Chan Jung Park

Monitoring-based temporal prediction of power entities in smart grid cities

This paper develops prediction models for groundwater and wind speed in smart grid cities, where the gradual penetration of electric vehicles keeps increasing load in its power system. As a major consumption source and a renewable energy source, water management systems and wind power generators make operation plans for their facilities based on temporal predictions of target objects. The massive data of day-by-day groundwater level readings and day-by-day wind speed records, which have been accumulated for about 10 years in Jeju City, are fed to 3-layer neural networks for modeling the change pattern. The trained neural network yields an accurate short-term prediction for groundwater level, with its maximum error less than 4.76 %. In addition, for wind speed predictions, 40 % of forecast errors are smaller than 5 %, in spite of its error spikes reaching 37 %. Those models are provided to the electric vehicle charging system to shift the peak load to less loaded intervals.

Design of a Performance Analyzer for Electric Vehicle Taxi Systems

This paper designs a performance analysis framework for electric vehicle taxis, aiming at promoting their wide deployment. Consisting of an event tracker, a stream handler, object interfaces, and strategy integrator, the analysis procedure can measure the performance of a dispatch and relocation strategy in terms of dispatch latency, customer waiting time, and the number of daily fast charging operations. Each pick-up and drop-off record from the actual call taxi system is associated with the corresponding taxi and charger object. It can host a new dispatch strategy to test and revise, while a specific road network and a future demand prediction model can be incorporated for better accuracy. This framework finds out that most battery charging can be done using slow chargers through the out-of-service intervals under the control of an intelligent coordinator for the fleet of member taxis, avoiding the significant increase in power load brought by fast charging operations.

Improvement of Jeju National Museum service using mobile RFID

Recently, various types of the museum services using RFID technology have been developed as RFID technology evolves. However, most of them are yet limited to transfer the simple information of exhibits or manage the exhibits of museums warehouse. With reflection on the island features of Jeju National Museum, we have developed a system, which can improve the service quality of the museum by providing the disruptive services using mobile RFID technology. The system we developed accumulates the information such as the number of viewing an exhibit, viewers age, and viewers gender, etc., whenever a viewer with an RFID reader connected to a PDA reads a tag on an exhibit. This information is directly stored in a database, which can provide an efficient search service, to bring better service quality of the museum. We have implemented the application program for the viewers pattern analysis based on the data retrieved from the database. Using this program, we can get the movement paths of the viewers. Also, we propose an approach to bring real-time customized interaction between a viewer and a curator to answer an intriguing question of a viewer on the spot where an exhibit is located. We expect that our approach proposed will be a disruptive model widely used for most museums because it can improve their services in spite of reducing the cost of their operation and management.

The Relative Importance of Weather Factors and the Predictions About the Groundwater Level in Jeju

The research related to the groundwater level in Jeju has focused on the amount of rainfall and its hydrogeological characteristics so far. However, since the sensor technology has been used in many areas to measure the various types of natural phenomena recently, it allows us to perform advanced analysis on the groundwater in Jeju. In this paper, we consider wind speed, evaporation, temperature, humidity, and rainfall as the factors that can have influences on the groundwater level. We describe how the factors can affect the groundwater and how artificial neural networks can predict the groundwater level. We perform multiple regression and hierarchical linear model analysis. And then, we calculate the relative importance of the weather factors. We use data accumulated from 2003 to 2009 and perform data-oriented analysis rather than theoretical analysis. We divide Jeju region into four basins such as north (Jeju), south (Seogwipo), west (Gosan), and east (Sungsan).

A Novel Approach to Provide Mobile RFID-Based Services Combined with Data Mining Technique

In advance of RFID technology, various types of services based on mobile RFID technology have been developed. However, there still exists limitation on RFID-based services such as transferring the simple information of exhibits at exhibition halls or managing the exhibits stored in warehouses. In order to improve the service qualities at exhibition centers, showrooms, or museums, we have developed a system, which can provide the disruptive services using mobile RFID technology. Our system accumulates the information such as how many people visit a specific exhibit, what the demographic information the visitors have, and what kinds of items the visitors see, and so on whenever visitors read a tag on an exhibit with an RFID reader connected to a mobile device. The accumulated information is stored in a database to provide an efficient search service and to bring a better service to the visitors. We have developed an application program for analyzing the visitors’ data retrieval pattern from the database. Using this program, our system can get the information about the movement paths of the visitors. Also, by adopting data mining technology, we also propose a novel approach to enhance RFID-based services so that visitors can have a customized real-time question-answering service from assistants in a certain exhibition hall or museum whenever they are intrigued to know about exhibits.