Abstract: By adopting the principle of system analysis and technique of mathematical modeling to knowledge expression system for wheat management, a dynamic knowledge model with temporal and spatial characters for digital wheat management was developed by extracting and formulating the fundamental relationships and quantitative algorithms of wheat growth indices and management criteria to variety types, ecological environments and production levels. With incorporation of the soft component characteristics, a component-based knowledge model system for digital wheat management was established on the platforms of Visual C++ and Visual Basic. The system realized two major functions as design of pre-sowing cultural plan and prediction of regulation index dynamics. The module of cultural plan design included the sub-models for determination of target yield and quality, suitable variety, sowing date, population density and sowing rate, fertilization strategy and water management, and the module of regulation index dynamics included the sub-models for suitable development stages, spike differentiation stages, dynamics of growth indices, source-sink indices and nutrient indices. Simulation studies on the knowledge model with the data sets of different eco-sites, cultivars, soil types and so on, and comparative field experiments indicated a good performance of the knowledge model system in decision-making and wide applicability. The present knowledge model system overcomes the shortcomings as specific site limitation and low quantification of traditional wheat cultivation patterns and expert systems, and lays a foundation for facilitating the digitization of wheat management.

Key words: Wheat, Knowledge model, Cultural plan, Regulation index, Model development, System testing