본 연구는 딸기 온실 내부의 방대한 환경인자를 활용하여 판별분석을 실시하고 딸기의 재배 베드 단수에 따른 온실 내부의 환경인자를 분석함으로 써, 딸기 분야에서 계측된 데이터의 활용성을 높이기 위한 기초자료로 활용할 목적으로 수행하였다. 그 결과는 다음과 같다. 환경인자별(온도, 습도, 이산화탄소 농도, EC 및 pH) 동질성 검정의 유의확률은 각각 0.0001, 8.2788E-38, 4.3310E-85, 1.3001E-16 및 0.0001로서 설정한 유의수준 0.05보다 낮게 나타났다. 그리고 분석결과 판별함수식은 F(x)1 = –60.5664 -0.1339×Temperature –0.0087×Humidity +0.0018×CO2 +0.1014×EC +8.3860×pH, F(x)2 = –12.4928 +0.1963×Temperature –0.0024×Humidity +0.0254×CO2 –0.0187×EC –0.3651×pH로 도출되었다. 판별함수식의 정확도는 대상 온실 A (81.1%) 및 B (96.1%)보다 대상 온실 C (100.0%)에서 높은 것으로 나타났다. 예측 가능한 대상 온실별(A, B 및 C) 분류함수는 각각 – 1836.7035 – 2.8733×Temperature + 0.1355×Humidity + 0.4186×CO2 + 7.4351×EC + 484.5901×pH, – 1710.8369 – 2.7701×Temperature + 0.1550×Humidity + 0.3937×CO2 + 7.2482×EC + 468.1477×pH, – 2291.7125 - 3.9756×Temperature + 0.0723× Humidity + 0.4177×CO2 + 8.1961×EC + 546.8476×pH로 나타났다. 특히 판별함수식을 근거로 환경인자별 새로운 측정값이나 자료를 입력하였을 때, 특정 그룹으로 분류가 가능함으로써 데이터의 특징을 파악할 수 있다. 이러한 환경인자는 식별을 용이하게 함으로써 환경인자 측정값의 활용도를 높여주는 방법이라고 판단된다.

This study conducted the discriminant analysis using significant environmental factors inside the strawberry greenhouse. The objective of this study was to analyze the environmental factors inside the greenhouse according to the number of beds in strawberry cultivation and to use it as basic data to increase the usability of the measured data in the strawberry field. The results showed that the significance probabilities of the homogeneity test for each environmental factor such as temperature, humidity, carbon dioxide concentration, EC and pH were 0.0001, 8.2788E-38, 4.3310E-85, 1.3001E-16, and 0.0001, respectively, which were lower than the significance level of 0.05. As a result of the analysis, the discriminant function formula was derived as F(x)1 = –60.5664 -0.1339×Temperature –0.0087×Humidity +0.0018×CO2 +0.1014×EC +8.3860×pH and F(x)2 = –12.4928 +0.1963×Temperature –0.0024×Humidity +0.0254×CO2 –0.0187×EC – 0.3651×pH. The accuracy of the discriminant function was higher in target greenhouse C (100.0%) than in target greenhouses A (81.1%)and B (96.1%). The predictable classification functions for each target greenhouse (A, B and C) were – 1836.7035 – 2.8733×Temperature + 0.1355×Humidity + 0.4186×CO2 + 7.4351×EC + 484.5901×pH, – 1710.8369 – 2.7701×Temperature + 0.1550×Humidity + 0.3937×CO2 + 7.2482×EC + 468.1477×pH, – 2291.7125 - 3.9756×Temperature + 0.0723×Humidity + 0.4177×CO2 + 8.1961×EC + 546.8476×pH, respectively. Specifically a new measured value or data for each environmental factor is input based on the discriminant function formula, it is possible to classify the data into a specific group, thereby identifying the characteristics of the data. This study revealed the environmental factors are a method of increasing the utilization of the new environmental factor measurement values by facilitating identification.

초록
Abstract
서론
재료 및 방법
    1. 실험대상 온실 및 작물
    2. 작물 재배
    3. 데이터 수집 방법
    4. 분석 방법
결과 및 고찰
    1. 통계 분석결과
References

• 김나은(경상국립대학교 바이오시스템공학과(스마트팜연구소) 대학원생) | Na-Eun Kim (Department of Bio-Systems Engineering, Gyeongsang National University (Institute of Smart Farm), Jinju, 52828, Korea)
• 김현태(경상국립대학교 바이오시스템공학과(스마트팜연구소) 교수) | Hyeon-Tae Kim (Department of Bio-Systems Engineering, Gyeongsang National University (Institute of Smart Farm), Jinju, 52828, Korea)
• 김대현(강원대학교 바이오시스템공학과 교수) | Dae-Hyun Kim (Department of Bio-Systems Engineering, Kangwon National University, ChunCheon, 24341, Korea)
• 윤용철(경상국립대학교 지역시스템공학과(스마트팜연구소) 교수) | Yong-Cheol Yoon (Department of Agricultural Engineering, Gyeonsang National University (Institute of Smart Farm), Jinju 52828, Korea) Correspondence author