WO2022027877A1 - 一种真空开关触头的运动速度检测方法及装置 - Google Patents

一种真空开关触头的运动速度检测方法及装置 Download PDF

Info

Publication number
WO2022027877A1
WO2022027877A1 PCT/CN2020/132471 CN2020132471W WO2022027877A1 WO 2022027877 A1 WO2022027877 A1 WO 2022027877A1 CN 2020132471 W CN2020132471 W CN 2020132471W WO 2022027877 A1 WO2022027877 A1 WO 2022027877A1
Authority
WO
WIPO (PCT)
Prior art keywords
arc
image
vacuum switch
moving
contact
Prior art date
Application number
PCT/CN2020/132471
Other languages
English (en)
French (fr)
Inventor
薛从军
李小钊
马占彪
刘世柏
赵芳帅
齐大翠
亓春伟
李锟
张杨
王宇浩
刘心悦
白丽娜
孙宇
王茜
董华军
马丽婷
唐朝端
Original Assignee
天津平高智能电气有限公司
平高集团有限公司
国家电网公司
大连交通大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 天津平高智能电气有限公司, 平高集团有限公司, 国家电网公司, 大连交通大学 filed Critical 天津平高智能电气有限公司
Publication of WO2022027877A1 publication Critical patent/WO2022027877A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/215—Motion-based segmentation
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01P—MEASURING LINEAR OR ANGULAR SPEED, ACCELERATION, DECELERATION, OR SHOCK; INDICATING PRESENCE, ABSENCE, OR DIRECTION, OF MOVEMENT
    • G01P3/00—Measuring linear or angular speed; Measuring differences of linear or angular speeds
    • G01P3/36—Devices characterised by the use of optical means, e.g. using infrared, visible, or ultraviolet light
    • G01P3/38—Devices characterised by the use of optical means, e.g. using infrared, visible, or ultraviolet light using photographic means
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/327—Testing of circuit interrupters, switches or circuit-breakers
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01H—ELECTRIC SWITCHES; RELAYS; SELECTORS; EMERGENCY PROTECTIVE DEVICES
    • H01H33/00—High-tension or heavy-current switches with arc-extinguishing or arc-preventing means
    • H01H33/60—Switches wherein the means for extinguishing or preventing the arc do not include separate means for obtaining or increasing flow of arc-extinguishing fluid
    • H01H33/66—Vacuum switches
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20024—Filtering details
    • G06T2207/20032—Median filtering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20172—Image enhancement details
    • G06T2207/20192—Edge enhancement; Edge preservation

