ارزیابی روش‌های درون‌یابی برای پهنه‌بندی نیازهای گرمایشی و سرمایشی گلخانه‌های کشور

Introduction

The science of knowing the environmental conditions to build greenhouses in suitable climatic locations is one of the most important studies in the field of greenhouse construction and maintenance. After all, by obtaining information about environmental factors, especially temperature, it is possible to choose the most suitable place for the construction of these structures, which requires a spatial study of the studied areas. Spatial investigations require the existence of rich and integrated data, but due to many limitations it is not possible to collect information at all points; therefore, information is only collected in areas that are more sensitive or generally represent a broad area. As a result, it is never possible to study an area continuously, because to conduct such studies we need an infinite collection of data. The solution available to solve this problem is the use of geostatistical software and Geographic Information Systems (GIS) and interpolation methods. In the present study, for the first time in the study and evaluation of different methods for interpolating the heat demand index for heating and cooling greenhouses, the most suitable methods for this purpose have been introduced. The application of the results of this research can be used not only in the greenhouse industry but also in other industries related to energy and its consumption.



Materials and Methods

Heating of greenhouses is also necessary in countries with a temperate climate in order to achieve maximum production of greenhouse products in terms of quantity and quality, as well as to increase efficiency and productivity. The heating and cooling demand for a given period was extracted and used based on the degree day index. Since information from meteorological stations is represented in the form of points, interpolation will be the basic element of information generalization for use of the selected models. Through the generalization of these properties, interpolated maps will be available, which, in addition to using these drawings for individual investigation and processing of the conditions of the region, will also be the basis for heat demand calculations. There are various methods of parameter interpolation, and five widely used categories and a total of 24 methods (with their subcategories) were evaluated and reviewed in this study. These categories are Inverse Distance Weighting (IDW), Global polynomial interpolation (GPI), Local polynomial interpolation (LPI), Radial basis function (RBF) and Kriging methods.



Results and Discussion

To evaluate the best interpolation method, available data was analyzed and validated for use in zone models. This information was used in 162 stations, which had relatively complete information for the period studied (20 years). The values of these stations for the annual heating demand data of the Iran’s greenhouses with a minimum value of zero and a maximum value of 3205 (degree days) have a skewness of 0.262 and are normally distributed. The best interpolation method for cooling and heating demand data is the IDW method. Among the potencies used in this procedure, the optimized power has the highest correlation and the best results. After that, the kriging methods and the RBF are each evaluated to interpolate the data related to the cooling and heating needs of the Iran's greenhouses. Examination of the other methods examined showed that with the GPI method, results improved by increasing the power up to the fifth order, and after has been shown decrease in correlation and an increase in error. Also with the method of the RBF, all sub-methods except TPS are in the same error range and the best result relates to the CRS method. According to the results, the best result for the LPI method relates to the exponential function, while other methods are in the same error range and in the next higher order. In general, the results of all three kriging methods used in this study are close to each other in the margin of error where ordinary kriging is relatively superior to other methods. Universal and Simple Kriging are also in the next ranks respectively.

Conclusion

Nowadays, most meteorological parameters have been studied by many researchers and they have recommended different interpolation methods for each parameter. Since the nature of the heating and cooling needs of greenhouses differs from other climatic parameters and is calculated and used cumulatively on an annual basis, for the purpose of interpolation and zoning it is necessary to determine and represent an appropriate method. In this study, by examining the types of common and widely used interpolation methods, the IDW method provided the best estimate of interpolation, after which ordinary kriging methods and RBF are placed in the next ranks.