英文摘要 | Plant Modelling uses a simple and abstract model to describe, depict the process of plant growth. There are already plenty of plant models. Plant growth is a very complex interactive process between plant and outer environment elements, including light, temperature, water, soil, etc. Therefore, plant modelling is still in the state of art, and is progressing quickly. Plant growth modelling is prospective in agriculture application, and attracting more and more interest. This paper presents an open-sourced Intelligent Management Software for Greenhouse Tomato Growth. It can make production plan, yield forcasting and optimization, to help tomato growth in some aspects. The main contributions of this software include following issues: ① support database access, can manage huge amount of data in an easy way. data-oriented system design, can support multi plant model in one system. ② implement 'Logistic' plant growth model in system, apply it in tomato growth management. The curve inferred from model can fit the measured data well. The curve's parameters can be obtained automatically or ad- justed manually, to facilitate its application in greenhouse, based on others' work, integrate 3D visualization function into system, which can support 'GreenLab' plant growth model. ③ developed a genetic algorithm toolbox 'GATS' based on scientific comput- ing language SCILAB. It can use four gene coding scheme, support multi- objective optimization and parallel processing, etc. This toolbox can be used in parameter estimation and optimization. ④ open source, easy to be extended, and can run on several platforms (Win- dows, Linux/Unix). It's based on open source software, which makes it possible that farms can use it freely, very fitful with the situation of China's agriculture. In a word, by developing a software described in this thesis, we have made a fruitful attempt on plant growth model application in agriculture. |
修改评论