Definitions

  • the application belongs to the technical field of vacuum switches, and in particular relates to a method and a device for detecting the movement speed of a contact of a vacuum switch.
  • the vacuum switch As the control and protection equipment of the power system, the vacuum switch has the advantages of long service life, no pollution, convenient maintenance, compact structure and light weight.
  • the vacuum switch mainly includes two parts: the vacuum interrupter and the operating mechanism.
  • the vacuum interrupter When the power system fails, how to realize the reliable operation of the operating mechanism is the key to successfully breaking the current in the interrupter.
  • the opening speed is one of the important factors affecting the breaking capacity of the vacuum switch. If the opening speed is too fast, the amplitude of the opening spring will increase, and then the bellows, shielding cover and other structures will be damaged; if the opening speed is too slow and the arcing time is too long, it will lead to serious contact burns. .
  • the detection methods of opening speed mainly include electromagnetic signal sensor detection method and photoelectric sensor detection method.
  • the detection method using the electromagnetic signal sensor is only suitable for the measurement requirements of the slow-speed operation of the operating mechanism, while the detection method of the photoelectric sensor is easily disturbed by external signals, and the sensor and the operating mechanism appear relative motion.
  • the related art realizes the method of calculating the contact position and moving speed of the vacuum switch.
  • This method obtains continuous images of the vacuum switch contact and arc movement, and adopts image analysis technology to determine the position of the upper and lower edges of the arc in each frame of the image (that is, the vacuum The position of the static contact of the switch and the position of the moving contact in the current frame), and then the moving speed of the moving contact can be calculated according to the position information and the frame difference time.
  • the method of grid division and threshold setting is directly used to analyze and obtain the upper and lower edge positions of the arc in each frame of image.
  • the embodiments of the present application provide a method and a device for detecting the movement speed of a vacuum switch contact, so as to solve the problem of a large amount of calculation processing caused by the prior art.
  • the technical solutions of the embodiments of the present application include:
  • the embodiment of the present application provides a method for detecting the movement speed of a vacuum switch contact, comprising the following steps:
  • the Laplacian edge detection operator is used to perform edge enhancement processing on the acquired arc images
  • the actual displacement difference of the moving contacts in the two arc images is determined, and the moving speed of the vacuum switch contacts is determined in combination with the time interval between the two arc images.
  • the application first uses the Laplace edge detection operator to perform edge enhancement processing on the obtained arc images of each frame, which strengthens the
  • the contrast between the lower edge and the background image of the arc image avoids the processing and analysis of the whole frame of the arc image, reduces the amount of calculation, and facilitates the rapid and accurate extraction of the upper and lower edges in the subsequent processing, making the edge detection more continuous and accurate. and clear, which effectively avoids the appearance of false edges, which makes the calculation of the movement speed of the vacuum switch contacts more accurate, facilitates the analysis of the influence of the movement characteristics of the operating mechanism on the arc shape, and provides technical support for the study of arc control theory.
  • the functional model of the Laplacian edge detection operator is in the form of a 5 ⁇ 5 matrix, and the used functional model is:
  • the Laplacian edge detection operator in the form of a 5 ⁇ 5 matrix is used to improve the edge detection accuracy.
  • the method further includes the step of performing noise reduction processing on the acquired arc image by using a median filtering algorithm.
  • the median filter algorithm can avoid the influence of image noise on the location of the upper and lower edges of the arc.
  • the actual displacement difference of the moving contacts in the two frames of arc images is:
  • s is the actual displacement difference of the moving contact in the two arc images
  • D is the diameter of the moving contact
  • N′ is the number of pixels of the moving contact
  • N is the image displacement difference of the moving contact in the two arc images.
  • An embodiment of the present application further provides a device for detecting the movement speed of a vacuum switch contact, including a memory and a processor, where the processor is configured to execute an instruction stored in the memory to realize the above-described movement speed detection of the vacuum switch contact method and achieve the same effect as this method.
  • FIG. 1 is a flowchart of an embodiment of a method for detecting the movement speed of a vacuum switch contact according to an embodiment of the present application
  • FIG. 2-1 is an arc image collected by the CMOS high-speed camera according to the embodiment of the present application.
  • Fig. 2-2 is the arc image after the arc image is imported into the LabVIEW software according to the embodiment of the present application;
  • 2-3 are arc images after median filtering processing according to an embodiment of the present application.
  • 2-4 are arc images after edge enhancement processing using Laplacian edge detection operator according to an embodiment of the present application.
  • Figure 3-1 is the arc image after edge enhancement processing using Robert edge detection operator
  • Figure 3-2 is the arc image after edge enhancement processing using Sobel edge detection operator
  • Figure 3-3 is the arc image after edge enhancement processing using Prewitt edge detection operator
  • Figure 3-4 is the arc image after edge enhancement processing using Laplacian edge detection operator
  • FIG. 4 is a structural diagram of a moving speed detection device for a vacuum switch contact according to an embodiment of the present application.
  • the embodiment of the method for detecting the movement speed of the contact of a vacuum switch is based on the LabVIEW virtual instrument development platform, using a high-speed camera to collect the moving image of the contact in the entire cycle, and using the image processing module of LabVIEW to detect the moving and static contacts
  • the edge is positioned with high precision, and the movement speed of the vacuum switch contact is calculated, which provides a technical basis for the dynamic research of the vacuum switch arc.
  • LabVIEW uses the graphical programming G language, and realizes programming through data flow, which can not only achieve real-time monitoring and direct editing of data flow, but also share data with other software.
  • step 101 a CMOS high-speed camera is used to acquire an arc image of the vacuum switch during the opening process.
  • CMOS high-speed camera is used to collect the arc image during the vacuum switch opening process.
  • the CMOS analog signal is converted into an image digital signal by A/D, and the acquired arc image ( N frames in total) are transmitted to the computer, as shown in Figure 2-2, the computer is installed with LabVIEW image processing software.
  • Step 102 using LabVIEW image processing software to preprocess the arc image.
