the cluster center: by traversing C k ={(m j ,n j )},j=1,2,…k are respectively (m j ,n j ) as the center and R as the radius to form a circular area: (xm j ) 2 +(yn j ) 2 =R 2 ; By traversing the moving point data A={(x i ,y i )}, i=1,2,…,β, cluster the moving point data in the circular area to form cluster data: Z={(x i ,y i )},i=1,2,…,∈, where, C3: Analyze the number of cluster data to determine whether the user is an operation and maintenance personnel or other suspicious personnel: By traversing the cluster data: Z = {(x i ,y i )}, i = 1, 2, ..., ∈, let the number of ∈>μ be ρ, where μ is the set threshold, if ρ>ω, where ω is the set threshold, then it is determined that the user belongs to other suspicious persons, and in this case, the process goes to step S403; Said S4 is specifically: S401: If the monitored object is determined to be an object, the steps are as follows: D1: The auxiliary control system processes the alarm and sends the alarm signal to relevant personnel; D2: Move point A to the object i ={(x i ,y i )}, i = 1, 2, ..., β for straight line fitting: Assume that the straight line equation is: Y1 = a + bx; where b is the slope and a is the intercept, according to the formula: Get the new equation of the line D3: Collect moving point A i The horizontal coordinate set {x i },i=1,2,…,β,with the time set T={it},i=1,2,…,β / ms Form a new data set {(it,x i )}, i=1,2,…,β, perform straight line fitting as in step D2 to obtain the straight line equation D4: According to the equation of the straight line and Predict the moving point position of the object at time (β+1)t and send the position to the inspection robot at the substation for search; S402: If it is determined that the monitored object is an operation and maintenance personnel, the operation and maintenance personnel are monitored for any misoperation by collecting device data; S403: If it is determined that the monitored object is another suspicious person, the auxiliary control system performs an alarm process.

2. The auxiliary control data analysis method for a smart substation according to claim 1, characterized in that: The S2 is specifically: S201: Extract the video surveillance data of camera q, and use the time wheel algorithm to take a screenshot of the video data every t / ms to form n image data sets L={l s },s=1,2,…,n; S202: overlapping the captured image data using a coincidence algorithm; S203: By traversing the image data set L, the image data l s ={(x i ,y i )},i=1,2,…,σandl s+1 ={(x i ,y i )}, i=1,2,…,α, perform image comparison and analyze whether new pixels are generated: if new pixels exist, it means that the object a is detected to have moved.

3. The auxiliary control data analysis method for a smart substation according to claim 2, characterized in that: The S202 is specifically as follows: A1: Establish two-dimensional plane coordinates and extract data l of image s in image dataset L s ={(x i ,y i )},i=1,2,…,σ;According to the formula: Get the texture features of image s; Among them, p represents the pixel point (x i ,y i )’s neighborhood, d p Represents the grayscale value of the neighborhood pixel, d c Represents the grayscale value of the center pixel; A2: By traversing the image dataset L = {l s }, s=1,2,…,n, perform pixel fusion on the image texture with the same pixel point to form a new image data l s+1 ={(x i ,y i )},i=1,2,…,α。 4. The auxiliary control data analysis method for a smart substation according to claim 1, characterized in that: The S302 is specifically as follows: B1: Calculate the angle between each two line segment vectors based on the position of the moving point: Take the difference between two consecutive points to form a line segment vector set: (u x ,u y )={(x i+1 -x i ,y i+1 -y i )}, i=1,2,…,β-1, then the set of angles between every two consecutive line segment vectors is obtained: B2: Analyze the regularity of the moving position based on the vector angle to determine whether the moving object is a person or an object: If there is θ i =180°, indicating that there is a straight line area at the moving position point, then it is determined that the moving object is a person, at this time, enter step S303; on the contrary, if there is no θ i =180°, indicating that there is no straight line area at the moving position point, and the moving object is determined to be an object, and the process goes to step S4.

Images