Improvement in an interpolation algorithm concerning multiple factors to predict drift distance

Research on droplet drift distance is one important part of precision agriculture, there are many factors influencing drift distance such as environment conditions and physical property of droplets group. According to previous research results based on other agricultural experts, those factors involving temperature, relative humidity of environment, velocity of wind, droplet sizes, discharge height from nozzle and droplet initial velocity play an essential role in determining how far droplet drift away. Dr Heping Zhu in Ohio State University have designed one quick method to predict droplet drift distance by assuming that each relations between those factors and drift distance is absolutely linear. However, error generating from assumption itself is obvious and it is entirely necessary to reduce any error caused by designed algorithm as much as possible. By analysis, such error could be reduced remarkably if partial improvement is made in Dr Heping Zhu's interpolation method. In this paper, relation between droplet size and drift distance is changed to be cubic because of its strong non-linear relation. Finally, we develop a software implemented with this new interpolation method, and then both of the results founded on the same but special combination of input variables are compared, and the result reveals that this new interpolation tally better with experiment data than the origin one and relative error based on those special combination of input variables are reduced obviously.