  • the preprocessing process sequentially performs gray-scale and binarization processing on the opening arc image, then uses a filtering algorithm to perform noise reduction processing, and uses a Laplacian edge detection operator to perform edge detection. Strengthen processing.
  • the filtering algorithm is a median filtering algorithm.
  • the median filtering algorithm can effectively smooth the impulse noise while protecting the sharp edges of the image.
  • the processed image is shown in Figure 2-3.
  • the function of median filter can be expressed as:
  • y is the median of the sequence x 1 , x 2 , x 3 ,...,x n .
  • the median filter is a neighborhood with odd pixels, and the pixel value of the center pixel of the neighborhood is replaced by the median.
  • the Laplacian edge detection operator is used to detect the edge of the arc image, and the edge of the arc image is enhanced.
  • the processed image is shown in Figure 2-4.
  • the edge enhancement can be achieved by using the Laplacian operator for edge detection.
  • the Laplacian edge detection operator is characterized by more continuous, accurate and clear edge detection, which effectively avoids the appearance of false edges.
  • the traditional Laplacian edge detection operator in the form of a 3 ⁇ 3 matrix can be used, and its function model is:
  • the coefficient of the center point is 0, the remaining adjacent coefficients are negative, and the sum of the coefficients is 0.
  • the algorithm in the form of a 3 ⁇ 3 matrix ignores some real edges in some non-special directions, such as when calculating the inside of the arc, the left and right ends of the arc column, etc. due to the different distances from the calculation point to the center point.
  • the Laplacian edge detection operator in the form of a 5 ⁇ 5 matrix is the most ideal for the processing of arc images. Therefore, the Laplacian edge detection operator in the form of a 5 ⁇ 5 matrix is selected in this embodiment, and its function model is:
  • Step 103 after preprocessing the arc image, perform edge extraction on it to determine the positions of the moving and static contacts.
  • the Find Straight Edge function in LabVIEW is used for edge extraction, and the positions of the upper and lower edges of the arc in the arc image are accurately located, that is, the position of the static contact and the position of the moving contact in the current frame of the arc image are determined.
  • the position of the arc image after edge extraction is shown in Figure 2-5, and then the displacement difference of the moving contact in the two adjacent arc images can be determined, which is called the image displacement difference of the moving contact.
  • the parameters such as edge strength, edge polarity, and search distance are set as follows: when searching for the lower edge, the search line direction of the target area ROI (Region of interest) is from bottom to top, and the edge polarity selection is from black to white. a little. Manually set the minimum edge strength to 50, the search interval to 1, and the operator size and projection width to 13.
  • the Vision processing package detects all pixel points, filters out the edge points that meet the conditions, and performs fitting.
  • searching for the top edge just change the search direction from top to bottom and keep the rest of the parameters unchanged, then the top edge of the contact image can be extracted.
  • the background brightness of the vacuum arc image is small and the gray value is extremely low.
  • the arc part has high brightness and extremely high gray value. If the edge strength value is too small, more false edges will be generated. If the intensity value is too large, although false edges can be suppressed, some real edges will be lost. It has been verified that the minimum edge strength is set to 50, and the detection effect of the arc edge is the best.
  • Step 104 Determine the actual displacement difference of the moving contacts in two adjacent frames of arc images.
  • a geometrically constrained calibration algorithm is used to calibrate the image displacement and the actual displacement of the contact, and the image displacement difference of the moving contact in the adjacent two frames of arc images is converted into the actual displacement difference.
  • the conversion formula between the image displacement difference of the moving contact and the actual displacement difference is shown in formula (4), and the displacement difference in pixel unit can be converted into the actual displacement difference in millimeter unit.
  • s is the actual displacement difference
  • the unit is mm
  • D is the diameter of the moving contact
  • the unit is mm
  • N' is the number of pixels of the moving contact
  • N is the image displacement difference of the moving contact.
  • Step 105 Calculate the ratio between the actual displacement difference of the moving contacts in the two adjacent arc images and the time interval between the two adjacent arc images, and determine the movement speed of the vacuum switch contacts.
  • an arc width threshold can be set.
  • the current arc image is subjected to subsequent processing and analysis only when the arc width of the arc image exceeds the arc width threshold.
  • the opening speed curve of the moving contact can be calculated, and then the average opening speed, rigid opening speed and other motion parameters of the vacuum switch can be calculated.
  • edge detection operators are used to process the arc image, so as to illustrate the effectiveness of the method for detecting the movement speed of the vacuum switch contact according to the embodiment of the present application.
  • Figures 3-1, 3-2, 3-3 and 3-4 are the arcs after edge enhancement processing using Robert edge detection operator, Sobel edge detection operator, Prewitt edge detection operator and Laplacian edge detection operator respectively image. It can be seen from the detection results that the Sobel operator and the Prewitt operator have similar working principles, so the detection results of the arc image edge are not much different. The edge of the image detected by these two detection operators is not clear enough, especially the Sobel operator is not sensitive to the pixels with large gray value in the arc image.
  • the arc edge detected by Robert operator is thinner, and the brightness is the lowest among several detection operators.
  • the edge of the image detected by the Laplacian operator has high definition, and the brightness is also the highest among several detection operators. Through experimental comparison, the Laplacian operator is the best for edge detection of arc images.
  • an embodiment of an apparatus for detecting movement speed of a vacuum switch contact includes a memory 401 , a processor 402 and an internal bus 403 . communication with each other.
  • the processor may be a processing device such as a microprocessor (Micro Control Unit, MCU), a programmable logic device (Field Programmable Gate Array, FPGA).
  • MCU Micro Control Unit
  • FPGA Field Programmable Gate Array
  • the memory can be all kinds of memories that use electrical energy to store information, such as random access memory (Random Access Memory, RAM), read-only memory (Read-Only Memory, ROM), etc.; all kinds of memories that use magnetic energy to store information, such as hard disks , floppy disk, magnetic tape, magnetic core memory, magnetic bubble memory, U disk, etc.; all kinds of memory that use optical means to store information, such as CD, DVD, etc.
  • RAM Random Access Memory
  • ROM read-only memory
  • ROM read-only memory
  • the processor can call the logic instructions in the memory to realize a method for detecting the movement speed of the contacts of the vacuum switch. The method is described in detail in the method embodiment.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Multimedia (AREA)
  • Electromagnetism (AREA)
  • Power Engineering (AREA)
  • Image Analysis (AREA)

