Electric Vehicle Charging Station incorporating with an Energy Management and Demand Response Technique
School of Electrical Engineering,
King Mongkut's Institute of Technology Ladkrabang (KMITL)
A study of Demand Response (DR) function and Energy management for an electrical vehicle (EV) charger is presented in this paper. The proposed technique is to prevent the electrical system within the rated power by controlling electric vehicle chargers loads to the electrical system. A power management method for limiting the charging current and charging time by using state of charge (SOC) as a priority point index is developed. The power management method is also developed as a calculation algorithm for determining the charging current limit rating for electric vehicle chargers. The simulation model is also developed for validating a DR management function of the electric vehicle charger according to Provincial Electricity Authority of Thailand (PEA) load characteristic data. The results show that the proposed DR function can manage the charging current of each electric vehicle charger appropriately. Also. the proposed technique can prevent the rated power demand of the transformer distribution. The study illustrates that this method is an effective protection for the distribution transformer and can be able to apply as a DR function to manage electrical energy more efficiently
Order Learning and Its Applications to Computer Vision
School of Electrical Engineering,
In this talk, we first discuss order learning to determine the order graph of classes, representing ranks or priorities, and classify an object instance into one of the classes. To this end, we design a pairwise comparator to categorize the relationship between two instances into one of three cases: one instance is ‘greater than,’ ‘similar to,’ or ‘smaller than" the other. Then, by comparing an input instance with reference instances and maximizing the consistency among the comparison results, the class of the input can be estimated reliably. Second, we present the deep repulsive clustering (DRC) algorithm of ordered data for effective order learning. To this end, we develop the order-identity decomposition (ORID) network to divide the information of an object instance into an order-related feature and an identity feature. Then, we group object instances into clusters according to their identity features using a repulsive term. Moreover, we estimate the rank of a test instance, by comparing it with references within the same cluster. Experimental results on facial age estimation, aesthetic score regression, and historical color image classification show that the proposed algorithm can cluster ordered data effectively and also yield excellent rank estimation performance.
Low Latency and Lightweight Video Computing in Edge Cloud Networks
Information Technology Center,
University of the Ryukyus
In this talk, I will give an overview of the edge-cloud-based “high-speed video computing technologies” in the Beyond 5G (B5G) era. In the future, the wideband and low latency data transfer function of the high-frequency band Beyond 5G will be linked with various computing resources in edge and cloud networks. The purpose is to realize low latency and lightweight video processing such as video editing and compression by using software in exceeding 10 Gbps edge cloud networks. I will mainly introduce two technologies. The first is high-speed processing technologies for real-time 8K ultra-high-definition (UHD) video. We have developed a system to store, process, and deliver uncompressed 8K UHD video in real-time using PC servers in edge cloud networks. The second is seamless video coding that takes privacy into consideration. I will talk about the basic concept. It enables video compression in edge cloud networks while the video data is scrambled. It also realizes low-delay transmission and lossless archiving in a unified architecture.