• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.4
• /
• pp.1193-1215
• /
• 2021

The measurement of the key product quality index plays an important role in improving the production efficiency and ensuring the safety of the enterprise. Since the actual working conditions and parameters will inevitably change to some extent with time, such as drift of working point, wear of equipment and temperature change, etc., these will lead to the degradation of the quality variable prediction model. To deal with this problem, the selective integrated moving windows based principal component regression (SIMV-PCR) is proposed in this study. In the algorithm of traditional moving window, only the latest local process information is used, and the global process information will not be enough. In order to make full use of the process information contained in the past windows, a set of local models with differences are selected through hypothesis testing theory. The significance levels of both T - test and χ² - test are used to judge whether there is identity between two local models. Then the models are integrated by Bayesian quality estimation to improve the accuracy of quality variable prediction. The effectiveness of the proposed adaptive soft measurement method is verified by a numerical example and a practical industrial process.

• Journal of Institute of Control, Robotics and Systems
• /
• v.18 no.11
• /
• pp.997-1004
• /
• 2012

In enterprise software projects, performance issues have become more critical during recent decades. While developing software products, many performance tests are executed in the earlier development phase against the newly added code pieces to detect possible performance regressions. In our previous research, we introduced the framework to enable automated performance anomaly detection and reduce the analysis overhead for identifying the root causes, and showed Statistical Process Control (SPC) can be successfully applied to anomaly detection. In this paper, we explain the special performance trend in which the existing anomaly detection system can hardly detect the noticeable performance change especially when a performance regression is introduced and recovered again a while later. Within the fixed number of sampling period, the fluctuation gets aggravated and the lower and upper control limit get relaxed so that sometimes the existing system hardly detect the noticeable performance change. To resolve the issue, we apply dynamically tuned sampling window size based on the performance trend, and Fuzzy theory to find an appropriate size of the moving window.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.27-27
• /
• 2017

신뢰성 있는 수개월 선행시간의 댐 유입량 예측은 가뭄 상황으로 진입하는 시점에서 효율적인 댐 운영을 위해 필수적이다. 최근 기후변화로 인한 강수량의 경년 및 계절 내 변동성이 증가됨에 따라서 기존의 과거 통계치를 이용한 댐 운영 의사결정은 많은 도전을 받고 있다. 최근 엘리뇨-남방진동(ENSO) 등의 전구기후지수와 지역수문기후와의 원격상관성을 활용하여 수개월 이후에 대한 수문조건을 통계적으로 예측하기 위한 연구가 시도되고 있다. 또한 매월 제공되는 역학적 예측모형으로부터 생산된 월단위 예측정보를 유량예측을 위한 유역모형에 활용하기 위하여 편이보정 및 상세화 기법이 개발되어 활용되고 있다. 본 연구에서는 댐 유입량 예측을 위해 SWAT 모형을 선정하였고 최장 6개월 선행 강수량 및 기온의 예측을 위해서 하이브리드 계절예측 시스템을 활용하였다. 이 시스템은 전지구역학적 예측모형의 자료를 편이보정을 거쳐 직접적으로 사용하는 단순 편이보정(Simple Bias Correction, SBC) 방법에 회귀모형을 이용하여 통계적인 방법으로 예측자료를 생산하는 전구기후지수 기반의 Climate Index Regression (CIR), 실시간 재분석자료 기반의 Observation-based Moving Window Regression (MWR-Obs), 역학적 예측모형의 예측자료 기반의 Moving Window Regression (MWR) 방법을 통합하여 사용하고 있다. 충주댐을 대상으로 우선 관측자료를 이용하여 SWAT 모형을 검 보정한 후, 관측기간에 대하여 하이브리드 시스템에 의한 예측 기상자료를 적용하여 모의된 댐 유입량과 관측 유입량과의 비교를 통해 예측성을 평가하였다. 본 연구는 다양한 기후정보를 활용하여 댐 유입량 예측에 있어서 예측성을 높이고자 시도되었으며, 도출된 결과는 향후 충주댐 운영에 유용한 정보를 제공할 수 있는 것으로 판단된다.

