水基聚合物包膜控释肥料养分释放模型研究

By research on nutrient release model of coated Controlled Release Fertilizers (CRFs), the nutrient release mechanism and influence factors can be understood, providing basis for optimize release parameters, also nutrient release profile can be predicted to make it better match with nutrient absorbed by crop, providing the theoretical foundation for the efficient development and application of CRFs. Existing models were developed under the assumption that the radius of coated fertilizer granules remains the same during the release period. These models apply to liposoluble polymer coated CRFs only, not suitable for waterborne polymer coated CRFs, whose granule radius increases significantly during the release period. However, waterborne polymer coated CRFs represent one of the important development trend in the future. Taking waterborne polymer coated CRFs as the research object, the release mechanism is clarified by determining diffusion coefficient dynamically through Electronic Speckle Pattern Interferometry (ESPI), and the mechanism model is established based on its release mechanism. In most case, the nutrient release profile determined for populations of CRFs is significantly different from the one predicted based on mechanism model. This is due to a large variation in properties of populations CRFs. Meanwhile, an empirical model is constructed in combination with an artificial neural network using step-scan Fourier Transform Infrared Photoacoustic Spectroscopy (FTIR-PAS). The implementation of the project not only provides theoretical basis for optimizing release parameters, thereby improving the efficiency of research and development of CRFs, but also create a new idea to accurately predict nutrient release profiles of coated CRFs.

研究包膜肥料的养分释放模型，可了解其控释机制和影响因素，为优化控释参数提供依据；也可预测包膜肥料的养分释放特征，使其更好地和作物的养分吸收相匹配，为高效研制和应用包膜肥料提供基础。现有的养分释放理论模型都是针对释放过程中颗粒半径不变的脂溶性聚合物包膜肥料，不适于颗粒半径明显增大的水基聚合物包膜肥料，而水基聚合物包膜肥料代表了包膜肥料未来的重要发展方向之一。本项目以水基聚合物包膜肥料为研究对象，利用电子散斑干涉技术动态测定包膜扩散系数的变化规律，阐明包膜肥料的控释机制，并推导理论模型。应用理论模型预测群体包膜肥料养分释放时，由于群体中包膜肥料的性质变异大致使预测效果一般都很差。本项目同时利用红外光声光谱的逐层扫描功能，结合人工神经网络构建养分释放的经验模型。本项目的实施，不仅为优化释放参数提供理论依据，进而提高包膜肥料研发效率,也为准确预测包膜肥料养分释放特征开创新的思路。

养分释放模型是包膜控释肥料研制和应用的理论基础。现有的养分释放模型都是针对释放过程中颗粒半径不变的脂溶性聚合物包膜肥料，不适用于释放过程中颗粒半径明显增大的水基聚合物包膜肥料，而水基聚合物包膜肥料代表了包膜肥料未来的重要发展方向之一。本项目以水基聚合物包膜肥料为研究对象，构建了适用于溶胀型水基聚合物包膜肥料养分释放的机理模型。肥料的养分累计释放率由颗粒半径表示，而颗粒半径随释放时间线性增加。肥料颗粒半径增大的速率，即溶胀速率，正比于包膜的亲水性，反比于包膜厚度。模型对颗粒半径有高的预测准确性（R 2 = 0.99，RMSE=0.06 mm），也可准确预测单颗肥料的累计养分释放率（R 2 = 0.86，RMSE = 2.28%）。然而模型用于群体包膜肥料养分释放预测时表现出了很大的差异，这主要是因为群体包膜肥料中包含了颗粒间很大的变异。包膜肥料几乎都是以群体的形式应用的，项目也构建了可准确预测群体包膜肥料养分释放的经验模型。模型考虑了颗粒间的差异，以包膜肥料的光声光谱信息，包衣量以及颗粒半径的分布频率为输入，以群体肥料养分累计释放曲线为输出，构建了偏最小二乘-支持向量机模型。模型可准确并有效地预测群体包膜肥料的累计养分释放曲线，即适用‘S’型累计释放曲线(R2=0.998，RMSE=1.15%)，也适用‘L’型累计释放曲线(R2=0.999 ，RMSE=0.54%)。通过本项目的实施，构建了水基聚合物包膜肥料养分释放的理论模型，阐明了包膜肥料的相关参数对养分释放的影响，为优化控释参数提供理论依据；也构建了可预测群体水基聚合物包膜肥料养分释放的经验模型，为准确高效预测包膜肥料养分释放特征开创新的思路，也为包膜肥料的工业化生产以及田间应用提供了指导思想。