Abstract

一种真空开关触头的运动速度检测方法及装置。该方法首先获取真空开关在分闸过程中的电弧图像；然后采用拉普拉斯边缘检测算子对获取的各帧电弧图像进行边缘加强处理；接着进行边缘提取，确定各帧电弧图像中电弧的上、下边缘的位置，以确定各帧电弧图像中动、静触头的位置，并结合所述两帧电弧图像之间的时间间隔，确定真空开关触头的运动速度。采用拉普拉斯边缘检测算子对获取的各帧电弧图像进行边缘加强处理，加强了上、下边缘与电弧图像的背景图像的明暗对比，避免了对整帧电弧图像的处理分析，减少了计算量。便于分析操动机构的运动特性对电弧形态的影响，为电弧调控理论的研究提供技术支持。

Description

一种真空开关触头的运动速度检测方法及装置
相关申请的交叉引用
本申请基于申请号为202010790046.3、申请日为2020年08月07日的中国专利申请提出，并要求该中国专利申请的优先权，该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本申请属于真空开关技术领域，具体涉及一种真空开关触头的运动速度检测方法及装置。
背景技术
真空开关作为电力系统的控制和保护设备，具有使用寿命长、无污染、维护方便、结构紧凑、重量轻等优点。真空开关主要包括真空灭弧室和操动机构两部分，当电力系统出现故障时，如何实现操动机构可靠动作是实现灭弧室成功开断电流的关键。分闸速度作为操动机构运动参数之一，是影响真空开关开断能力的重要因素之一。若分闸速度过快，将导致分闸弹振幅度增大，进而破坏波纹管、屏蔽罩等结构；若分闸速度过慢，燃弧时间过长，则导致触头灼损严重等情况发生。
目前采用的分闸速度检测方法主要有电磁信号传感器检测方法及光电传感器检测方法。利用电磁信号传感器的检测方法，只适用于操动机构慢速工作的测量需求，而光电传感器的检测方法，容易受外界信号的干扰，传感器与操动机构出现相对运动情况。
相关技术实现了计算真空开关触头位置和运动速度的方法，该方法获取真空开关触头和电弧运动的连续图像，采用了图像分析技术，确定每帧 图像中电弧上、下边缘位置(即真空开关的静触头的位置和动触头在当前帧所处的位置)，进而便可根据这些位置信息和帧差时间便可计算得到动触头的运动速度。该方法在对电弧图像进行预处理后，直接采用网格划分、设定阈值的方法来分析得到每帧图像中电弧上、下边缘位置，但由于电弧图像受外界信号的干扰，图像质量不佳，不利于后期的识别与提取上、下边缘处理；而且该种处理方式需要对整个电弧图像进行分析处理，计算处理量较大。
发明内容
本申请实施例提供了一种真空开关触头的运动速度检测方法及装置，用以解决现有技术造成的计算处理量较大的问题。
为解决上述技术问题，本申请实施例的技术方案包括：
本申请实施例提供了一种真空开关触头的运动速度检测方法，包括如下步骤：
获取真空开关在分闸过程中的电弧图像；
采用拉普拉斯边缘检测算子对获取的各帧电弧图像进行边缘加强处理；
对边缘加强处理后的各帧电弧图像进行边缘提取，确定各帧电弧图像中电弧的上、下边缘的位置，以确定各帧电弧图像中动、静触头的位置；
根据电弧图像中动、静触头的位置，确定两帧电弧图像中动触头的实际位移差，并结合所述两帧电弧图像之间的时间间隔，确定真空开关触头的运动速度。
上述技术方案的有益效果为：本申请在获取得到真空开关在分闸过程中的电弧图像后，首先采用拉普拉斯边缘检测算子对获取的各帧电弧图像进行边缘加强处理，加强了上、下边缘与电弧图像的背景图像的明暗对比，避免了对整帧电弧图像的处理分析，减少了计算量，方便后续处理中快速 准确地提取得到上、下边缘，使边缘检测更加连续、准确和清晰，有效避免了伪边缘的出现，进而使得计算得到真空开关触头的运动速度更加准确，便于分析操动机构的运动特性对电弧形态的影响，为电弧调控理论的研究提供技术支持。
作为方法的进一步改进，所述拉普拉斯边缘检测算子的函数模型为5×5矩阵形式，所使用的函数模型为：
Figure PCTCN2020132471-appb-000001
其中，
Figure PCTCN2020132471-appb-000002
分别是x、y方向的函数模型。
采用5×5矩阵形式的拉普拉斯边缘检测算子，提高了边缘检测精度。
作为方法的进一步改进，还包括采用中值滤波算法对获取的电弧图像进行降噪处理的步骤。采用中值滤波算法可避免图像噪声对定位电弧上、下边缘位置造成的影响。
作为方法的进一步改进，所述两帧电弧图像中动触头的实际位移差为：
Figure PCTCN2020132471-appb-000003
式中，s为两帧电弧图像中动触头的实际位移差，D为动触头直径，N′为动触头像素个数，N为两帧电弧图像中动触头的图像位移差。
