• Korean Journal of Agricultural Science
• /
• v.47 no.4
• /
• pp.1029-1037
• /
• 2020

The machine learning algorithm has been widely used in water-related fields such as water resources, water management, hydrology, atmospheric science, water quality, water level prediction, weather forecasting, water discharge prediction, water quality forecasting, etc. However, water quality prediction studies based on the machine learning algorithm are limited compared to other water-related applications because of the limited water quality data. Most of the previous water quality prediction studies have predicted monthly water quality, which is useful information but not enough from a practical aspect. In this study, we predicted the dissolved oxygen (DO) using recurrent neural network with long short-term memory model recurrent neural network long-short term memory (RNN-LSTM) algorithms with hourly- and daily-datasets. Bugok Bridge in Oncheoncheon, located in Busan, where the data was collected in real time, was selected as the target for the DO prediction. The 10-month (temperature, wind speed, and relative humidity) data were used as time prediction inputs, and the 5-year (temperature, wind speed, relative humidity, and rainfall) data were used as the daily forecast inputs. Missing data were filled by linear interpolation. The prediction model was coded based on TensorFlow, an open-source library developed by Google. The performance of the RNN-LSTM algorithm for the hourly- or daily-based water quality prediction was tested and analyzed. Research results showed that the hourly data for the water quality is useful for machine learning, and the RNN-LSTM algorithm has potential to be used for hourly- or daily-based water quality forecasting.

• Journal of Environmental Science International
• /
• v.21 no.8
• /
• pp.953-963
• /
• 2012

The annual variations of the urban heat island in Busan is investigated using surface temperature data measured at 3 automatic weather stations(AWSs) for the 5 years period, 2006 to 2010. Similar to previous studies, the intensity of the urban heat island is calculated using the temperature difference between downtown(Busanjin, Dongnae) and suburb(Gijang). The maximum hourly mean urban heat island are $1.4^{\circ}C$ at Busanjin site, 2300LST and $1.6^{\circ}C$ at Dongnae site, 2100LST. It occurs more often at Dongnae than Busanjin. Also the maximum hourly mean urban heat island appears in November at both sites. The urban heat island in Busan is stronger in the nighttime than in the daytime and decreases with increasing wind speed, but it is least developed in summer. Also it partly causes the increasement of nighttime PM10 concentration.

• Journal of Environmental Health Sciences
• /
• v.14 no.1
• /
• pp.39-45
• /
• 1988

The stochastic variations were analyzed periodicity by autocorrelation, variance spectrum and Fourier series. These time series included hourly and hourly mean observations on DO, water temperature and air temperature which measured by automatic recording instrument at Guii from 1, Jan., 1986 to 23, Feb., 1986. The results of study were as follows: l. Autocorrelation coef. (lag time 120) DO($\varrho_1$= 0.9705), WT($\varrho_1$ = 0.9890), and AT($\varrho_1$ = 0.9874) were deeply related. DO and AT clearly showedr 24-hour periodicities while WT showed 23-26 hour periodicity. 2. Spectral density showed high at 24 hour in eech item and all of them showed weak peak at 12 hour. 3. The explained variance, which was a measure of the contribution of periodic function to the original time series, varied high 90.8 - 94.7%. This results showed that water qualities at Guii were affected deterministic components.

• Proceedings of the SAREK Conference
• /
• 2008.11a
• /
• pp.440-445
• /
• 2008

In this paper, new methodology is proposed to estimate the cooling load using areas of building group and predicted weather data. Only three parameters such as maximum, minimum temperature and building area are necessary to obtain hourly distribution of cooling load for the next day. The maximum and minimum temperature that are used for input parameters can be obtained from forecasted weather data. The areas of building group are used for setting several parameters that are used for estimate cooling loads. Benchmarking building(research building) is selected to validate the performance of the proposed method, and the estimated cooling loads in hourly bases are calculated and compared with the measured data for benchmarking building. The estimated results show fairly good agreement with the measured data for benchmarking building.

• Solar Energy
• /
• v.7 no.1
• /
• pp.23-29
• /
• 1987

TAP system, a kind of natural convective space heating collector, has a good heat loss by night. The aim of this paper is to induce and to study an hourly heat flow theory by response factors analysis with TAP in inside and outside insulated construction, to compare and evaluate on thermal performance an hourly natural temperature, heated room temperature and heating load in aboved-mention constructions with computer simulation. The results of the study can be summarized as follows. According that there is no TAP and with TAP, it is inside insulated construction and outside insulated construction, daily natural range of temperature each shows $12.5^{\circ}C$ and $16.7^{\circ}C$, $2.7^{\circ}C$ and $3.7^{\circ}C$, daily heated range of temperature with noramal control heating system each shows $6.6^{\circ}C$ and $12.1^{\circ}C$, $1.7^{\circ}C$ and $3.1^{\circ}C$, heating hours each show 10 hr and 7 hr, 9 hr and 4 hr and heating energy saving percentage in january 123% and 79%, 100% and 40%. Therefore, energy saving percentage shows that outside insulated construction saves about 54% in comparision with inside insulated construction.

