• Resources Recycling
• /
• v.30 no.4
• /
• pp.3-10
• /
• 2021

The field of recycling for waste electronic components, which is the typical example of an urban mine, requires the development of useful sorting techniques. In this study, a sorter based on image identification by deep learning was developed to select electronic components into four groups. They were recovered from waste printed circuit boards and should be separated to depend on the difference after treatment. The sorter consists of a workstation with GPU, camera, belt conveyor, air compressor. A small piece (less than 3.5 cm) of electronic components on the belt conveyor (belt speed: 6 cm/s) was taken and learned as teaching data. The accuracy of the image identification was 96% as kinds and 99% as groups. The optimum condition of sorting was determined by evaluating accuracies of image identification and recovery rates by blowdown when changing the operating condition such as belt speed and blowdown time of compressed air. Under the optimum condition, the accuracy of image classification in groups was 98.7%. The sorting rate was more than 70%.

• Proceedings of the KIEE Conference
• /
• 1999.07g
• /
• pp.3060-3063
• /
• 1999

We made a multi-layer conveyor sterilizing system using UV rays to prolong a storage period of dried marine products. The system is composed of 5 layer conveyor belt of 3 m long each, thus 15 m long in total. It is designed to radiate UV rays for 10 minutes the dried marine products, and can process 300 kg/hour. Eight UV lamps of 30 cm long and 15 W of power with 254 nm wave length are installed in parallel, 15 cm high above in each belt, and 40 lamps in total. The whole system is covered by 1 mm thick metal plates to block the UV rays except a front side which is covered with 5 mm thick plastic plates coated with UV protection film to survey inside of the system. A sterile ratio is about 99.97% with 10 minutes sterilization using this system.

• Journal of Power System Engineering
• /
• v.21 no.4
• /
• pp.42-50
• /
• 2017

A velocity controller using a modified model reference adaptive controller (M-MRAC) and a projection operator for a fish sorting belt conveyor system with uncertainty parameters, input saturation and bounded disturbances is proposed in this paper. To improve the tracking performance and robustness of the proposed controller in the presence of bounded disturbances, the followings are done. Firstly, the reference model for the conventional model reference adaptive controller (CMRAC) is replaced by a modified reference model for a M-MRAC to reduce unexpected high frequency oscillation in control input signal when the adaptation rate is increased. Secondly, estimated parameters in an adaptive law are varied smoothly under bounded external disturbances and a projection operator is utilized in an adaptive law for the proposed M-MRAC controller to be robust. Thirdly, an auxiliary error vector is introduced for compensating the error dynamics of the system when the saturation input occurs. Finally, the experimental results are shown to verify the better effectiveness and performance of the proposed controller under the bounded disturbance and saturated input than that of a CMRAC.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 1996.06c
• /
• pp.314-324
• /
• 1996

A high speed NMR quality evaluation sensor was designed , constructed and tested . The device consists of an NMR spectrometer coupled to a conveyor system. The conveyor was run at speeds ranging from 0 to 250 mm/s. Spectral of avocado fruits and one-dimensional magnetic resonance images of pickled olives were acquired while the samples were moving on a conveyor belt mounted through a 20Tesla NMR magnet with a 20 mm diameter surface coil and a 150 mm diameter imaging coil respectively. Fro a magnetic resonance spectrum analysis, motion through variations in the magnetic field tends to narrow spectral line width just like using sample rotation in high resolution NMR to narrow spectral line width. Spectrum analysis was used to detect the dry weight of avocado fruits using the ratio oil and water resonance peaks. Good correlations maximum r=0.970@ 50 mm/s and minimum r=0.894@250mm/s ) between oil and water resonance peak ratio and dry weight of avocados were observed at speeds ra ging from0 to 250mm/s. For the application of magnetic resonance imaging (MRI) method, the projections were used to distinguish between pitted and non-pitted olives . Effect of fruit position in the coil was tested and coil degree effects were noticed when projects were generated under dynamic conditions. Various belt speeds (up to 250mm/s) were tested and detection results were compared to static measurements. Higher classification errors were occurred at dynamic conditions compared to errors while olives were at rest.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.644-647
• /
• 2003

This paper describes a neural network based machine vision system designed for inspecting yellow beans in real time. The system consists of a camera. lights, a belt conveyor, air ejectors, and a computer. Beans are conveyed in four lines on a belt and their images are taken by a monochrome line scan camera when they fall down from the belt. Beans are separated easily from their background on images by back-lighting. After analyzing the image, a decision is made by a multilayer artificial neural network (ANN) trained by the error back-propagation (EBP) algorithm. We use the global mean, variance and local change of gray levels of a bean for the input nodes of the network. In an our experiment, the system designed could process about 520kg/hour.