本申请实施例还提供了一种真空开关触头的运动速度检测装置，包括存储器和处理器，所述处理器用于执行存储在存储器中的指令以实现上述介绍的真空开关触头的运动速度检测方法，并达到与该方法相同的效果。
附图说明
此处的附图被并入说明书中并构成本说明书的一部分，示出了符合本 公开的实施例，并与说明书一起用于解释本公开的原理。
图1是本申请实施例的真空开关触头的运动速度检测方法实施例的流程图；
图2-1是本申请实施例的CMOS高速相机采集的电弧图像；
图2-2是本申请实施例的将电弧图像导入LabVIEW软件后的电弧图像；
图2-3是本申请实施例的进行中值滤波处理后的电弧图像；
图2-4是本申请实施例的采用拉普拉斯边缘检测算子进行边缘加强处理后的电弧图像；
图2-5是本申请实施例的提取上、下边缘后的电弧图像；
图3-1是采用Robert边缘检测算子进行边缘加强处理后的电弧图像；
图3-2是采用Sobel边缘检测算子进行边缘加强处理后的电弧图像；
图3-3是采用Prewitt边缘检测算子进行边缘加强处理后的电弧图像；
图3-4是采用Laplacian边缘检测算子进行边缘加强处理后的电弧图像；
图4是本申请实施例的真空开关触头的运动速度检测装置的结构图。
具体实施方式
下面结合附图及实施例，对发明的一种真空开关触头的运动速度检测方法和一种真空开关触头的运动速度检测装置进行详细说明。
方法实施例：
本申请实施例的一种真空开关触头的运动速度检测方法实施例，基于LabVIEW虚拟仪器开发平台，利用高速摄像机采集整个周期内触头运动图像，应用LabVIEW的图像处理模块对动、静触头边缘进行高精度定位，计算得到真空开关触头的运动速度，为真空开关电弧动态研究提供技术基础。LabVIEW采用的是图形化编程G语言，通过数据流实现编程，不仅能够做到对数据流的实时监控和直接编辑，还能做到与其他软件进行数据共享， 相较于采用MATLAB的C语言编程更加形象、直观，编程环境简单，能有效提高运算效率，在图像处理过程中不需要编写复杂的公式及运算程序。通过参数选择来提高图像处理的准确性，缩短了运算及编程时间。
下面结合图1，对整个检测方法进行详细说明。
步骤101，应用CMOS高速相机获取真空开关在分闸过程中的电弧图像。
具体地，利用CMOS高速像机对真空开关分闸过程中的电弧图像进行采集，如图2-1所示，CMOS的模拟信号经A/D转换成图像数字信号，并将获取的电弧图像(共计N帧)传输至计算机，如图2-2所示，该计算机中安装有LabVIEW图像处理软件。
步骤102，利用LabVIEW图像处理软件，对电弧图像进行预处理。
具体地，该预处理过程对分闸电弧图像依次进行灰度化和二值化处理，然后采用滤波算法进行降噪处理，采用拉普拉斯边缘检测算子(Laplacian边缘检测算子)进行边缘加强处理。
电弧图像在图像采集及传送过程中，受到光线、大气运动等外界因素的影响，图像质量下降，使图像产生椒盐噪声等污染。因此在接收到电弧图像之后，需采用滤波算法对电弧图像中的椒盐噪声进行处理。本实施例中，滤波算法为中值滤波算法，中值滤波算法能够在有效平滑脉冲噪声的同时，保护图像尖锐的边缘，对电弧图像的两级有很好的保护作用，处理效果较理想，其处理后的图像如图2-3所示，。中值滤波的函数可以表示为：
Figure PCTCN2020132471-appb-000004
其中，y为序列x 1,x 2,x 3,...,x n的中值。当选定邻域为一维时，中值滤波为一个含有奇数像素点的邻域，邻域中心像素点的像素值用中值代替。
图像进行降噪处理后，采用拉普拉斯边缘检测算子对电弧图像进行边缘检测，对电弧图像进行边缘加强处理，处理后的图像如图2-4所示。当图像中出现一个灰度值高的像素点，通过用Laplacian算子进行边缘检测后，能能够实现边缘强化的作用。拉普拉斯边缘检测算子特点是边缘检测更加连续、准确和清晰，有效避免了伪边缘的出现。可选用传统的3×3矩阵形式的拉普拉斯边缘检测算子，其函数模型为：
Figure PCTCN2020132471-appb-000005
其中，中心点的系数为0，其余相邻系数为负，系数总和为0。
实验中发现，该3×3矩阵形式的算法在一些非特殊方向，例如在计算电弧内部、弧柱左右两端等区域时，因计算点到中心点的距离不同，而忽略部分真实边缘。经多次实验发现，选用5×5矩阵形式的拉普拉斯边缘检测算子对电弧图像的处理效果最理想。故本实施例中选用5×5矩阵形式的拉普拉斯边缘检测算子，其函数模型为：
Figure PCTCN2020132471-appb-000006
其中，
Figure PCTCN2020132471-appb-000007
