• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.2019-2023
• /
• 2005

In this paper, an autonomous omni directional cleaning robot which recognizes an obstacle and a battery charger is introduced. It utilizes a robot vision, ultra sonic sensors, and infrared sensors information along with appropriate algorithm. Three omni-directional wheels make the robot move any direction, enabling a faster maneuvering than a simple track typed robot. The robot system transfers command and image data through Blue-tooth wireless modules to be operated in a remote place. The robot vision associated with sensor data makes the robot proceed in an autonomous behavior. An autonomous battery charger searching is implemented by using a map-building which results in overcoming the error due to the slip on the wheels, and camera and sensor information.

• Journal of Korean Society of Occupational and Environmental Hygiene
• /
• v.33 no.4
• /
• pp.517-527
• /
• 2023

Objectives: We examine cases of chemical poisoning that occurred in the cleaning of metal parts and the regulations on halogenated solvents in other countries and propose regulations necessary to prevent chemical poisoning from halogenated solvents. Methods: We collected cases of chemical poisoning through the website of the Korea Occupational Safety and Health Agency. A review of the literature was conducted focusing on regulations related to halogenated solvents in the United States and the European Union, particularly for cleaning metal parts. Among the Material Safety Data Sheets submitted to the government, MSDS containing eleven substances were extracted to confirm the composition and product use. We investigated cleaning methods for metal parts used in South Korea. For the hazard classification, the European Chemicals Agency or Japan's NITE's website was used. Results: In the case of poisoning, the cleaning methods involving trichloromethane were dipping and dry, which was not found in the literature. It was confirmed that many halogenated solvents and dimethyl carbonate were used for metal cleaning in South Korea. In vapor degreasing using TCE in the USA, even if the facility is strictly managed, such as by installing cooling coils in open cleaning facilities, the risk of exposure to TCE is considered to be not only carcinogenic but also a concern for acute and chronic effects. In comparison, exposure through Korean work methods such as dipping and drying operations is inevitably much higher. Conclusions: The transition to water-based cleaning with low-hazard chemicals should be a priority in the cleaning process. In the case of metal parts that require precise cleaning, if the use of a halogenated solvent is inevitable, a closed degreasing facility should be used to minimize exposure. The current regulations in the Occupational Safety and Health Act, the Chemical Substances Control Act, and the Air Environment Conservation Act do not require cleaning facilities to minimize emissions. To protect the health of workers using halogenated solvents to clean metal parts, regulations that require a fundamental reduction in exposure will be necessary.

• Clean Technology
• /
• v.19 no.2
• /
• pp.184-188
• /
• 2013

To increase flash point which is related to flammability, seven unprecedented new cleaning agents containing fluoride atoms have been invented. These newly synthesized cleaning agents's physical properties which were conducted by Korea Institute of Petroleum Management by using a standard method showed excellent values. Particularly, flash point of newly synthetic cleaning agents is more higher than that of fluoride free compound. A specimen for cleaning ability was prepared by cutting in $60mm{\times}40mm$ size of stainless steel plate. The surface of the above specimens was applied with four kinds of contaminants, such as paraffin based drawing oil, flux abietic acid, water-insoluble cutting oil, and lubricating oil. Contaminated specimens were immersed in new compounds (1-7) for 1 to 5 minutes to dissolve oil in the cleaning agent. Although the data indicate that all compounds (1-7) exhibit lower cleaning ability toward cutting oil, it is observed that in the case of the present study more than 80% of pollutants on the surface were almost removed within 5 minutes.

• Membrane Journal
• /
• v.27 no.6
• /
• pp.519-527
• /
• 2017

Membrane fouling is the main issue hindering the expansion of low pressure membrane processes for surface water treatment. Therefore, applying periodic hydraulic cleaning for fouling control should be well optimized. Better understanding of membrane fouling associated with periodic hydraulic cleaning would be useful to optimize membrane cleaning strategies. By comparing experimental permeability data with the classical Hermia blocking laws, this study aims at analyzing membrane fouling and understanding dominant fouling mechanisms occurring when filtering a synthetic surface water solution with a pressurized membrane process during six filtration cycles of 30 min each, separated with cyclic cleaning of 1 min by backwashing and forward flushing separately and combined. When applying single cleaning technique, membrane fouling during the first cycles was controlled by complete blocking mechanism while the last cycles were dominated by cake formation. Nevertheless, when combining cleaning technique better membrane regeneration was obtained and fouling was mainly due to cake formation.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.4 no.1
• /
• pp.119-124
• /
• 2004

Real data as web log file tend to be incomplete. But we have to find useful knowledge from these for optimal decision. In web log data, many useful things which are hyperlink information and web usages of connected users may be found. The size of web data is too huge to use for effective knowledge discovery. To make matters worse, they are very sparse. We overcome this sparse problem using Markov Chain Monte Carlo method as multiple imputations. This missing value imputation changes spare web data to complete. Our study may be a useful tool for discovering knowledge from data set with sparseness. The more sparseness of data in increased, the better performance of MCMC imputation is good. We verified our work by experiments using UCI machine learning repository data.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2001.10a
• /
• pp.411-415
• /
• 2001

