Estimation of the Temperatures in an Experimental Infrared Heated Greenhouse Using Neural Network Models

Estimation of the Temperatures in an Experimental Infrared Heated Greenhouse Using Neural Network Models

Angeliki Kavga, Vassilis Kappatos
DOI: 10.4018/jaeis.2013040102
OnDemand:
(Individual Articles)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

A high quality greenhouse control demands continuous measurements of indoor and outdoor greenhouse conditions. In this work, neural network models were used for reducing the cost and the time of temperature measurements, estimating two of the most important parameters in the operation of the greenhouse, namely the inside air temperature and the cover temperature of greenhouse. An extensive experimental investigation was carried out in an infrared heated greenhouse, using the experimental data to train and validate the neural network models. The modelling results show that the estimated temperatures have been in good agreement with the experimental data with accuracy ranging from 95.64% to 97.67%.
Article Preview
Top

Introduction

Greenhouses are increasingly used as a modern way of agricultural production. The main purpose of greenhouses is to improve the environmental conditions in which the plants are grown, achieving better quality and more quantity of the production. The most significant cost component in greenhouse operation is the energy consumption. Therefore, the reduction of energy consumption for heating purposes constitutes a task of crucial importance for greenhouse operators throughout the world (Bot, 2001; Tiwari, 2003).

Conventional greenhouse heating systems are based either on circulation of hot water through a piping system or on the direct use of air heaters (Van de Braak, 1988; Teitel et al., 1999). The above methods, in order to achieve the required plant temperature, have to heat the internal greenhouse air to the same or even to a slightly higher temperature than the value targeted for the plants. This practice results in increased heat losses due to convection and radiation through the cover, and also due to the leakages through openings, caused by unavoidable construction defects.

To reduce energy consumption, some straightforward measures can be applied. They include, insulation improvement by means of double glazing (Gupta & Chandra, 2002), insulation of side walls (Singh & Tiwari, 2002; Gupta & Chandra, 2002), introduction of thermal screens (Kittas et al., 2003; Ghosal & Tiwari, 2004), use of different types of cover materials, use of zigzag covering that restricts transmission losses (Swinkels et al., 2001) etc. Although, the above methods do result to energy savings, they do not question the basic premise of heating the entire greenhouse environment and then letting the plants gain energy from it.

An alternative method for reducing energy consumption in greenhouse heating could emerge by using Infrared Radiation (IR). The main advantage of IR heating is the direct delivery of thermal energy from the source to the canopy, thus eliminating the need to increase the inside air temperature in order to deliver the necessary heat by convection. As a result, cover and inside air may ideally remain at significantly lower temperatures than the target value for the plants, and the heat losses are significantly reduced (Nelson, 2003). However the use of infrared radiation for greenhouse heating has been so far scarcely investigated. Few works (Itagi & Takahashi, 1978; Blom & Ingratta, 1981; Rotz & Heins, 1982) have been published in the early 80’s, investigating the suitability of low intensity IR for greenhouse heating. The investigations are mainly experimental and provide a preliminary proof of the IR concept. Energy savings ranging from 33% to 41% are reported (Blom & Ingratta, 1981) by using an infrared heating system, compared to the conventional heating method. Current trends (Kavga et al., 2009; 2012) indicate that IR is worth reconsidering as a reasonable alternative to conventional forced air or pipe heating.

Complete Article List

Search this Journal:
Reset
Volume 15: 1 Issue (2024): Forthcoming, Available for Pre-Order
Volume 14: 1 Issue (2023)
Volume 13: 2 Issues (2022): 1 Released, 1 Forthcoming
Volume 12: 4 Issues (2021)
Volume 11: 4 Issues (2020)
Volume 10: 4 Issues (2019)
Volume 9: 4 Issues (2018)
Volume 8: 4 Issues (2017)
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2015)
Volume 5: 4 Issues (2014)
Volume 4: 4 Issues (2013)
Volume 3: 2 Issues (2012)
Volume 2: 2 Issues (2011)
Volume 1: 2 Issues (2010)
View Complete Journal Contents Listing