http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
채수원(Soowon Chae),배상무(Sangmu Bae),남유진(Yujin Nam) 대한설비공학회 2021 대한설비공학회 학술발표대회논문집 Vol.2021 No.6
It is difficult to construct a building load model because standards of occupant and equipment pattern is not clearly defined. Therefore, in this study, building load model is suggested based on occupant and equipment patterns for standard building model. The building load model was contructed by energy simulation and calculated heating and cooling energy demand considering design factors such as architectural conditions and energy consumption. As the result, the cooling energy demand of first floor was 13% lower and the heating energy demand was 4% higher compared with middle floor. In addition, the cooling and heating energy demand of top floor were 7% and 12% higher, respectively.
인공지능 기반 태양광열 시스템의 최적 운전제어 알고리즘 개발에 관한 연구
채수원(Soowon Chae),배상무(Samgmu Bae),남유진(Yujin Nam) 대한설비공학회 2022 대한설비공학회 학술발표대회논문집 Vol.2022 No.6
Recently, in Korea, the development of new and renewable energy utilization technology is actively progressing in order to realize carbon neutrality and to cope with the problem of energy supply and demand. Among them, photovoltaic-thermal (PVT) is attracting attention as a technology that can simultaneously produce heat and electricity to recover waste heat and increase power production efficiency. However, the existing PVT system has a problem in that waste heat recovery and power generation efficiency is low because the circulating water is controlled by ON-OFF or PID (Proportional integral derivative control) controller based on the set temperature. Therefore, this study developed an optimal control algorithm to increase the waste heat recovery and power generation efficiency of PVT. The optimal control algorithm was determined through the collection and power generation trends according to the flow rate of circulating water, and was developed using machine learning. As a result of comparing the normalized data with the development model, the maximumerror was confirmed as Cv(RMSE) was 23.3 and the correlation coefficient was 0.81, indicating stable control. On the other hand, energy saving was evaluated on the basis of heat-collecting production compared to power consumption, and it was confirmed that 59.1% was improved compared to the notation model. Therefore, the optimal algorithm developed in this study is expected to contribute to the reduction of pump power and energy consumption of hot water supply as it enables efficient operation of PVT.
인공지능 기반 태양광열 시스템의 최적 운전제어 알고리즘 개발에 관한 연구
채수원(Soowon Chae),배상무(Samgmu Bae),남유진(Yujin Nam) 대한설비공학회 2022 대한설비공학회 학술발표대회논문집 Vol.2022 No.6
Recently, in Korea, the development of new and renewable energy utilization technology is actively progressing in order to realize carbon neutrality and to cope with the problem of energy supply and demand. Among them, photovoltaic-thermal (PVT) is attracting attention as a technology that can simultaneously produce heat and electricity to recover waste heat and increase power production efficiency. However, the existing PVT system has a problem in that waste heat recovery and power generation efficiency is low because the circulating water is controlled by ON-OFF or PID (Proportional integral derivative control) controller based on the set temperature. Therefore, this study developed an optimal control algorithm to increase the waste heat recovery and power generation efficiency of PVT. The optimal control algorithm was determined through the collection and power generation trends according to the flow rate of circulating water, and was developed using machine learning. As a result of comparing the normalized data with the development model, the maximumerror was confirmed as Cv(RMSE) was 23.3 and the correlation coefficient was 0.81, indicating stable control. On the other hand, energy saving was evaluated on the basis of heat-collecting production compared to power consumption, and it was confirmed that 59.1% was improved compared to the notation model. Therefore, the optimal algorithm developed in this study is expected to contribute to the reduction of pump power and energy consumption of hot water supply as it enables efficient operation of PVT.
