植物生長過程中受到許多環境因素影響，如：溫度、濕度、光照、風速及降雨量等。近年來由於氣候變遷導致極端氣候發生頻率上升，如高溫、乾旱或暴雨，維持農作物正常生長與品質面臨更多困難，同時也突顯了農民監控作物週遭即時環境數據的重要性。文心蘭是台灣極為重要的切花產業，溫度對於其切花之產量與品質影響甚鉅。因此，本研究以文心蘭為例，研究如何以有效方式監控及估算其生長網室之溫度。本研究決定裝設網室感應器時，檢驗五種安裝方式與不同安裝數目，並評選最佳感測器安裝方式與支數。
所採用空間推論方式為克利金法，該法因為可最大使用空間資訊使其推論更為精確，目前廣泛使用於各種環境資訊之空間推論。在環境溫度監控部分，現有相關文獻多於網室內安裝環境感測器，且以節能省電或通訊傳輸之角度來決定感測器安裝位置；較少研究分析如何安裝能有效獲得最準確的環境資訊。因此，本研究比較了五種感測器安裝位置的方法：極端值法、最多數據資訊法(熵)、標準差法、均勻布設法和隨機布設法，以其溫度估算準確度決定最佳之感測器安裝位置選取方法。
研究區域為台中市新社區的文心蘭網室，網室面積長寬約為180公尺與30公尺。以均勻分布的方式於網室內設置44支氣候感測器，共4行11列。為了比較不同季節是否影響研究結果，研究週期分別選擇了2020年4月23日至4月29日(春)以及 11月23日至11月29日(冬)。而在感測器布設支數選擇中，實際可用支數為42支，本研究亦測試當感應器減少至30與20支進行分析。
研究成果顯示，根據溫度估算準確度(均方根誤差)，感測器布設位置使用極端值法之選取方法結果最佳，誤差範圍約0到0.5℃。換言之，農民若有預算規劃於溫室內安裝感測器，可使用極端值法選取感測器之安裝位置，此研究成果亦可推廣於其他溫室作物生長環境的感測器安裝位置選取。

Climate conditions such as temperature, humidity, and light play important roles in agriculture. In recent years, climate change increases the frequency of extreme weather, such as high temperatures, droughts, or heavy rain affect agriculture production and quality. A sensor system monitoring real-time microclimate can help farmers protect crops from extreme weather and then maintain the production quality. We used oncidium as our study object to observe how to estimate the shade net house temperature efficiently. The study tested five sensor placement methods to get the best installation approach and the suitable number of sensors. We aim to determine the optimal sensor placement of the oncidium shade net house used in this study and the optimal sensor placement method.
We used Kriging for spatial interpolation because this approach is widely used in most environmental information-related studies. The study gap of existing sensor studies is that most sensor systems were designed based on energy-saving or communication transmission. Therefore, we compare the performance of these five sensor placement methods, which are the extreme value-based method, information entropy-based method, standard deviation-based method, uniform distribution method, and random distribution method. Furthermore, the accuracy of temperature prediction is used to determine the optimal sensor placement method, and then the optimal sensor placement method was used to decide the optimal sensor placement of the shade net house.
This study was located on an oncidium shade net house in Xinshe District, Taichung City. The size of the shade net house is 180 meters x 30 meters and 44 sensors are installed in the shade net house by uniform distribution (4 rows and 11 columns). To compare the study results in different seasons, multiple study periods were selected from April 23rd – 29th (spring) and November 23rd – 29th (winter). In the selection of the number of sensors to be installed, the actual number of available sensors is 42. We chose 20 and 30 sensors based on five sensor placement methods to analyze the accuracy of the different numbers of sensors.
The results show that the extreme value-based method tends to obtain the optimal result based on the accuracy of temperature prediction (RMSE value), and its range of error is about 0 ~ 0.5℃. Based on our result, farmers can install the sensors and monitor the temperature in the shade net house accurately by using the extreme value-based method. The study results can also be applied to the sensor placement of other shade net houses.

Appelhans, T., Mwangomo, E., Hardy, D.R., Hemp, A., Nauss, T. (2015) Evaluating machine learning approaches for the interpolation of monthly air temperature at Mt. Kilimanjaro, Tanzania. Spat. Stat., 14, 91–113.
Cressie, N. (1990) The origins of kriging. Math Geol 22, 239–252.
C. Gascuel-Odoux, P. Boivin (1994). Variability of variograms and spatial estimates due to soil sampling: a case study. Geoderma, 62, 165-182.
C. R. Bojacá, R. Gil, S. Gómez, A. Cooman, E. Schrevens (2009). Analysis of Greenhouse Air Temperature Distribution Using Geostatistical Methods. Transactions of the ASABE. 52(3): 957-968.
Franklin, T.B. (1955) Climates in miniature. A study of micro-climate and environment.
