Curve Number 및 Convolution Neural Network를 이용한 유출모형의 적용성 평가

송철민,이광현,Song, Chul Min,Lee, Kwang Hyun 응용생태공학회 2020 Ecology and resilient infrastructure Vol.7 No.2

Despite the various artificial neural networks that have been developed, most of the discharge models in previous studies have been developed using deep neural networks. This study aimed to develop a discharge model using a convolution neural network (CNN), which was used to solve classification problems. Furthermore, the applicability of CNN was evaluated. The photographs (pictures or images) for input data to CNN could not clearly show the characteristics of the study area as well as precipitation. Hence, the model employed in this study had to use numerical images. To solve the problem, the CN of NRCS was used to generate images as input data for the model. The generated images showed a good possibility of applicability as input data. Moreover, a new application of CN, which had been used only for discharge prediction, was proposed in this study. As a result of CNN training, the model was trained and generalized stably. Comparison between the actual and predicted values had an R² of 0.79, which was relatively high. The model showed good performance in terms of the Pearson correlation coefficient (0.84), the Nash-Sutcliffe efficiency (NSE) (0.63), and the root mean square error (24.54 ㎥/s). 본 연구는 유출모형 연구를 위해 주로 사용되었던 DNN에서 벗어나, 다양한 신경망을 이용하여 유출모형을 개발하고 모형의 적합성을 나타내고자 하였다. 이를 위해 분류문제에만 사용되었던 CNN을 활용하였는데, 본 모형의 입력자료로 일반적으로 CNN에서 사용하는 사진을 이용할 수 없으며, 연구의 특성상 유역조건 및 강우 등의 영향이 반영된 수치적(numerical) 이미지(image)를 사용해야 하는 난해점이 있다. 이를 해결하고자 NRCS의 CN을 사용하여 이미지를 생성했으며, CNN 모형의 입력자료로 충분히 활용 가능함을 나타냈다. 이에 더하여, 유출 추정을 위해서만 사용되어왔던 CN의 새로운 용도를 제시할 수 있었다. 모형의 학습 및 검정 결과, 전반적으로 안정적으로 모형의 학습 및 일반화가 이루어졌으며, 관측값과 산정값간의 관계를 나타내는 R²는 0.79로 비교적 높은 값이 나타났다. 또한, 모형의 평가결과는 Pearson 상관계수, NSE, 및 RMSE 등이 각각 0.84, 0.65 및 24.54 ㎥/s으로 나타나, 전반적으로 양호한 모형의 산정성능을 보인것으로 나타났다.

독단적이지만 합리적이며 변칙적이지만 체계적인 관개의 복잡계

송철민,송석호 한국농공학회 2024 韓國農工學會誌 : 전원과 자원 Vol.66 No.1

Convolution Neural Network 기반의 유출량 모의를 위한 자료구축 방법

송철민 한국농공학회 2022 韓國農工學會誌 : 전원과 자원 Vol.64 No.3

척추기저동맥부전환자를 천마구등음(天麻鉤藤飮)으로 치료한 치험 1례

송철민,장통영,임은경,이윤재,이정섭,신선호,이인,문병순,정현애,Song, Cheol-Min,Chang, Tong-Young,Rhim, Eun-Kyung,Lee, Yun-Jae,Lee, Jung-Sub,Shin, Sun-Ho,Lee, In,Moon, Byung-Soon,Jeong, Hyun-Ae 대한한방내과학회 2005 大韓韓方內科學會誌 Vol.26 No.3

Vertebrobasilar insuffciency(VBI) is also refered to as vertebrobasilar transient ischemic attack(V-B TIA). Vertebrobasilar insufficiency is often undiagnosed or misdiagnosed due to complexity of the symptoms. Transcranial doppler(TCD) can be used to possible to record the blood flow velocities in the cerebral arteries through the intact cranium. Transcranial doppler and magnetic resonance angiography(MRA) are both noninvasive techniques that can be used in vertebrobasilar insufficiency, but transcranial doppler is more sensitive with respect to evaluating stenotic lesion of the arteries than magnetic resonance angiographv. Diagnostic values of transcranial doppler were assessed in patients with vertebrobasilar insufficiency. A case of vertebrobasilar insufficiency is here reported. A 62-year-old man with dizzness, nausea and tremor because of vertebrobasilar insufficiency was admitted at Wonkwang Oriental Medicine Hospital. He was treated with the herbal medicine Chunmagudung-um and with acupunture. Results showed improvement of symptoms and favorable transcranial doppler readings. The process used in this successful treatment is here described.

