• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.10
• /
• pp.117-124
• /
• 2017

A Smart Grid is a system that can efficiently use energy by exchanging real-time information in both directions between a consumer and a power supplier using ICT technology on an existing power network. DR(Demand response) is an arrangement in which electricity users can sell the electricity they save to the electricity market when the price of electricity is high or the power system is crisis. In this study, we developed a power meter data transmission device and power IT system that measure the demand information in real-time using a smart meter and transmit it to a cloud server. The power meter data transmission device developed in this study uses alight sensor connected to a Raspberry Pi 3 to measure the number of blinking lamps on the KEPCO meter per unit of power, in order to provide reliable data without any measurement errors with respect to the KEPCO power data. The power measurement data transmission device uses the standard communication protocol, OpenADR 2.0b. The measured data is transmitted to the power IT system, which consists of the VEN, VTN, and calculation program, via the LTE WiFi communication network and stored in its MySQL DB. The developed power measurement data transmission device issues a power supply instruction and performs a peak reduction DR when a power system crisis occurs. The developed power meter data transmission device has the advantage of allowing the user to adjust it every 1 minute, where as the existing smart metering time is fixed at once every 15 minutes.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.2
• /
• pp.139-147
• /
• 2021

This paper introduces and implements a Sound Event Detection (SED) system based on weakly-supervised learning where only part of the data is labeled, and analyzes the effect of parameters. The SED system estimates the classes and onset/offset times of events in the acoustic signal. In order to train the model, all information on the event class and onset/offset times must be provided. Unfortunately, the onset/offset times are hard to be labeled exactly. Therefore, in the weakly-supervised task, the SED model is trained by "strongly labeled data" including the event class and activations, "weakly labeled data" including the event class, and "unlabeled data" without any label. Recently, the SED systems using the mean-teacher model are widely used for the task with several parameters. These parameters should be chosen carefully because they may affect the performance. In this paper, performance analysis was performed on parameters, such as the feature, moving average parameter, weight of the consistency cost function, ramp-up length, and maximum learning rate, using the data of DCASE 2020 Task 4. Effects and the optimal values of the parameters were discussed.