• Korean journal of applied entomology
• /
• v.55 no.2
• /
• pp.139-148
• /
• 2016

A seasonal outlook for crop insect pests is most valuable when it provides accurate information for timely management decisions. In this study, we investigated probable tele-connections between climatic phenomena and pest infestations in Korea using a statistical method. A rice insect pest, brown planthopper (BPH), was selected because of its migration characteristics, which fits well with the concept of our statistical modelling - utilizing a long-term, multi-regional influence of selected climatic phenomena to predict a dominant biological event at certain time and place. Variables of the seasonal climate forecast from 10 climate models were used as a predictor, and annual infestation area for BPH as a predictand in the statistical analyses. The Moving Window Regression model showed high correlation between the national infestation trends of BPH in South Korea and selected tempo-spatial climatic variables along with its sequential migration path. Overall, the statistical models developed in this study showed a promising predictability for BPH infestation in Korea, although the dynamical relationships between the infestation and selected climatic phenomena need to be further elucidated.

• Journal of Electrical Engineering and Technology
• /
• v.1 no.4
• /
• pp.466-471
• /
• 2006

This raper presents a new optimization method combining the genetic algorithm with the response surface method for the optimum design of a Brushless Direct Current motor. The method utilizes a regression function approximating an objective function and the window moving and zoom-in method so as to complement disadvantages of both the genetic algorithm and response surface method. The results verify that the proposed method is powerful and effective in reducing cogging torque by optimizing only a few decisive design factors compared with the conventional stochastic methods.

• International Journal of Control, Automation, and Systems
• /
• v.6 no.5
• /
• pp.639-650
• /
• 2008

Electricity price forecasting has become an integral part of power system operation and control. In this paper, a wavelet transform (WT) based neural network (NN) model to forecast price profile in a deregulated electricity market has been presented. The historical price data has been decomposed into wavelet domain constitutive sub series using WT and then combined with the other time domain variables to form the set of input variables for the proposed forecasting model. The behavior of the wavelet domain constitutive series has been studied based on statistical analysis. It has been observed that forecasting accuracy can be improved by the use of WT in a forecasting model. Multi-scale analysis from one to seven levels of decomposition has been performed and the empirical evidence suggests that accuracy improvement is highest at third level of decomposition. Forecasting performance of the proposed model has been compared with (i) a heuristic technique, (ii) a simulation model used by Ontario's Independent Electricity System Operator (IESO), (iii) a Multiple Linear Regression (MLR) model, (iv) NN model, (v) Auto Regressive Integrated Moving Average (ARIMA) model, (vi) Dynamic Regression (DR) model, and (vii) Transfer Function (TF) model. Forecasting results show that the performance of the proposed WT based NN model is satisfactory and it can be used by the participants to respond properly as it predicts price before closing of window for submission of initial bids.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2016.05a
• /
• pp.7-7
• /
• 2016

인도네시아 산불에 의한 연무는 동남아시아 인접한 국가들에 있어서 심각한 환경문제 중 하나이다. 국제적으로 심각한 문제를 야기하는 인도네시아의 산불은 건조기에 강수량이 적게 내리는 극심한 가뭄 조건에서 발생한다. 건조기 강수량을 모니터링 하는 것은 산불 발생 가능성을 예측하기 위해 중요하지만 산불을 사전에 예방하고 영향을 최소화하기에는 부족하다. 따라서 산불에 대한 선제적 사전예방을 위해서는 수개월의 선행예측 기간을 갖는 조기경보 시스템이 절실하다. 따라서 본 연구는 인도네시아 산불에 대한 선제적 대응을 위한 강수량 예측시스템을 개발하고 예측성을 평가하여 동남아시아 지역의 화재 연무 조기경보 시스템의 시제품(Prototype)을 개발하는데 있다. 강수량 예측을 위해서 APEC 기후센터의 계절예측정보의 활용 정도에 따라서 4가지 서로 다른 방법을 통합하여 사용하였다. 예측정보 기반의 방법들로는 대상지역의 강수량 예측을 위해서 대상 지역 상공의 계절예측 강수자료를 보정을 통해 직접적으로 사용하는 SBC (Simple Bias Correction) 방법과 대상 지역 상공의 강수 예측자료를 사용하는 대신에 지역 강수량과 높은 상관 관계를 보이는 다른 지역의 대리변수를 예측인자로 사용하는 MWR (Moving Window Regression) 방법이 있다. 또한 예측자료의 사용 없이 과거자료 기반의 기후지수(Climate Index) 중에서 지체시간을 고려하여 지역 강수량과 높은 상관관계를 갖는 경우 예측에 활용하는 관측자료 기반의 CIR (Climate Index Regression) 방법과 예측기반 MWR과 관측기반의 CIR 방법에서 선정된 예측인자를 동시에 활용하는 ITR (Integrated Time Regression) 방법이 사용되었다. 장기 강수량 예측은 보르네오 섬의 4개 지역에서 3개월 이하의 선행예측기간에 대하여 0.5 이상의 TCC (Temporal Correlation Coefficient)의 값을 보여 양호한 예측성능을 보였다. 예측된 강수량 자료는 위성기반 관측 강수량 및 관측 탄소 배출량 관계에서 결정된 강수량의 임계값과의 비교를 통해 산불발생 가능성으로 환산하였다. 개발된 조기경보 시스템은 산불 발생에 가장 취약한 해당지역의 건조기(8월~10월) 강수량을 4월부터 예측해 산불 연무에 대한 조기경보를 수행한다. 개발된 화재 연무조기경보 시스템은 지속적인 개선을 통해 현장 실효성을 높여 동남아국가 정부의 화재 및 산림관리자들에게 보급함으로써 동남아의 화재 연무로 인한 환경문제 해결에 기여할 수 있으리라 판단된다.