실제 건물 적용을 고려한 태양광열 및 공기열원 융복합 시스템의 동적 해석
채수원(Chae Soowon),배상무(Bae Sangmu),남유진(Nam Yujin) 한국태양에너지학회 2022 한국태양에너지학회 논문집 Vol.42 No.2
As regulations on renewable energy are strengthened and zero-energy buildings are becoming mandatory, research on the commercialization of renewable energy for heating and cooling buildings is accelerating. However, geothermal, photovoltaic, and solar heat are mostly applied as single systems, and there are few cases in which the disadvantages of individual systems are overcome or where the advantages are maximized. The purpose of this study is to develop an analysis model that can respond stably to heating and cooling loads in buildings and to analyze system performance. To stabilize the control of cooling, heating, and hot water supply, the operation method is divided into four cycles. In seasons when cooling and heating loads are significantly decreased, heating and cooling are performed directly by the heat pump without using storage tanks. In addition, a case study using late-night power was conducted to investigate efficient electricity use. Without late-night power, the heat pump coefficient of performance (COP) during the heating period was 2.5 and the monthly average heat exchange rate (HER) was 240 kWh; meanwhile, the COP during the cooling period was 3.99 and the HER was 880 kWh. The energy self-sufficiency rate during the heating period was 2.02 times higher than the cooling period. However, as a result of using late-night power, the electricity rate and COP during heating decreased by 2.5% and 2.04%, respectively, and the COP during cooling increased by 0.1%. This study presents basic data for the implementation of renewable energy systems and the design of predictive models.
인공신경망 기반 태양광열 및 공기열원 히트펌프 시스템의 최적제어에 관한 연구: 부하측 순환수 유량제어
채수원(Soowon Chae),배상무(Samgmu Bae),남유진(Yujin Nam) 대한설비공학회 2022 대한설비공학회 학술발표대회논문집 Vol.2022 No.11
This study developed a predictive model to predict the optimal flow rate based on artificial neural network to maximize the performance of the air heat source heat pump system and analyzed the heating and cooling performance of the heat pump using dynamic simulation. The target for controlling the optimal flow rate in the system is the flow rate of water flow rate on the load side at heat pump, and learning data were prepared by selecting the flow rate with the highest system COP as the optimal value according to the outdoor air conditions and heating and cooling load. The coefficient of variation of the root mean square error of the prediction model was 0.69 %, and the normalized mean bias error was 0.19%, which satisfies the criteria recommended by ASHRAE. To evaluate the applicability of the verified predictive model, a dynamic simulation model was constructed based on the actual building system. The target building is equipped with an integrated photovoltaic-thermal and air source heat pump system to implement a zero energy building. As a result of analyzing the heating and cooling performance of the optimal model and the conventional model using a dynamic simulation model, it was confirmed that the average annual coefficient of performance was improved by 3.1 %. The results indicate that the new control method is more energy efficient and has an advantage in implementing zero energy building.
민정국(JeongGuk Min),채수원(SooWon Chae),김권희(KwonHee Kim) 한국자동차공학회 2004 한국자동차공학회 춘 추계 학술대회 논문집 Vol.- No.-
Finite element analyses for vehicle crash usually take a lot of time and efforts for model modifications and computations. Alternatives to finite element models are 3D frame models in which the characteristic of joints represent plastic deformation resistances between two rigid bodies. In a commercial code MADYMO deformation resistances are embodied either by sets of point restraints or by sets of kinematic joints. In spite of versatility, point restraints are dit1icult to set up and the joints are simple but prone to overestimation of deformation resistances. This paper suggests a 3D frame model with combined joint restraint and card an restraint to simulate crash modes with simultaneous bending and axial crush.
실제 건물적용을 위한 태양광열-공기열원 히트펌프 융복합 시스템의 실증실험
배상무(Sangmu Bae),오진환(Jinhwan Oh),채수원(Soowon Chae),남유진(Yujin Nam) 대한설비공학회 2022 대한설비공학회 학술발표대회논문집 Vol.2022 No.6
Renewable energy systems are being actively applied to realize zero energy building(ZEB). However, a single technology centered on solar-thermal, photovoltaic and ground source heat pump is mostly being used. This single technology was difficult to stably respond to the energy demand of buildings. Therefore, in this study, the integrated system combining photovoltaic-thermal and air source heat pump was proposed for overcoming the disadvantages of single technology and realizing the ZEB. In order to verify the accuracy performance of integrated system, the real-scale experiment plant and monitoring system was constructed. The heating performance of integrated system was analyzed based on monitoring data. As the result, the average coefficient of performance of the integrated system was calculated to be 2.50. The performance of the integrated system could be improve when the contribution of the PVT module is higher.