Korea's facility farming is steadily developing in that it has the advantage of being able to maintain the appropriate growth temperature for crops and can prepare for climate change and natural disasters. Through this, the facility cultivation area is also increasing. In addition, production per un...
Korea's facility farming is steadily developing in that it has the advantage of being able to maintain the appropriate growth temperature for crops and can prepare for climate change and natural disasters. Through this, the facility cultivation area is also increasing. In addition, production per unit area is also increasing due to improvements in crop cultivation technology. However, Korea's institutional horticulture facilities are inferior to those of advanced horticultural countries such as the Netherlands. Although more than 90% of them are outdated facilities that have been around for more than 10 years, various cutting-edge greenhouses have recently been distributed in Korea, and among them, semi-enclosed greenhouses are the most popular. A semi-closed greenhouse is a greenhouse that creates a suitable cultivation environment for crops by combining the incoming outside air with the internal air existing in an enclosed space. Due to these characteristics, semi-closed greenhouses require an in-house cooling and heating system to maintain an appropriate growth temperature for crops. Large, high-tech greenhouses have the advantage of being easy to automate and improve productivity, but a lot of heating and cooling energy is invested to maintain the optimal growth temperature for crops and a uniform temperature inside the greenhouse. Considering the increasing cost and depletion of fossil fuels such as oil used for facility cooling and heating, it is essential to calculate the cooling and heating load of greenhouses to ensure management stability for facility cultivation farms.
To calculate the cooling and heating load of a greenhouse, various factors such as weather data, crops, and covering materials must be considered. Among them, crop energy exchange is essential for calculating the cooling and heating load of greenhouses. To date, research on crop energy exchange has been conducted in the form of a regression equation, but there are limitations in measuring and considering the factors of the target equation. To overcome these limitations, this study attempted to develop a crop energy exchange model using time series sensor data commonly available in greenhouses. Bi-LSTM was used to design the model, and evapotranspiration and greenhouse internal meteorological data were collected for tomatoes grown in a two-unit semi-closed greenhouse in the Jeonju area and used for model learning. The crop energy exchange model using artificial intelligence was learned as the optimal model and then applied to building energy simulation.
This study attempted to calculate the appropriate heating and cooling capacity through dynamic energy load analysis for a semi-closed greenhouse. and attempted to dynamically analyze the energy exchange and load of the greenhouse using building energy simulation, and TRNSYS (Ver. 18, SEL, USA) was used among several commercial programs. The model design was divided into stages, including greenhouse structure and crop energy exchange. To design the model, actual measurements of the greenhouse shape, investigation of internal facility operating conditions such as heat pumps, ventilation windows, and thermal curtains, and field experiments were conducted. In addition, the evapotranspiration rate of crops and the temperature inside the greenhouse were measured and used as verification data for the model. As a result of verifying the energy exchange component of the crop using the evapotranspiration rate of a single crop inside the greenhouse, it showed high statistical indices with MAE 5.58 and RMSE 8.56. Through this, it was determined that the energy exchange component of the crops designed in this study was suitable for application to the energy exchange model of the entire greenhouse, and the methodology of this study can be applied when designing an energy exchange model for any crops grown inside the greenhouse. It is believed that there is. In addition, as a result of verifying the overall energy exchange model using the temperature inside the greenhouse, high statistical indicators were shown with a coefficient of determination of 0.94 and a degree of agreement of 0.98, confirming that the target model was designed appropriately for calculating energy load. To implement crop energy exchange in a greenhouse, a dynamic energy exchange model was designed using the analyzed values using an artificial intelligence LSTM model. Verification was conducted through field experiments in the Jeonju area, the target greenhouse. In addition, a 10-year cooling and heating load analysis in the Jeonju area was conducted using a verified BES model, and the appropriate cooling and heating load capacity of the target greenhouse was presented. The target area, Jeonju, was analyzed to have a maximum cooling load of 196,876 kJ/hr and a maximum heating load of 422,435 kJ/hr over the past 10 years.
