Abstract
Can we detect electric discharge states in gases based on the information on visual images? This article proposes a new kind of method where we build several detection models for different states of corona discharge by applying four kinds of machine learning algorithms to extract color, brightness, and shape information characteristics of visible images taken by a digital camera. Every model is then tested on a new set of images to measure its performance. The four different machine learning algorithms are support vector machine (SVM), K-nearest neighbor regression (KNN), single layer perceptron (SLP), and decision tree (DT) algorithms. The prediction results show that the color features perform best among all three types of features and the KNN algorithm performs best among all four algorithms. This article also presents a discussion on how to choose the optimal detection areas of images for better detection performance. Our approach shows consistent results across different cameras and camera settings. The results demonstrate that even if only the visible light spectrum emitted from plasma is captured, the color method can provide sufficient discharge information for economic and convenient use in discharge state detection because the species producing visible radiation are affected by radiation in all bands.
-
Author Keywords
- Color information,
- discharge state,
- gray-level histogram (GLH),
- machine learning,
- red,
- green,
- blue (RGB),
- visible image
-
IEEE Keywords
- Discharges (electric),
- Image color analysis,
- Machine learning algorithms,
- Corona,
- Machine learning,
- Support vector machines,
- Prediction algorithms
Introduction
The ELECTRIC discharge is widespread in nature and is commonly used in the operation of industrial equipment. In an effort to understand the properties of electrical discharge, researchers have investigated several physical quantities of discharge, including the voltage, current, optical spectrum, ultrahigh-frequency electromagnetic waves, number of discharges, phase angles (for ac discharge), etc. However, such research works have been carried out without considering the relevant optical characteristics of discharge images even though the initial studies were based on the image, such as that of the original meaning of corona. As a matter of fact, optical measurements are better suited than electromagnetic measurements for the purpose of determining discharge geometries. If the optical characteristics of the discharge image can be incorporated in the diagnosis, the reliability of traditional recognition methods may be increased.
In the past, the traditional research of the discharge image was conducted from a qualitative point of view, such as morphology description, strong or weak light intensity, etc. [1]–[5]. However, only a few studies focused on quantitative evaluation can be found. With the development of computer techniques, the digital image processing methods have been applied extensively to study discharge characteristics, such as breakdown paths, discharge area, etc., especially in the ultraviolet (UV), which can help tackle complex problems by using statistical techniques [6], [7] or the fractal theory [8]–[10]. Although the images obtained by a high-speed camera (nanosecond time scale) can provide some details of a single discharge [11]–[13], the essence of gas discharge remains random under the same macroscopic physical conditions. On the other hand, the discharge used in some industrial applications is a collection of a large number of micro discharges. Therefore, a statistical evaluation of discharge images covering a large number of stochastic processes on a long time scale is still of great significance, compared with the research methods of high-speed cameras. [14].
The color information produced by optical radiation has not been widely used in the study of discharge images. In 2000, Russell and Jones [15] proposed the use of chromatic attributes to directly monitor the stability of plasma states. However, the studies were then limited only to the use of optical-electrical detection techniques, which can only be applied to a relatively large area for achieving a general understanding [15]. In 2009, Koppisetty et al. [16] attempted to establish a correlation of color information of the visual images with the progress of partial-vacuum breakdown. In 2016, Serrano et al. [17] used color information to monitor arc welding. Developments in nonthermal plasmas have stagnated. We conducted research on the color difference in the corona and surface discharge and filed for patents on using the color information to detect the discharge state [18]–[20]. In 2017, Prasad and Reddy [21], [22] introduced a method for extracting color information from discharge images, which was then converted to brightness metrics to study the relationship with discharge power, which is important progress. To summarize, utilizing color information in the study of the spatial distribution of nonthermal plasma discharge is an emerging area of research, connected to the ground through a resistor (R = 50 ). The frequency of the applied ac voltage is 50 Hz. A high-voltage probe (Tektronix P6015A) is used to measure the applied voltage and the resistor is used to measure the current. The discharge images are taken with a Nikon D800, the ISO is at 2000, 3200, and 5000, respectively, and the exposure times at 2, 4, and 6 s, respectively. Fig. 1(b) and (c) shows the typical images of the corona caused by different voltages.
Conclusion
The prediction shows that the color features perform the best among all the three characteristics information and the KNN algorithm performs the best among all four algorithms. The model shows consistently good performance with different cameras and camera settings as well. Discharge produces radiation of UV, visible, and near-infrared wavelengths. Past studies focused mostly on the UV spectrum, yet the measurement of the light spectrum demonstrates that the radiation intensity of the visible spectrum can be high as well. Objects that can produce radiation of the visible spectrum are affected by radiation across all spectra. Because of this, even though our RGB-GLH method uses information only from the visible spectrum, it is still able to encompass discharge status-related information across spectra, thus enabling us to build a more successful model. The RGB color information characteristics method can also be applied to other discharge types other than corona.
About KSRA
The Kavian Scientific Research Association (KSRA) is a non-profit research organization to provide research / educational services in December 2013. The members of the community had formed a virtual group on the Viber social network. The core of the Kavian Scientific Association was formed with these members as founders. These individuals, led by Professor Siavosh Kaviani, decided to launch a scientific / research association with an emphasis on education.
KSRA research association, as a non-profit research firm, is committed to providing research services in the field of knowledge. The main beneficiaries of this association are public or private knowledge-based companies, students, researchers, researchers, professors, universities, and industrial and semi-industrial centers around the world.
Our main services Based on Education for all Spectrum people in the world. We want to make an integration between researches and educations. We believe education is the main right of Human beings. So our services should be concentrated on inclusive education.
The KSRA team partners with local under-served communities around the world to improve the access to and quality of knowledge based on education, amplify and augment learning programs where they exist, and create new opportunities for e-learning where traditional education systems are lacking or non-existent.
FULL Paper PDF file:
A Corona Recognition Method Based on Visible Light Color and Machine LearningBibliography
author
Year
2020
Title
A Corona Recognition Method Based on Visible Light Color and Machine Learning
Publish in
in IEEE Transactions on Plasma Science, vol. 48, no. 1, pp. 31-35, Jan. 2020
Doi
10.1109/TPS.2019.2947289
PDF reference and original file: Click here
Somayeh Nosrati was born in 1982 in Tehran. She holds a Master's degree in artificial intelligence from Khatam University of Tehran.
Professor Siavosh Kaviani was born in 1961 in Tehran. He had a professorship. He holds a Ph.D. in Software Engineering from the QL University of Software Development Methodology and an honorary Ph.D. from the University of Chelsea.
Nasim Gazerani was born in 1983 in Arak. She holds a Master's degree in Software Engineering from UM University of Malaysia.
