• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 한국공작기계학회 2004년도 춘계학술대회 논문집
• /
• pp.253-257
• /
• 2004

Neural networks are works are used in the framework of sensor based tracking control of robot manipulators. They learn by practice movements the relationship between PSD(an analog Position Sensitive Detector) sensor readings for target positions and the joint commands to reach them. Using this configuration, the system can track or follow a moving or stationary object in real time. Furthermore, an efficient neural network architecture has been developed for real time learning. This network uses multiple sets of simple back propagation networks one of which is selected according to which division(corresponding to a cluster of the self-organizing feature map) in data space the current input data belongs to. This lends itself to a very training and processing implementation required for real time control.

• Journal of the Korean Society of Safety
• /
• 제14권2호
• /
• pp.163-169
• /
• 1999

Safety education should not only prevent workers from industrial accidents but also contribute to improve the productivity of manufacturing plants or construction sites. In practice this do not happen because workers do not realize the importance of safety education. This study aims to suggest a methodology to improve safety education of construction sites by surveying conditions of safety education and the safety consciousness of workers and engineers. The results showed that most education except regular educations were nominally carried out. Lectures and audio-visual education were mainly used as educational methods. After trainees attended the education session they completed a written survey, the most dissatisfied factor about safety education was education circumstances, of which rate was around 36%. The proportion of construction engineers who thought that safety management was contributable to cost reduction was 35%, to construction period 20%, and to quality enhancement 48%. Based on these results, this research pointed out the need to review training manuals, the development of educational programs, improvement of educational facilities to improve safety education of construction sites, and finally to discussed these issues.

• Proceedings of the Korean Professional Engineer Association Conference
• /
• 한국기술사회 1984년도 제14회 한일기술사 합동 심포지움 참관기
• /
• pp.77-82
• /
• 1984

The small group activities (SGA) was introduced into some of enterprises in Korea in 1967 and plant division, Saemaul undong Headquarters had encouraged Quality Control Circles (QCC) within the manufacturing corporations in assistance of central government. Registered small groups as of august 31, 1984 amounted up to 78,243 units and the number of members were 791,512. Meanwhile, small and medium industries have mostly introduced small group activities considerably later and less actively managed than large business. Main reasons for the less effectiveness of the activities are due to lack of management skills and less awareness of it from management and workers group. Effective small group activities are presumed to be successful only with labor management cooperation on the basis of human-oriented management philosophy. The small group activities are also prevalent in service sector. More derivative methods have been developed and more members are willingly participating in training programs. The small group which is basically a horizontal organization unit, promotes communication within the whole organization. In consideration of the social circumstances and traditions, the flexible model of the small group activities suitable to the corporate environment, will contribute to industrial development.

• Journal of Power System Engineering
• /
• 제17권6호
• /
• pp.11-17
• /
• 2013

Esterified bio-diesel oil is normally used as blend oil of 3% that and 97% diesel fuel in Korea. Since specifics of it is similar to that of diesel fuel, availability of non-esterified bio-diesel oil that has a lower expenses of manufacturing is worthy of attention. However, bio-diesel oil has a demerit which it emits typically more NOx emission than diesel fuel. In this study, characteristic tests using blending oil with 95% gas oil and 5% bio-diesel oil were achieved at lower common rail pressure in order to improve this demerit. It was noticed that non-esterified bio-diesel oil has more similar characteristics to diesel fuel than esterified bio-diesel oil and it emits more NO emission by fuel NO mechanism.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 한국공작기계학회 2001년도 추계학술대회(한국공작기계학회)
• /
• pp.99-104
• /
• 2001

The neural network method has been introduced to design a controller for linear motor feed system and system operation status was monitored. It is most difficult to achieve controller gain tuning because of the information limit. Regardless of the system structure, conventional control gain could be adjusted minimizing the resulting error for both position and velocity using the proposed method. Slight performance deterioration was observed at the small value of training epoch. Different controller performance for position was observed with respect changed sampling time. Actuated system performance was monitored using neural network signal processing and operational status was predicted with the rate of 80% approximately.