分别是x、y方向的函数模型。
步骤103，将电弧图像进行预处理后，对其进行边缘提取，确定动、静触头位置。
本实施例中采用LabVIEW中的Find Straight Edge函数进行边缘提取，准确定位到电弧图像中电弧的上、下边缘的位置，即确定了静触头的位置、以及动触头在当前帧电弧图像中所处的位置，边缘提取后的电弧图像如图2-5所示，进而便可确定相邻两帧电弧图像中动触头的位移差，称为动触头 的图像位移差。
其中，边缘强度、边缘极性、搜索间距等参数设置如下：在搜索下边缘时，目标区域ROI(Region of interesting)的搜索线方向从下到上，边缘极性选择由黑到白，查找第一点。手动设置边缘最小强度50，搜索间隔为1，算子尺寸以及投影宽度设置13。通过以上参数设置，Vision处理包对所有像素点进行检测，筛选出符合条件的边缘点，进行拟合。在搜索上边缘时，只需将搜索方向改为从上到下，其余参数设置不变，便可提取出触头图像上边缘。真空电弧图像的背景亮度较小，灰度值极低。而电弧部分亮度较大，灰度值极高。若边缘强度值太小会产生较多的伪边缘。若强度值太大，虽然能抑制伪边缘，但会丢失部分真实边缘。经过验证将最小边缘强度设为50，对电弧边缘的检测效果最理想。
步骤104，确定相邻两帧电弧图像中动触头的实际位移差。
具体地，采用几何约束的标定算法对触头的图像位移与实际位移进行标定，将相邻两帧电弧图像中动触头的图像位移差转化为实际位移差。其中，动触头的图像位移差与实际位移差的转换公式如式(4)所示，可以将像素单位位移差转化为毫米单位的实际位移差。
Figure PCTCN2020132471-appb-000008
其中，s为实际位移差，单位为mm；D为动触头直径，单位为mm；N′为动触头像素个数；N为动触头的图像位移差。
步骤105，计算相邻两帧电弧图像中动触头的实际位移差与相邻两帧电弧图像之间的时间间隔的比值，确定真空开关触头的运动速度。
具体地，根据相邻两帧电弧图像中动触头的实际位移差s以及相邻两帧电弧图像的时间间隔T，计算相邻两帧间平均速度v，如式(5)所示，并将帧间平均速度v作为瞬时速度。
Figure PCTCN2020132471-appb-000009
另外，在检测动、静触头位置的过程中，若在电弧拉弧的起始阶段，电弧面积较小，通过定位动、静触头上下边缘的方法检测触头的运动速度存在误差。故可设定一个电弧宽度阈值，在步骤103中，当电弧图像的电弧宽度超过该电弧宽度阈值时才对当前电弧图像进行后续的处理与分析。
根据该方法便可计算得到动触头的分闸速度曲线，进而计算得到真空开关的平均分闸速度、刚分速度及其他运动参数，为分析操动机构的运动特性对电弧形态的影响及电弧调控理论的研究提供技术支持。
下面采用其他边缘检测算子对电弧图像进行处理，以说明本申请实施例的真空开关触头的运动速度检测方法的有效性。
如图3-1、3-2、3-3、3-4分别是采用Robert边缘检测算子、Sobel边缘检测算子、Prewitt边缘检测算子和Laplacian边缘检测算子进行边缘加强处理后的电弧图像。通过检测结果可以看出，Sobel算子与Prewitt算子因为工作原理相似，所以对电弧图像边缘的检测结果相差不大。这两种检测算子检测出来的图像边缘清晰度不够，特别是Sobel算子对电弧图像内部灰度值较大的像素点不敏感。Robert算子检测出来的电弧边缘较细，且亮度是几种检测算子中最低的。而Laplacian算子检测出来的图像边缘清晰度高，亮度也是几种检测算子中最高的。通过实验对比，选用Laplacian算子对电弧图像进行边缘检测的效果最好。
装置实施例：
本申请实施例的一种真空开关触头的运动速度检测装置实施例，如图4所示，包括存储器401、处理器402和内部总线403，处理器402和存储器401之间通过内部总线403完成相互间的通信。
处理器可以为微处理器(Micro Control Unit，MCU)、可编程逻辑器件(Field Programmable Gate Array，FPGA)等处理装置。
存储器可为利用电能方式存储信息的各式存储器，随机存取存储器(Random Access Memory，RAM)、只读存储器(Read-Only Memory，ROM)等；利用磁能方式存储信息的各式存储器，例如硬盘、软盘、磁带、磁芯存储器、磁泡存储器、U盘等；利用光学方式存储信息的各式存储器，例如CD、DVD等。当然，还有其他方式的存储器，例如量子存储器、石墨烯存储器等。
处理器可以调用存储器中的逻辑指令，以实现一种真空开关触头的运动速度检测方法。在方法实施例中对该方法做了详细介绍。

