|M.Sc Student||Gips Elad|
|Subject||Advanced Mathematical Model for Predicting Quality of|
Wine Grapes and its use for Improving Vineyard
|Department||Department of Civil and Environmental Engineering||Supervisor||Professor Raphael Linker|
|Full Thesis text|
In the field of wine-grapevine, the use of dynamic plant models for Decision Support Systems (DSS) is still limited due to the complexity of the desired yield features, which include not only fruit quantitative properties but also, with major importance, qualitative properties. Although widely addressed in many viticultural researches, the complex relations between these different properties have not yet been implemented in a dynamic computer model.
The goal of the research was to develop such a model and a model-based Decision Support System for the wine-grapevine grower. The research was based on an existing generic dynamic plant model - STICS, which was adjusted using data from field experiments conducted at Dolev vineyards, Cv. Cabernet Sauvignon. The research included (a) model calibration using sensitivity analysis, (b) formulating methodology for development of qualitative berry elements modules, for elements that are still unmodeled, (c) formulation of a grower-friendly optimal control problem and (d) formulation of a control scheme for a decision support system, including multiple seasons static optimization and seasonal on-line control algorithms.
The results demonstrated the feasibility of the proposed approach but also revealed the limitations of the current model, i.e. its current accuracy is not sufficient and additional experimental data is required to improve the model.