농촌 소유역에서의 지표수와 지하수의 영양물질 농도 비교

송철민,김진수,오광영,권성일 한국수자원학회 2007 한국수자원학회 학술대회 Vol.2007 No.1

하천변 양배추 밭에서의 영양물질의 거동

송철민,김진수,장훈,Song, Chul-Min,Kim, Jin-Soo,Jang, Hoon 한국농공학회 2010 한국농공학회논문집 Vol.52 No.3

This study was conducted to investigate the dynamics of nutrients such as total nitrogen (TN), nitrate nitrogen ($NO_3$-N) total phosphorous (TP), and phosphate phosphorous ($PO_4$-P) in outflow from a cabbage farmland in a mixed land-use watershed. The TN concentrations in groundwater showed twice peaks in late July 2006 and late March 2007 (3.8, 4.7 mg/L, respectively), when it rained shortly after fertilizer application, indicating that nitrogen leaching is greatly influenced by fertilization and rainfall. The mean concentrations of TN and $NO_3$-N in surface water were not significantly higher than those in groundwater, while the mean concentrations of TP and $PO_4$-P in surface water were significantly (p < 0.05) were higher than those in groundwater. The TN concentrations in groundwater were generally higher than those in surface water during fertilization and early growing season due to the effect of fertilization, but vice versa in the other periods. In contrast, the TP concentrations in groundwater were always lower than those in surface water due to the sorption of particulate phosphorous by soil. The ratio of TN load in baseflow to that in total TN load (39 %) was much greater than the TP ratio (7 %), suggesting that baseflow contribute to nitrogen export. Therefore, proper fertilization management should be taken to reduce nitrogen load through baseflow.

‘코로나19’ 방송보도의 과장표현 양상에 대한 탐색적 연구

송철민,이지화,이서현 한국언론정보학회 2023 한국언론정보학보 Vol.118 No.-

This study has intended to examine the type of exaggerated expressions used in reports broadcasting on ‘COVID-19’ in order to ascertain whether the media is properly fulfilling its role as a crisis communicator. To achieve this, 9 broadcasting companies, including groundwave broadcasting, comprehensive programming channels, and news channels, which have the obligation of providing disaster broadcasts pursuant to Article 40 of the <Basic Act on Broadcasting and Communication Development>. The analysis period was 2 years and 3 months from January 20, 2020, when the first COVID-19 patient occurred in Korea, to March 31, 2022. As drift analysis, the speech space was classified into headlines, reporter’s comments, and news reports, and the exaggerated expressions were classified into maximization, augmentation, reduction, overgeneralization, and contrast. The result of analysis displayed 570 cases of exaggerated expressions among 1,035 cases that were analyzed, which equates to 55.1%. In view of this among each speech space, the ratio of exaggerated expressions to the total number of sentences was the highest in the reporter’s comments followed by news reports. In headlines, the ratio of using maximization was the highest, followed by augmentation, contrast, and reduction, and overgeneralization was not found. In the reporter’s comments, augmentation was the most frequently used, followed by maximization, contrast, overgeneralization and reduction. In news reports, the ratio of using augmentation was the highest, followed by maximization, contrast, overgeneralization, and reduction. Through research results, it was found that many exaggerated expressions were used in reports on infectious disease like idioms. Furthermore, it was confirmed that although the implementation of reporting guidelines on infectious diseases is recommended, it was found that many exaggerated expressions that are idiomatically used are not included in the reporting guidelines.

