• KIEAE Journal
• /
• v.15 no.4
• /
• pp.105-110
• /
• 2015

Purpose: This study aimed at finding the optimal input variables of the artificial neural network-based predictive model for the optimal controls of the indoor temperature environment. By applying the optimal input variables to the predictive model, the required time for restoring the current indoor temperature during the setback period to the normal setpoint temperature can be more precisely calculated for the cooling season. The precise prediction results will support the advanced operation of the cooling system to condition the indoor temperature comfortably in a more energy-efficient manner. Method: Two major steps employing the numerical computer simulation method were conducted for developing an ANN model and finding the optimal input variables. In the first process, the initial ANN model was intuitively determined to have input neurons that seemed to have a relationship with the output neuron. The second process was conducted for finding the statistical relationship between the initial input variables and output variable. Result: Based on the statistical analysis, the optimal input variables were determined.

• Proceedings of the IEEK Conference
• /
• 2002.06a
• /
• pp.367-370
• /
• 2002

In this paper, we propose a new node architecture LonWorh control Network for home network system environmint using power line communications. Using conventional Lon Work technology is a many disputable points for home network. LonWork network system needs high-cost development equipment. Moreover, conventional Lon Work system can not implement high-grade algorithms and variety application operation. because of the limitation of processing ability in Neuron chip. For that reason, the proposed structure is applicable to low-cost and more complex applications which are impossible in home network using conventional Lonworks structure. The proposed structure is implemented with some hardware and かone software for power line home network. The physical layer and the MAC layer of the LonTalk protocol within ton Work are implemented in hardware, which decreases the development costs communication processor. The upper of link layer of the LonTalk protocol is implemented with software, which decreases the development costs of software and increases the flexibility of tile system and increases the extension of the system. We verified the commercial feasibility of the proposed system through the power line tests with the existing LonWorks network in home network. As a result, it is concluded that the proposed architecture provides increasing flexibility and decreasing cost of the system.

• Tunnel and Underground Space
• /
• v.30 no.1
• /
• pp.63-75
• /
• 2020

When tunnel is excavated via drilling and blasting, the excessive overbreak is the primary cause of personal or equipment safety hazards and increasing the cost of the tunnel operation owing to additional ground supports such as shotcrete. The practical management of overbreak is extremely difficult due to the complex causative mechanism of it. The study examines the relationship between rock mass characteristics (unsupported face condition, uniaxial compressive strength, face weathering and alteration, discontinuities- frequency, condition and angle between discontinuities and tunnel contour) and the depth of overbreak through using feed-forward artificial neuron networks. Then, Overbreak Resistance Factor (ORF) has been developed based on the weights of rock mass parameters to the overbreak phenomenon. Also, a new concept of tunnel overbreak management system using ORF has been suggested.

• Journal of Practical Engineering Education
• /
• v.15 no.2
• /
• pp.381-388
• /
• 2023

Through artificial neural network education using spreadsheets, non-major undergraduate students can understand the operation principle of artificial neural networks and develop their own artificial neural network software. Here, training of the operation principle of artificial neural networks starts with the generation of training data and the assignment of correct answer labels. Then, the output value calculated from the firing and activation function of the artificial neuron, the parameters of the input layer, hidden layer, and output layer is learned. Finally, learning the process of calculating the error between the correct label of each initially defined training data and the output value calculated by the artificial neural network, and learning the process of calculating the parameters of the input layer, hidden layer, and output layer that minimize the total sum of squared errors. Training on the operation principles of artificial neural networks using a spreadsheet was conducted for undergraduate non-major students. And image training data and basic artificial neural network development results were collected. In this paper, we analyzed the results of collecting two types of training data and the corresponding artificial neural network SW with small 12-pixel images, and presented methods and execution results of using the collected training data for Orange machine learning model learning and analysis tools.

• The Journal of Korean Medicine
• /
• v.25 no.4
• /
• pp.95-107
• /
• 2004

Objective : The inappropriate or excessive apoptosis has been known to be associated with neurodegenerative disorders including intracranial hemorrhage(ICH). Paeoniae radix, in traditional Korean medicine, has played its role as blood­nourisher and yin-astringent. In the present study, the effect of Paeoniae radix on the inhibition of neurodegeneration in the brain of rats after artificial ICH and on the resulting apoptosis was investigated. Methods : 30 rats were divided into 6 equal groups ; the sham-operation group, the hemorrhage-induction group, the hemorrhage-induction with 10, 50, 100, and 200 mg/kg Paeoniae radix-treated group, respectively. Stereotactic surgery was performed and collagenase was infused to induce ICH in the region of CA1 of hippocampus of rats. The sham group took only saline infusion. For 7 days after the surgery, 4 testing groups had intraperitoneal injections of Paeoniae radix extract. The step-down inhibitory avoidance task, measurement of neurodegeneration degree in the CA1 region of the hippocampus, and detection of caspase-3 and newly generated cells in the dentate gyrus were done after animal sacrifice. Results : Rats receiving Paeoniae radix extract showed increased latency time in the inhibitory avoidance task. The extension of neuron-deprived areas in the CA1 region was significantly suppressed in the Paeonia treated groups. Also expressions of caspase-3 in the CA1 region and cortex were significantly inhibited in the Paeonia treated groups. The cell proliferation was evaluated by means of BrdU methods and proved to be decreased in the Paeonia treated groups. Conclusion : These results suggest that Paeoniae radix has potential to suppress short-tenn memory loss after devastating neurologic accidents. Also it was proved that Paeoniae radix has a neuroprotective effect and alleviates central nervous complications following intracerebral hemorrhage. Furthermore, it may imply that this medicinal plant can be widely used for vascular dementia and other neurodegenerative disorders.