Claims (5)

  1. 一种真空开关触头的运动速度检测方法，包括如下步骤：
    获取真空开关在分闸过程中的电弧图像；
    采用拉普拉斯边缘检测算子对获取的各帧电弧图像进行边缘加强处理；
    对边缘加强处理后的各帧电弧图像进行边缘提取，确定各帧电弧图像中电弧的上、下边缘的位置，以确定各帧电弧图像中动、静触头的位置；
    根据电弧图像中动、静触头的位置，确定两帧电弧图像中动触头的实际位移差，并结合所述两帧电弧图像之间的时间间隔，确定真空开关触头的运动速度。
  2. 根据权利要求1所述的真空开关触头的运动速度检测方法，其中，所述拉普拉斯边缘检测算子的函数模型为5×5矩阵形式，所使用的函数模型为：
    Figure PCTCN2020132471-appb-100001
    其中，
    Figure PCTCN2020132471-appb-100002
    分别是x、y方向的函数模型。
  3. 根据权利要求1或2所述的真空开关触头的运动速度检测方法，其中，还包括采用中值滤波算法对获取的电弧图像进行降噪处理的步骤。
  4. 根据权利要求1所述的真空开关触头的运动速度检测方法，其中，所述两帧电弧图像中动触头的实际位移差为：
    Figure PCTCN2020132471-appb-100003
    式中，s为两帧电弧图像中动触头的实际位移差，D为动触头直径，N′ 为动触头像素个数，N为两帧电弧图像中动触头的图像位移差。
  5. 一种真空开关触头的运动速度检测装置，包括存储器和处理器，所述处理器用于执行存储在存储器中的指令以实现如权利要求1～4任一项所述的真空开关触头的运动速度检测方法。
PCT/CN2020/132471 2020-08-07 2020-11-27 一种真空开关触头的运动速度检测方法及装置 WO2022027877A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010790046.3 2020-08-07
CN202010790046.3A CN114066928A (zh) 2020-08-07 2020-08-07 一种真空开关触头的运动速度检测方法及装置

Publications (1)

Publication Number Publication Date
WO2022027877A1 true WO2022027877A1 (zh) 2022-02-10

Family

ID=80119957

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/132471 WO2022027877A1 (zh) 2020-08-07 2020-11-27 一种真空开关触头的运动速度检测方法及装置

Country Status (2)

Country Link
CN (1) CN114066928A (zh)
WO (1) WO2022027877A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116660741A (zh) * 2023-07-26 2023-08-29 天津宜科自动化股份有限公司 一种用于智能基站的断路器关断检测方法
CN117349710A (zh) * 2023-12-04 2024-01-05 七七七电气科技有限公司 一种真空灭弧室开断能力的预测方法、电子设备及介质