• Journal of Sensor Science and Technology
• /
• v.17 no.2
• /
• pp.106-113
• /
• 2008

Perceiving a 3D moving target is important for automated machines including robots in a factory. Although stereovision or ultrasonic ranging is often employed for the purpose, the former requires high-cost systems, and the latter suffers from low-resolution. In this paper, we propose a PSD sensor array to acquire the 3D profile of an object conveyed by a belt. Rectangular parallelepiped target objects at random positions and orientations are assumed. Nine PSD sensors are configured in a $3{\times}3$ array above the belt, and fired in a sequence so that cross-talk can be avoided. Time gaps among sensor readings are compensated to get a result that is obtainable when sensors' measurements are made simultaneously along a sensing line. The system designed is tested with various objects in experiments.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2000.10a
• /
• pp.157-160
• /
• 2000

This study is concerned with how to optimize the Input sequence of product models with sequence -dependent setup time in Mixed-Model Assembly Line (MMAL) using conveyor system. Usually MMAL consists of a number of stations linked by conveyor belt and each station has a work zone limited by upstream and downstream boundaries. To avoid improper interference between operators in the adjacent stations and excess of machine moving range, operators are forced to complete their operations within their predetermined work zone. In this study, our goal is to determine sequence of models for minimizing the total work unfinished within their work zone. A generalized formulation of the product sequencing problem in MMAL is presented and we developed an optimal procedure using Branch & Bound. Also we developed a heuristic procedure using local search.

• The Journal of Korea Robotics Society
• /
• v.4 no.4
• /
• pp.320-326
• /
• 2009

This paper proposed a new magnetic field detection algorithm to detect metal pieces in food producing processes. This algorithm can detect mixed metal pieces by sensing magnetic field. Some metal pieces are passed through an over-current circuit to magnetize them. The magnetic field sensor can detect the change in the magnetic field on theconveyor belt caused by the flow of the metal pieces in the food product. However, such a method detects the output of signals that change their amplitude and phase according to the movement of the conveyor belt with the food product, in which the equilibrium of the positive signal that is created in the receiver coil loses its balance due to the magnetized material. This includes not only the signal elements resulting from the effect of the alternating magnetic fields of the mixed metals, but also the signal elements resulting from the effect of the alternating magnetic fields of the examined object itself.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10b
• /
• pp.1273-1275
• /
• 1996

Moisture measurements of s intering materials were performed using the simulator which can simulate various conditions taking place at the belt-conveyor of sintering process. The moisture sensor used in this experiment is NIR moisture meter(JE-330, KETT Electric Lab.). Using the simulator, we measured the moisture contents of sintering materials at different conditions. We found that the moisture content decreased as the rotation speed increased and the size of the sintering materials became smaller.

• Journal of Biosystems Engineering
• /
• v.41 no.3
• /
• pp.201-207
• /
• 2016

Purpose: A water-spraying type automatic glochids removal system for cactus (Opuntia humifusa) stem was developed, and its performance was evaluated. The system was developed to reduce intensive human labor in removing glochids from cactus stem skin without inducing damage prior to further processing into value-added products. Methods: The developed system consists of conveyor and water-spraying systems. The conveyor system delivers cactus stems through water-spraying compartments and finally to a collecting box. In order to remove the glochids, rotating nozzles spray water over all areas (i.e., front and back faces and sides) of the cactus skin under controlled water pressure. Operating conditions such as conveyor speed, water pressure, angle of water-spraying nozzles, distance between conveyor belt and rotating nozzles, and angle of cactus flipping slide were adopted from our previous study and applied on the system design and manufacturing. The performance of glochids removal was evaluated by counting the number of glochids on the cactus stem before and after processing on the system. Results: The developed system performed efficiently and effectively under the pre-studied operating conditions except for the angle of cactus flipping slide. The new system had a glochids removal ratio of 94.1% without damaging the cactus skin. Considering the original number (approximately 30-60) of glochids, the remaining number was low (1-4), and most of them were found at the side edge of the cactus stem. This system can remove glochids from 360 cactus stems in 1 hr regardless of cactus size. Conclusions: The performance of the new system in glochids removal without damaging cactus skin is superior to any other existing device (i.e., brush type, rubber-friction type, and agitation type). The system is expected to be applied in cactus (O. humifusa) processing facilities.