• Journal of the Korea Computer Industry Society
• /
• 제8권4호
• /
• pp.221-228
• /
• 2007

Many kinds of microcontrollers such as 8051, PIC16 and Atmega series are used for the automatic control system, home appliances and communication equipments manufacturing. It is very important to understand the basic operational principles of microcontrollers and their design concepts. There are many kinds of educational microcontroller trainers and also they are designed and assembled very complicatedly. For the students or developers, it is very difficult to catch the basic operation schemes and apply the techniques to the control system. And also it requires much cost and time for the various kinds of trainers purchasing. In this paper, standardization of pins layout and peripheral modules for the general microcontroller usage was introduced and tested with 89C2051, 89C51, PIC16F84, PIC16F877, Atmega8535 and Atmega128, etc. As a result of test, it was found that saving the cost and time using this suggested device was possible. And also it was very effective way to understand microcontroller design and programming techniques.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 한국공작기계학회 2003년도 추계학술대회
• /
• pp.137-142
• /
• 2003

A fundamental study for developing a system of fault diagnosis of a pump is performed by using neural network. The acoustic signals were obtained and converted to frequency domain for normal products and artificially deformed products. The signals were obtained in various driving frequencies in order to obtain many types of data from a limited number of pumps. The acoustic data in frequency domain were managed to multiples of real driving frequency with the aim of easy comparison. The neural network model used in this study was 3-layer type composed of input, hidden, and output layer. The normalized amplitudes at the multiples of real driving frequency were chosen as units of input layer, Various sets of teach signals made from original data by eliminating some random cases were used in the training. The average errors were approximately proportional to the number of untaught data. The results showed neural network trained by acoustic signals can be used as a simple method far a detection of machine malfunction or fault diagnosis.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 한국공작기계학회 2001년도 춘계학술대회 논문집(한국공작기계학회)
• /
• pp.239-243
• /
• 2001

Neural networks are used in the framework of sensorbased tracking control of robot manipulators. They learn by practice movements the relationship between PSD (an analog Position Sensitive Detector) sensor readings for target positions and the joint commands to reach them. Using this configuration, the system can track or follow a moving or stationary object in real time. Furthermore, an efficient neural network architecture has been developed for real time learning. This network uses multiple sets of simple backpropagation networks one of which is selected according to which division (corresponding to a cluster of the self-organizing feature map) in data space the current input data belongs to. This lends itself to a very training and processing implementation required for real time control.

• International Journal of Precision Engineering and Manufacturing
• /
• 제6권3호
• /
• pp.37-44
• /
• 2005

Gait control capacity for most trans-femoral prostheses is significantly different from that of a normal person, and training is required for a long period of time in order for a patient to walk properly. People become easily tired when wearing a prosthesis or orthosis for a long period typically because the gait angle cannot be smoothly adjusted during wearing. Therefore, to improve the gait control problems of a trans-femoral prosthesis, the proper gait angle is estimated through surface EMG(electromyogram) signals on a normal leg, then the gait posture which the trans-femoral prosthesis should take is calculated in the neural network, which learns the gait kinetics on the basis of the normal leg's gait angle. Based on this predicted angle, a postural control method is proposed and tested adaptively following the patient's gait habit based on the predicted angle. In this study, the gait angle prediction showed accuracy of over $97\%$, and the posture control capacity of over $90\%$.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• 제13권5호
• /
• pp.81-86
• /
• 2004

A fundamental study for developing a system of fault diagnosis of a pump is performed by using neural network. The acoustic signals were obtained and converted to frequency domain for normal products and artificially deformed products. The signals were obtained in various driving frequencies in order to obtain many types of data from a limited number of pumps. The acoustic data in frequency domain were managed to multiples of real driving frequency with the aim of easy comparison. The neural network model used in this study was 3-layer type composed of input, hidden, and output layer. The normalized amplitudes at the multiples of real driving frequency were chosen as units of input layer. Various sets of teach signals made from original data by eliminating some random cases were used in the training. The average errors were approximately proportional to the number of untaught data. The results showed neural network trained by acoustic signals can be used as a simple method for a detection of machine malfuction or fault diagnosis.