Design Optimization of Proportional Plus Derivative Band Parameters Used in Greenhouse Ventilation by Response Surface Methodology

• Published in: Weon'ye gwahag gi'sulji Vol. 38; no. 2; pp. 187 - 200
• Main Authors: Jung, Dae-Hyun, Kim, Hak-Jin, Kim, Joon Yong, Lee, Taek-Sung, Park, Soo Hyun
• Format: Journal Article
• Language: English
• Published: 한국원예학회HST 01.01.2020; 한국원예학회
• Subjects: 농학; control coefficient; automated ventilation; experimental design; temperature control; greenhouse management
• ISSN: 1226-8763; 2465-8588; 2465-8588; 1226-8763
• DOI: 10.7235/HORT.20200018

Ventilation is one of the most important factors in greenhouse cultivation because the inside temperatureof greenhouses rises rapidly in warm climates, and this high temperature may be detrimental to plantgrowth. The ventilation system in multi-span greenhouses is often controlled by a proportionalalgorithm. The relationship between temperature changes in the greenhouse and ventilation is notlinear. Conventional ventilation strategies mostly employ a proportional band (P-band) control, whichcontains various settings related to solar radiation, outside temperature, and wind velocity. In practice,these settings may have a relationship with the set ventilation temperature. However, it is difficultto find optimized settings because they must be changed according to the greenhouse location, shape,and local climate changes. In this study, we attempted to optimize the P-band parameters for outsidetemperature, solar radiation, and wind speed affecting greenhouse ventilation by surface responseanalysis. Surface response analysis was used to confirm the quadratic response of the relationship betweenX and Y, which was designed using the central composite design. A total of 32 experimental factorswere designed and the results were analyzed by empirical tests. In addition, derivative calculationswere added to respond to rapid changes of temperature in the greenhouse and used for ventilationparameters. The results showed that optimized parameters were successfully applied to temperaturecontrol and we obtained better performance than without optimized parameters, with a root meansquared error of ± 1.25°C. It is anticipated that the study results will be useful in providing thenumerical guidelines for the influence factors of the ventilation control settings. The study resultswill be available immediately for the influence factors of the ventilation control settings. KCI Citation Count: 0