• Korean Journal of Clinical Laboratory Science
• /
• v.48 no.1
• /
• pp.41-48
• /
• 2016

Motor evoked potential of spinal surgery is known to cause damage due to the movement path of the continuous scan operation and surgery can be performed with minimized disability after surgery. However, if it is not at all formed at the wave motion evoked potential can occur during surgery and, in some cases the size of the waveform to be measured is very small and intermittent. In this case, the surgery cannot provide information about whether there is neurological damage. Increased intensity of the wave-induced motion of the dislocation does not occur if it appears in a very small amplitude stimulus, but changing the inspection area that electrical stimulation of the waveform changes could not be found. However, stimulation of a wide area in the cerebral cortex was found to occur with a waveform in the patients who underwent examination. Through this study, we propose a useful motor evoked potential test. From November to December 2015 three spine surgery patients visited Samsung Medical Center as neurosurgery patients with omission discomfort, gait disturbance, and no symptom of strength before surgery. In spine surgery patients with motor grade weakness, when motor evoked potential waveform has not been measured, in examination of the site of electrical stimulation of the cerebral cortex from entering the C3+C5/C4+C6 or C3+C1/C4+C2 if by the activity of more motor neuron unit, it was found that the waveform is better formed.

• Journal of Korea Multimedia Society
• /
• v.4 no.3
• /
• pp.257-269
• /
• 2001

Because of increasing number of integrated circuit, there is many tries of making chip of neural network and some chip is exit. but this is not prefer because YLSI technology can't support so large hardware. So imitation of whole system of neural network is more prefer. There is common procedure in signal processing as in the neural network and pattern recognition. That is multiplication of large amount of signal and reading LUT. This is identical with some operation of MLP, and need iterative and large amount of calculation, so if we make this part with hardware, overall system's velocity will be improved. So in this paper, we design neutral network, not neuron which can be used to many other fields. We realize this part by following separated bits addition method, and it can be appled in the real time parallel process processing.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.7 no.5
• /
• pp.34-45
• /
• 1997

In this paper, we present hardware implemcntation of self-organizing feature map (SOFM) neural networkwith constant adaptation gain and binary reinforcement function on FPGA. Whereas a tnme-varyingadaptation gain is used in the conventional SOFM, the proposed SOFM has a time-invariant adaptationgain and adds a binary reinforcement function in order to compensate for the lowered abilityof SOFM due to the constant adaptation gain. Since the proposed algorithm has no multiplication operation.it is much easier to implement than the original SOFM. Since a unit neuron is composed of 1adde $r_tracter and 2 adders, its structure is simple, and thus the number of neurons fabricated onFPGA is expected to he large. In addition, a few control signal: ;:rp sufficient for controlling !he neurons.Experimental results show that each componeni ot thi inipiemented neural network operates correctlyand the whole system also works well.stem also works well.

• Journal of Practical Engineering Education
• /
• v.13 no.2
• /
• pp.233-242
• /
• 2021

In this paper, as a liberal arts course for non-majors, we proposed a supervised learning artificial neural network parameter optimization method and a basic education method for activation function to design a basic artificial neural network subject curriculum. For this, a method of finding a parameter optimization solution in a spreadsheet without programming was applied. Through this training method, you can focus on the basic principles of artificial neural network operation and implementation. And, it is possible to increase the interest and educational effect of non-majors through the visualized data of the spreadsheet. The proposed contents consisted of artificial neurons with sigmoid and ReLU activation functions, supervised learning data generation, supervised learning artificial neural network configuration and parameter optimization, supervised learning artificial neural network implementation and performance analysis using spreadsheets, and education satisfaction analysis. In this paper, considering the optimization of negative parameters for the sigmoid neural network and the ReLU neuron artificial neural network, we propose a training method for the four performance analysis results on the parameter optimization of the artificial neural network, and conduct a training satisfaction analysis.

• Journal of Ginseng Research
• /
• v.48 no.1
• /
• pp.52-58
• /
• 2024

Background: Skeletal muscle denervation leads to motor neuron degeneration, which in turn reduces muscle fiber volumes. Recent studies have revealed that apoptosis plays a role in regulating denervation-associated pathologic muscle wasting. Korean red ginseng (KRG) has various biological activities and is currently widely consumed as a medicinal product worldwide. Among them, ginseng has protective effects against muscle atrophy in in vivo and in vitro. However, the effects of KRG on denervation-induced muscle damage have not been fully elucidated. Methods: We induced skeletal muscle atrophy in mice by dissecting the sciatic nerves, administered KRG, and then analyzed the muscles. KRG was administered to the mice once daily for 3 weeks at 100 and 400 mg/kg/day doses after operation. Results: KRG treatment significantly increased skeletal muscle weight and tibialis anterior (TA) muscle fiber volume in injured areas and reduced histological alterations in TA muscle. In addition, KRG treatment reduced denervation-induced apoptotic changes in TA muscle. KRG attenuated p53/Bax/cytochrome c/Caspase 3 signaling induced by nerve injury in a dose-dependent manner. Also, KRG decreases protein kinase B/mammalian target of rapamycin pathway, reducing restorative myogenesis. Conclusion: Thus, KRG has potential protective role against denervation-induced muscle atrophy. The effect of KRG treatment was accompanied by reduced levels of mitochondria-associated apoptosis.