DEVELOPMENT OF FARM POND OPERATIONAL MODEL FOR IRRIGATION SCHEDULING OF HORTICULTURAL CROPS
Loading...
Date
2010
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
IARI, DIVISION OF AGRICULTURAL ENGINEERING
Abstract
Water is a one of the most vital and important resources on the planet earth. In the
present situation of increasing human population, there is an increasing stress on available
water resources. To meet the demand of growing population food production has to grow at
the rate of growing population, which will lead to increase in agricultural water demand and
therefore there is ever increasing need for irrigation water for better agricultural outputs.
Therefore under these circumstances water resource management is of great importance.
Regulation of operating policies of storage reservoir is vital component of water resource
management. Therefore, there is a need to find out better operating policies of storage
reservoirs mainly at the farm level in the form of farm ponds. The operating policies for on
farm reservoir ensure judicious use of water at field level. Small reservoirs failed to serve the
purpose of judicious water management due to faulty operating policies, which leads to poor
distribution of water in space and time. In on farm reservoir the parameters like demand,
supply and storage were highly uncertain and instantaneous. Therefore to develop operating
rules of farm pond the soft computing techniques like fuzzy logic (FL) and Artificial Neural
Network (ANN) can be employed effectively, which takes care of uncertainty and unreliability
at the field level. So the present study was carried out to derive operational model for a farm
pond of 3000 m3
capacity at Center for protected cultivation technology (CPCT), Indian
Agricultural Research Institute, New Delhi, India. The farm pond was the important source of
irrigation water of the farm of the area 10 ha. The Neuro-Fuzzy approach was used to develop
the operational model and to derive operational rules for proper irrigation scheduling of the
horticultural crops grown at CPCT. Based upon the inputs like crop water requirement,
evaporation losses and farm pond inflow the Neuro-Fuzzy model predicting outflow of the
reservoir was developed. Sugeno type Fuzzy inference system (FIS) was developed. The
operating rules which were of the form “If-Then” were also developed using Adaptive NeuroFuzzy
Inference System (ANFIS). Out of twenty seven rules only four best rules were
selected.
The developed operational model using Neuro-Fuzzy technique was validated
statistically with new sets of input-output data set. The result of validation revealed that the
developed Neuro-Fuzzy model was able to predict outflow with high accuracy. Coefficient of
59
determination (R2
) between observed and predicted value was found to be 0.98. The model
efficiency was also found to be very high i.e. 0.97. These results showed that the model has
high accuracy and predictability. The linguistic form of operating rules of the developed soft
computing Neuro-fuzzy model would be easy for the user to perceive the process. The
operating policies of the farm pond would lead to sustainable water management at field level.
Description
T-8447
Keywords
cotton, planting, bacteria, exudates, transgenics, vegetative propagation, biological development, proteins, application methods, sowing