• Proceedings of The Korean Society of Agricultural and Forest Meteorology Conference
• /
• 2016.09a
• /
• pp.118-126
• /
• 2016

• Journal of the Korean Society for Geothermal and Hydrothermal Energy
• /
• v.16 no.1
• /
• pp.17-25
• /
• 2020

Ground-coupled heat pump (GCHP) systems have become an efficient alternative to conventional cooling and heating methods due to their higher energy using efficiency. Although some experimental and simulation works related to performance analysis of GCHP systems for commercial buildings have been done, relatively little has been reported on the performance evaluation of GCHP systems for school buildings. The purpose of this simulation study is to evaluate the performance of a hypothetical GCHP system for a school building in Seoul. We collected various data of building specifications and construction materials for the building and then modeled to calculate hourly building loads with SketchuUp and TRNSYS V17. In addition, we used GLD (Ground Loop Design) V2016, a GCHP system design and simulation software, to design the GCHP system for the building and to simulate temperature of circulating water in ground heat exchanger. The variation of entering source temperature (EST) into the system was calculated with different prediction time and then each result was compared. For 20 years of prediction time, EST for baseline design (Case A) based on the hourly simulation results were outranged from the design criteria.

• Asian Journal of Atmospheric Environment
• /
• v.9 no.1
• /
• pp.12-21
• /
• 2015

Air pollution is a challenging issue in some of the large cities in developing countries. In this regard, data interpretation is one of the most important parts of air quality management. Several methods exist to analyze air quality; among these, we applied the Multilayer Perceptron (MLP) and Radial Basis Function (RBF) methods to predict the hourly air concentration of benzene in 14 districts in the municipality of Tehran. Input data were hourly temperature, wind speed and relative humidity. Both methods determined reliable results. However, the RBF neural network performance was much closer to observed benzene data than the MLP neural network. The correlation determination resulted in 0.868 for MLP and 0.907 for RBF, while the Index of Agreement (IA) was 0.889 for MLP and 0.937 for RBF. The sensitivity analysis related to the MLP neural network indicated that the temperature had the greatest effect on prediction of benzene in comparison with the wind speed and humidity in the study area. The temperature was the most significant factor in benzene production because benzene is a volatile liquid.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.40 no.12
• /
• pp.793-799
• /
• 2016

In a hybrid fuel cell system, low-temperature reforming technology, which uses waste heat as a heat source, is applied to improve system efficiency. A low temperature reformer is required to optimize geometry in low thermal conditions so that the reformer can achieve the proper methane conversion rate. This study analyzed internal temperature distributions and the reaction patterns of a reformer by considering the change of the shape factor on the limited heat supply condition. Unlike the case of a high temperature reformer, analysis showed that the reaction of a low temperature reformer takes place primarily in the high temperature region of the reactor exit. In addition, it was confirmed that the efficiency can be improved by reducing the GHSV (gas hourly space velocity) or increasing the heat transfer area in the radial direction. Through reacting characteristic analysis, according to change of the aspect ratio, it was confirmed that a low temperature reformer can improve the efficiency by increasing the heat transfer in the radial direction, rather than in the longitudinal direction.

• Smart Structures and Systems
• /
• v.3 no.3
• /
• pp.341-356
• /
• 2007

A good understanding of normal modal variability of civil structures due to varying environmental conditions such as temperature and wind is important for reliable performance of vibration-based damage detection methods. This paper addresses the quantification of wind-induced modal variability of a cable-stayed bridge making use of one-year monitoring data. In order to discriminate the wind-induced modal variability from the temperature-induced modal variability, the one-year monitoring data are divided into two sets: the first set includes the data obtained under weak wind conditions (hourly-average wind speed less than 2 m/s) during all four seasons, and the second set includes the data obtained under both weak and strong (typhoon) wind conditions during the summer only. The measured modal frequencies and temperatures of the bridge obtained from the first set of data are used to formulate temperature-frequency correlation models by means of artificial neural network technique. Before the second set of data is utilized to quantify the wind-induced modal variability, the effect of temperature on the measured modal frequencies is first eliminated by normalizing these modal frequencies to a reference temperature with the use of the temperature-frequency correlation models. Then the wind-induced modal variability is quantitatively evaluated by correlating the normalized modal frequencies for each mode with the wind speed measurement data. It is revealed that in contrast to the dependence of modal frequencies on temperature, there is no explicit correlation between the modal frequencies and wind intensity. For most of the measured modes, the modal frequencies exhibit a slightly increasing trend with the increase of wind speed in statistical sense. The relative variation of the modal frequencies arising from wind effect (with the maximum hourly-average wind speed up to 17.6 m/s) is estimated to range from 1.61% to 7.87% for the measured 8 modes of the bridge, being notably less than the modal variability caused by temperature effect.