Greg Simpson, Yi Hwa Wu (2014). Accuracy and Effort of Interpolation and Sampling: Can GIS Help Lower Field Costs? ISPRS Int. J. Geo-Inf., 1317-1333.
Jermin Jeaunita, T. C., Sarasvathi, V. (2019). Fault tolerant sensor node placement for IoT based large scale automated greenhouse system. International Journal of Computing and Digital Systems, 8(2), 189– 197.
Lu, G.Y., Wong, D.W. (2008) An adaptive inverse-distance weighting spatial interpolation technique. Comput. Geosci., 34, 1044–1055.
Li, C., Lu, Z., Ma, T., Zhu, X. (2009) A simple kriging method incorporating multiscale measurements in geochemical survey. J. Geochem. Explor., 101, 147–154.
Liang, X., Schilling, K., Zhang, Y.-K., Jones, C. (2016) Co-Kriging Estimation of Nitrate-Nitrogen Loads in an Agricultural River. Water Resour. Manag., 30, 1771–1784.
Lata, S. H. K. V. (2017). Selection of sensor number and locations in intelligent greenhouse. In 2017 3rd International Conference on Condition Assessment Techniques in Electrical Systems (CATCON) (p. 5). India: Rupnagar.
Lee, S. Y., Lee, I. B., Yeo, U. H., Kim, R. W., & Kim, J. G. (2019b). Optimal sensor placement for monitoring and controlling greenhouse internal environments. Biosystems Engineering, 188, 190– 206.
Matheron, G. (1963) Principles of geostatistics. Econ. Geol., 58, 1246–1266.
Mcbratney, A.B., Webster, R. (1986) Choosing functions for semi-variograms of soil properties and fitting them to sampling estimates. J. Soil Sci., 37, 617–639.
Malla, G. (2008) Climate change and its impact on Nepalese agriculture. J. Agric. Environ., 9, 62–71.
Mercer, L.D., Szpiro, A.A., Sheppard, L., Lindström, J., Adar, S.D., Allen, R.W., Avol, E.L., Oron, A.P., Larson, T., Liu, L.J., et al. (2011) Comparing universal kriging and land-use regression for predicting concentrations of gaseous oxides of nitrogen (NOx) for the Multi-Ethnic Study of Atherosclerosis and Air Pollution (MESA Air). Atmos. Environ., 45, 4412–4420.
M. P. Fanti, G. Faraut, J. Lesage and M. Roccotelli, (2018) An Integrated Framework for Binary Sensor Placement and Inhabitants Location Tracking. IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 48, no. 1, pp. 154-160, Jan.
Oliver, M.A., Webster, R. (1990) Kriging: A method of interpolation for geographical information systems. Int. J. Geogr. Inf. Syst., 4, 313–332.
Oktavia, E., Mustika, I.W. (2016) Inverse distance weighting and kriging spatial interpolation for data center thermal monitoring. In Proceedings of the 2016 1st International Conference on Information Technology, Information Systems and Electrical Engineering (ICITISEE), Yogyakarta, Indonesia, 23–24 August, pp. 69–74
Pahuja, R., Verma, H.K., Uddin, M. (2013) A wireless sensor network for greenhouse climate control. IEEE Pervasive Comput., 12, 49–58.
Pahuja, R., Verma, H. K., & Uddin, M. (2017). An intelligent wireless sensor and actuator network system for greenhouse microenvironment control and assessment. Journal of Biosystems Engineering, 42(1), 23–43.
Pathak, T.B., Maskey, M.L., Dahlberg, J.A., Kearns, F., Bali, K.M., Zaccaria, D. (2018) Climate change trends and impacts on California Agriculture: A detailed review. Agronomy, 8, 1–27.
Roscoe, J.T. (1975) Fundamental Research Statistics for the Behavioural Sciences, 2nd edition. New York: Holt Rinehart & Winston
van Beers, W., Kleijnen, J. (2003) Kriging for interpolation in random simulation. Journal of the Operational Research Society, 54(3), 255–262.
van Zoest, V., Osei, F.B., Hoek, G., Stein, A. (2020) Spatio-temporal regression kriging for modelling urban NO2 concentrations. Int. J. Geogr. Inf. Sci., 34, 851–865.
Wu Xuewen, Yan Luming. (2007) Setting parameters and choosing optimum semivariogram models of ordinary kriging interpolation—A case study of spatial interpolation to January average temperature of Fujian province. Geo-information Science，2007，9（3）:104-108
Wu, T., Li, Y. (2013) Spatial interpolation of temperature in the United States using residual kriging. Appl. Geogr., 44, 112–120.
Zhang, J., Li, X., Yang, R., Liu, Q., Zhao, L., Dou, B. (2017) An extended kriging method to interpolate near-surface soil moisture data measured by wireless sensor networks. Sensors., 17, 1390.