Korea's facility farming is steadily developing in that it has the advantage of being able to maintain the appropriate growth temperature for crops and can prepare for climate change and natural disasters. Through this, the facility cultivation area is also increasing. In addition, production per unit area is also increasing due to improvements in crop cultivation technology. However, Korea's institutional horticulture facilities are inferior to those of advanced horticultural countries such as the Netherlands. Although more than 90% of them are outdated facilities that have been around for more than 10 years, various cutting-edge greenhouses have recently been distributed in Korea, and among them, semi-enclosed greenhouses are the most popular. A semi-closed greenhouse is a greenhouse that creates a suitable cultivation environment for crops by combining the incoming outside air with the internal air existing in an enclosed space. Due to these characteristics, semi-closed greenhouses require an in-house cooling and heating system to maintain an appropriate growth temperature for crops. Large, high-tech greenhouses have the advantage of being easy to automate and improve productivity, but a lot of heating and cooling energy is invested to maintain the optimal growth temperature for crops and a uniform temperature inside the greenhouse. Considering the increasing cost and depletion of fossil fuels such as oil used for facility cooling and heating, it is essential to calculate the cooling and heating load of greenhouses to ensure management stability for facility cultivation farms.
To calculate the cooling and heating load of a greenhouse, various factors such as weather data, crops, and covering materials must be considered. Among them, crop energy exchange is essential for calculating the cooling and heating load of greenhouses. To date, research on crop energy exchange has been conducted in the form of a regression equation, but there are limitations in measuring and considering the factors of the target equation. To overcome these limitations, this study attempted to develop a crop energy exchange model using time series sensor data commonly available in greenhouses. Bi-LSTM was used to design the model, and evapotranspiration and greenhouse internal meteorological data were collected for tomatoes grown in a two-unit semi-closed greenhouse in the Jeonju area and used for model learning. The crop energy exchange model using artificial intelligence was learned as the optimal model and then applied to building energy simulation.
This study attempted to calculate the appropriate heating and cooling capacity through dynamic energy load analysis for a semi-closed greenhouse. and attempted to dynamically analyze the energy exchange and load of the greenhouse using building energy simulation, and TRNSYS (Ver. 18, SEL, USA) was used among several commercial programs. The model design was divided into stages, including greenhouse structure and crop energy exchange. To design the model, actual measurements of the greenhouse shape, investigation of internal facility operating conditions such as heat pumps, ventilation windows, and thermal curtains, and field experiments were conducted. In addition, the evapotranspiration rate of crops and the temperature inside the greenhouse were measured and used as verification data for the model. As a result of verifying the energy exchange component of the crop using the evapotranspiration rate of a single crop inside the greenhouse, it showed high statistical indices with MAE 5.58 and RMSE 8.56. Through this, it was determined that the energy exchange component of the crops designed in this study was suitable for application to the energy exchange model of the entire greenhouse, and the methodology of this study can be applied when designing an energy exchange model for any crops grown inside the greenhouse. It is believed that there is. In addition, as a result of verifying the overall energy exchange model using the temperature inside the greenhouse, high statistical indicators were shown with a coefficient of determination of 0.94 and a degree of agreement of 0.98, confirming that the target model was designed appropriately for calculating energy load. To implement crop energy exchange in a greenhouse, a dynamic energy exchange model was designed using the analyzed values using an artificial intelligence LSTM model. Verification was conducted through field experiments in the Jeonju area, the target greenhouse. In addition, a 10-year cooling and heating load analysis in the Jeonju area was conducted using a verified BES model, and the appropriate cooling and heating load capacity of the target greenhouse was presented. The target area, Jeonju, was analyzed to have a maximum cooling load of 196,876 kJ/hr and a maximum heating load of 422,435 kJ/hr over the past 10 years.
주제어
#인공지능 온실 스마트팜 동적에너지 작물모델링
※ AI-Helper는 부적절한 답변을 할 수 있습니다.