Expert system to select a proper chemigation-injection system according to field condition

The aim of this research is to help in selecting an appropriate chemical injector in certain situations. Systems under investigation included: (1) Bypass pressure differential tank, (2) Piston injection pump, (3) With suction pipe of irrigation pump, (4) Diaphragm injection pump, (5) Separate electric centrifugal pump, (6) Venturi, and (7) Independent power injection unit developed by Lithy (2012). Evaluation of systems for selection depend on farm resources and conditions (water source, power source, labor, system pump capacity, distance from pump to field, irrigation time, irrigation system, operating pressure, required injection rate, initial cost, area, safety consideration). Results of expert system (ES) approach were validated through consulting with domain expert and references. Each chemical injector type was given a score for every resource item. The highest summation for any system, indicate its suitability for the set of conditions composed. Four actual trial cases were put under validation: Namely case (A), (B), (C), and (D). Results corroborated that using of a separate electric centrifugal pump was the most appropriate choice, with score weight of 13.25 which is about 43% above the average of all the choices in case (A), for big scale farm with electric power supply, well water source, and unlimited irrigating time. Whereas, for cases (B) and (C) the most appropriate proved to be the independent power injection unit developed by Lithy (2012) with score 14.5 and marginal advantages over other systems due to unavailable power source, limited irrigation time and high rate of chemical injection required. The choice with suction pipe of irrigation pump recorded the highest score weight of 15.25 which is about 37% above the average of all the choices in case (D) with condition of underground water reservoir and unlimited irrigation time. In conclusion the expert system proposed is valid in different cases including extreme representative situations.