• Journal of Environmental Impact Assessment
• /
• v.32 no.5
• /
• pp.338-352
• /
• 2023

As of 2019, Korea's fine dust is the most severe among 38 OECD countries, and in the same year, 「the Framework on Disaster and Safety Management」 was revised to define fine dust as a social disaster. Currently, the government is working to achieve its emission reduction goals by preparing a comprehensive fine dust management plan (2022-2023) consisting of a total of five areas, 42 tasks, and 177 detailed tasks. However, it is necessary to come up with measures in consideration of the various spatial characteristics of the city, not just as a source of emission. Therefore, in this study, the shape of the city was classified using the LCZ (Local Climate Zone) classification system into 17 types by building type and land cover type in Busan, and the average annual PM10 and PM2.5 concentration were mapped using the IDW technique. In addition, Fragstats and Moving Window were used to quantify the LCZ classification system. Finally, correlation analysis and regression analysis were conducted to analyze the relationship between the LCZ classification system and PM10 and PM2.5. As a result, it was confirmed that the type of low height of the building and the type of green space with trees had a positive effect on the concentration of PM10 and PM2.5. Therefore, this study is expected to be used as basic data to establish fine dust reduction policies based on efficient spatial planning.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.11
• /
• pp.310-318
• /
• 2018

Recently, the importance of big data analysis is increasing as a large amount of data is generated by various devices connected to the Internet with the advent of Internet of Things (IoT). Especially, it is necessary to analyze various large-scale IoT streaming sensor data generated in real time and provide various services through new meaningful prediction. This paper proposes a real-time indoor PM10 concentration prediction LSTM model based on streaming data generated from IoT sensor using AWS. We also construct a real-time indoor PM10 concentration prediction service based on the proposed model. Data used in the paper is streaming data collected from the PM10 IoT sensor for 24 hours. This time series data is converted into sequence data consisting of 30 consecutive values from time series data for use as input data of LSTM. The LSTM model is learned through a sliding window process of moving to the immediately adjacent dataset. In order to improve the performance of the model, incremental learning method is applied to the streaming data collected every 24 hours. The linear regression and recurrent neural networks (RNN) models are compared to evaluate the performance of LSTM model. Experimental results show that the proposed LSTM prediction model has 700% improvement over linear regression and 140% improvement over RNN model for its performance level.

• Journal of Climate Change Research
• /
• v.4 no.4
• /
• pp.359-369
• /
• 2013

In this paper, correlations between driving parameters and $CO_2$ of light-duty vehicles have been analyzed. Three test vehicles equipped with PEMS (Portable Emission Measurement System) have been driven in real-road in urban areas of Seoul. Averaged vehicle speed, RPA(Relative Positive Acceleration) and stop ratio have been selected as main driving parameters. The analysis have been conducted in interrupted and uninterrupted road types. Averaged values in various driving conditions have been calculated with distance based moving averaging window method. The multiple linear regression method have been applied to account for correlation between driving parameters and $CO_2$ emissions. This approach has shown statistically that $CO_2$ emission per distance (g/km) have tendencies to be increased as decreased averaged vehicle speed and increased RPA and stop ratio. Compared with uninterrupted traffic, interrupted traffic have shown the lower vehicle speed and the higher RPA and stop ratio. These characteristics of driving parameters in interrupted traffic should cause the higher $CO_2$ emission per distance.