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115077605A (zh) * 2022-06-06 2022-09-20 昂顿科技（上海）有限公司 智能变电站管理系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE3539748A1 (de) * 1985-11-09 1987-05-21 Sachsenwerk Ag Pruefeinrichtung fuer vakuumschaltkammern
CN102520328A (zh) * 2011-12-26 2012-06-27 大连理工大学 一种可调速操动的真空电弧观测装置
CN103604942A (zh) * 2013-09-05 2014-02-26 大连交通大学 计算真空开关触头位置和运动速度的方法
CN109949263A (zh) * 2017-12-18 2019-06-28 西安交通大学 一种气体电弧电极喷溅烧蚀的原位观测方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE3539748A1 (de) * 1985-11-09 1987-05-21 Sachsenwerk Ag Pruefeinrichtung fuer vakuumschaltkammern
CN102520328A (zh) * 2011-12-26 2012-06-27 大连理工大学 一种可调速操动的真空电弧观测装置
CN103604942A (zh) * 2013-09-05 2014-02-26 大连交通大学 计算真空开关触头位置和运动速度的方法
CN109949263A (zh) * 2017-12-18 2019-06-28 西安交通大学 一种气体电弧电极喷溅烧蚀的原位观测方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
LIU SHUNXIN: "Design and Characteristics Analysis for the Image Collection System of Vacuum Arcs", MASTER’S DISSERTATION, 31 December 2006 (2006-12-31), XP055893712 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116660741A (zh) * 2023-07-26 2023-08-29 天津宜科自动化股份有限公司 一种用于智能基站的断路器关断检测方法
CN116660741B (zh) * 2023-07-26 2023-11-14 天津宜科自动化股份有限公司 一种用于智能基站的断路器关断检测方法
CN117349710A (zh) * 2023-12-04 2024-01-05 七七七电气科技有限公司 一种真空灭弧室开断能力的预测方法、电子设备及介质
CN117349710B (zh) * 2023-12-04 2024-03-22 七七七电气科技有限公司 一种真空灭弧室开断能力的预测方法、电子设备及介质

Also Published As

Publication number Publication date
CN114066928A (zh) 2022-02-18

Similar Documents

Publication Publication Date Title
WO2022027877A1 (zh) 一种真空开关触头的运动速度检测方法及装置
CN108062525B (zh) 一种基于手部区域预测的深度学习手部检测方法
CN113313677B (zh) 一种卷绕锂电池x光图像的质量检测方法
JPH10506206A (ja) 医・生物学標本の自動焦点装置
WO2021109697A1 (zh) 字符分割方法、装置以及计算机可读存储介质
CN109685766A (zh) 一种基于区域融合特征的布匹瑕疵检测方法
CN110728668B (zh) 一种小目标形态保持的空域高通滤波器
WO2020243992A1 (zh) 坡口缺陷检测方法及装置
CN110991452B (zh) 一种车位框检测方法、装置、设备及可读存储介质
CN105719275A (zh) 一种并行组合的图像缺陷分割方法
CN105787912B (zh) 一种基于分类的阶跃型边缘亚像素定位方法
CN110032946A (zh) 一种基于机器视觉的铝/铝泡罩包装药片识别与定位方法
CN110853077B (zh) 一种基于形态变化估计的自适应红外动态帧特征提取方法
CN114140384A (zh) 基于轮廓拟合与形心跟踪的提升钢丝绳横向振动图像识别算法
CN105930811B (zh) 基于图像处理的手掌纹理特征检测方法
CN111915558A (zh) 一种高压输电线销钉状态检测方法
CN103955917A (zh) 一种长间隙空气电弧图像边缘检测方法
Xiang et al. Research on target detection method based on HSV fusion Gaussian mixture model
CN108549889B (zh) 一种简易的印刷体数字识别方法
CN112288049B (zh) 基于模板匹配和限制连通域指针定位的指针表读数方法
CN114445358A (zh) 手机中框边缘外观缺陷检测方法、系统、存储介质及计算机设备
CN109784176B (zh) 车载热成像行人检测RoIs提取方法和装置
WO2015102945A1 (en) Image processing-based method and apparatus
KR102659027B1 (ko) 이동 객체 검출 방법, 이를 이용하는 이동 객체 검출 장치 및 프로그램
CN114092458B (zh) 一种基于改进NanoDet深度网络的发动机浓烟淡烟自动检测方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20948318

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20948318

Country of ref document: EP

Kind code of ref document: A1