Hydrothermal liquefaction (HTL), a thermo-chemical conversion process, uses water as a reaction medium at elevated pressure and temperature, to convert biomass to renewable liquid fuel and recovers fertilizer-rich water. To assess the techno-economic screening of HTL oils from various feedstock, it is crucial to have information on molecular composition of the feed and products. There are limitations of existing analytical methods to identify and quantify all the molecules present in the bio-fuel. Therefore, there is a need to find alternate ways to quantify the molecular composition of feed and expected products. The modeling work on bio-oil is developed and validated on mathematical approach using simple analytical results like CHNO along with structural analysis of oil like Fourier-transform infrared, nuclear magnetic resonance analysis for HTL derived oil from microalgae. This mathematical framework is further extended to predict the molecular composition of oil obtained from HTL of feedstocks like mixed plastic waste, sludge, and so on. A multi-dimensional molecular matrix is developed based on the distributions of side chains, aromatic rings, and olefinic carbon on top of core molecules. The parameters of the distributions are estimated computationally using global optimization algorithm (genetic algorithm) and local optimization algorithm to predict a mixture composition that matches closely with bulk properties of the product.

中文翻译：

---

计算热液液化衍生可再生原油分子组成的数学方法

水热液化 (HTL) 是一种热化学转化过程，在升高的压力和温度下使用水作为反应介质，将生物质转化为可再生液体燃料并回收富含肥料的水。为了评估从各种原料中筛选 HTL 油的技术经济性，获得有关原料和产品分子组成的信息至关重要。现有分析方法在识别和量化生物燃料中存在的所有分子方面存在局限性。因此，需要寻找替代方法来量化饲料和预期产品的分子组成。生物油的建模工作是通过数学方法开发和验证的，使用简单的分析结果，如 CHNO 以及油的结构分析，如傅里叶变换红外，来自微藻的 HTL 衍生油的核磁共振分析。这个数学框架进一步扩展到预测从混合塑料废物、污泥等原料的 HTL 中获得的油的分子组成。基于侧链、芳环和烯烃碳在核心分子顶部的分布，开发了多维分子矩阵。使用全局优化算法（遗传算法）和局部优化算法以计算方式估计分布参数，以预测与产品的整体特性密切匹配的混合物组成。基于侧链、芳环和烯烃碳在核心分子顶部的分布，开发了多维分子矩阵。使用全局优化算法（遗传算法）和局部优化算法以计算方式估计分布参数，以预测与产品的整体特性密切匹配的混合物组成。基于侧链、芳环和烯烃碳在核心分子顶部的分布，开发了多维分子矩阵。使用全局优化算法（遗传算法）和局部优化算法以计算方式估计分布参数，以预测与产品的整体特性密切匹配的混合物组成。