Advances in computer and sensor technology have made it possible to obtain superlarge manufacturing process data in real time, letting us to extract meaningful information from these superlarge data sets. We propose a systematic data analysis procedure which field engineers can apply easily to manufacture quality products. The procedure consists of data cleaning and data analysis stages. Data cleaning stage is to construct a database suitable for statistical analysis from the original superlarge manufacturing process data. In the data analysis stage, we suggest a graphical easy-to-implement approach to extract practical information from the cleaned database. This study will help manufacturing companies to achieve six sigma quality.

• The Journal of Korea Robotics Society
• /
• v.4 no.1
• /
• pp.74-80
• /
• 2009

This paper presents a remote monitoring and simulation system for a building cleaning mobile robot. It provides a tool of convenient 3D graphical map construction including network camera image viewer and status information of the robot. The 3D map is reconstructed from existing 2D building CAD data with DXF format using OpenGL graphic API. Through this system, it is possible to monitor and control the cleaning mobile robot from remote place. A practical experiment is performed to show the reliability and convenience of the monitoring system. The proposed system is expected to give efficient way of control and monitoring to building cleaning mobile robot.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.3
• /
• pp.1639-1658
• /
• 2019

Named entity recognition (NER) is an important technique for improving the performance of data mining and big data analytics. In previous studies, NER systems have been employed to identify named-entities using statistical methods based on prior information or linguistic features; however, such methods are limited in that they are unable to recognize unregistered or unlearned objects. In this paper, a method is proposed to extract objects, such as technologies, theories, or person names, by analyzing the collocation relationship between certain words that simultaneously appear around specific words in the abstracts of academic journals. The method is executed as follows. First, the data is preprocessed using data cleaning and sentence detection to separate the text into single sentences. Then, part-of-speech (POS) tagging is applied to the individual sentences. After this, the appearance and collocation information of the other POS tags is analyzed, excluding the entity candidates, such as nouns. Finally, an entity recognition model is created based on analyzing and classifying the information in the sentences.

• Journal of Fisheries and Marine Sciences Education
• /
• v.26 no.1
• /
• pp.34-48
• /
• 2014

Marine algal assemblages on the artificial reefs at three stations (Haengwon, Geumneung, Pyoseon in Jeju-do) and the growth pattern of Ecklonia cava with water depth were studied before and after rocky cleaning. Nine algal species occurred at three artificial reefs before rocky cleaning in July 2012, however, 19 algal species were found at three artificial reefs after rocky cleaning in July 2013. In particular, 13 of 19 species in 2013 were replaced by different species which were not found in July 2012. Algal biomass rapidly increased in July 2013 after rocky cleaning. The nMDS plot based on the presence and absence data of macro-algal assemblages on the artificial reefs showed that the species compositions between artificial reefs were similar to each other before rocky cleaning operations in July of 2012. However, after rocky cleaning, the species composition of macro-algal assemblages in Haengwon region was similar to that in Pyoseon region while that in Geumneung region was different to those in Haengwon and Pyseon regions in July 2013. It is needed to clarify the suitable water depth for transplantation of perennial Phaeophyta E. cava. According to the data on seasonal changes in total length, total weight, blade length, blade width, blade weight of E. cava, it began to grow rapidly from April and reached to the maximum value in June, and then degeneration of the blade occurred in July and continued to early August. Although the total length of E. cava at 10m water depth was larger than those at 5 m and 15 m water depths, there was no significant difference statistically among water depths by ANOVA test. However, in the values of total weight, blade length, blade width, and blade weight, the growth patterns at 5 m water depth were larger than those at 10 m and 15 m. Thus, it is assumed that the most suitable water depth for transplantation and its maximum growth and effective release of zoospores of E. cava will be 7~10 m water depth.

• International Journal of Contents
• /
• v.16 no.2
• /
• pp.102-108
• /
• 2020

The purpose of this study was to estimate and analyze the potential radiation dose that the future visitors and the cleaning staff will be exposed to when the KRR-1 reactor is converted into a memorial hall. The radiation doses were estimated using the RESRAD-BUILD software, where case, building, receptor, shielding, and source parameters were applied as the input data. Also, the basic data for the assessment of the radiation doses were determined in an indirect manner using the data on the waste generated during the decommissioning process of the reactor. The assessment results indicate that the potential radiation dose to the visitors and the cleaning staff will be less than 1 mSv, the annual dose limit for the general public. However, if anyone for a significant period of time is close to the reactor, the overall dose will increase. The radiation dose for the future visitors and the cleaning staff was determined to be lower than the annual dose limit for the general public. Given such a risk, systematic measures, such as periodic monitoring or limiting hours, are imperative.