분포형 CN 기반 토지피복별 유출가중치를 이용한 오염부하량 능형회귀모형 개발

송철민,김진수 한국농공학회 2018 한국농공학회논문집 Vol.60 No.1

The purpose of this study was to develop a ridge regression (RR) model to estimate BOD and TP load using runoff weighted value. The concept of runoff weighted value, based on distributed curve-number (CN), was introduced to reflect the impact of land covers on runoff. The estimated runoff depths by distributed CN were closer to the observed values than those by area weighted mean CN. The RR is a technique used when the data suffers from multicollinearity. The RR model was developed for five flow duration intervals with the independent variables of daily runoff discharge of seven land covers and dependent variables of daily pollutant load. The RR model was applied to Heuk river watershed, a subwatershed of the Han river watershed. The variance inflation factors of the RR model decreased to the value less than 10. The RR model showed a good performance with Nash-Sutcliffe efficiency (NSE) of 0.73 and 0.87, and Pearson correlation coefficient of 0.88 and 0.93 for BOD and TP, respectively. The results suggest that the methods used in the study can be applied to estimate pollutant load of different land cover watersheds using limited data.

농업지역에서의 BOD 및 TP 부하량 예측을 위한 수문학적 이미지와 Convolution Neural Network의 적용성 평가

송철민 ( Chul Min Song ),김진수 ( Jin Soo Kim ) 한국농공학회 2020 한국농공학회 학술대회초록집 Vol.2020 No.-

본 연구는 오염부하량 예측을 위해 1차원 입력자료를 사용한 기존의 ANN (Artificial Neural Network) 모형에서 벗어나, 주로 분류문제 해결에 사용되어 왔던 2차원 입력자료를 사용하는 CNN (Convolution Neural Network) 모형을 이용하여 오염부하량 예측에 적용하고 모형의 성능을 평가하였다. CNN 모형의 개발을 위해서는 유역조건이나 수문학적 조건과 같은 오염부하량의 영향인자가 반영된 수치적 이미지가 요구되는데, 선행연구에서 CNN 모형의 입력자료로 사용되어 왔던 일반적인 사진자료는 오염부하량 영향인자의 특성을 나타내지 못해 사용할 수 없다. 이를 위해, 본 연구에서는 수문학적 이미지(hydrological image)를 CNN 모형의 입력자료로 활용하였다. 수문학적 이미지는 토지이용, 토양도 및 강우 현상을 반영할 수 있는 이미지로, SCS-CN 방법론을 적용한 수치적 이미지이며, 임의의 유역에 대해 2차원 공간에서 무차원의 수문학적 속성을 갖는 grid의 집합으로 정의할 수 있다. 이를 활용하여 본 연구에서는 전반적으로 안정적인 CNN 모형의 학습결과와 일반화를 이끌어낼 수 있었다. 실측값과 예측값 간의 관계를 나타내는 R2는 BOD 부하량 모형이 0.88, TP 부하량 모형이 0.75으로 나타났으며, 모형의 성능을 판단하기 위한 기준으로 Pearson 상관계수, NSE 및 RMSE를 등을 검토한 결과, BOD 부하량 모형은 0.94, 0.83 및 1,240.95kg/day로, TP 부하량 모형은 0.87, 0.79 및 80.48kg/day로 보여, 전반적으로 CNN 모형의 성능이 우수한 것으로 나타났다. 본 연구의 방법론을 통해 오염부하량 예측 분야에 CNN 모형의 사용가능성이 충분한 것으로 판단하였으며, 향후 원격탐사 또는 미계측 유역에서의 오염부하량추정 등의 다양한 분야에 적용 가능할 것으로 생각된다.

분산이질 설계변경관리 시스템의 객체지향 모델링

송철민(Cheol Min Song),전병철(Bingzhe Quan),윤청(Cheong Youn) 한국정보과학회 1998 한국정보과학회 학술발표논문집 Vol.25 No.1A

현재 클라이언트/서버 구조는 가장 활발히 연구되고 응용되고 있는 소프트웨어 개발 구조로서, 이러한 구조를 가지는 시스템을 구축하는데 있어 최근에는 분산이질 환경이 자리를 잡아가고 있다. 또한 최근 기업들간에 제품 경쟁력을 유지하기위한 방안으로서 정보의 관리는 중요한 자산으로 인식되고 있으며, 이러한 정보를 분산시키고 공유하는 기술이 관심의 초점이 되고 있다. 본 고에서는 OMT를 기반으로한 객체지향 모델링 기법 그리고 CORBA를 기반으로 한 분산객체기술을 바탕으로 분산이질 설계변경관리 시스템을 위한 모델링 방법을 제시하며, 이를 클라이언트/서버 구조를 가진 실제 응용 프로그램을 분석, 설계하는 방안을 제시하고 있다.