The excessive reliance on fossil fuels has resulted in an energy crisis, environmental pollution, and health problems, calling for alternative fuels such as biodiesel. Here, we review computational chemistry and machine learning for optimizing biodiesel production from waste. This article presents computational and machine learning techniques, biodiesel characteristics, transesterification, waste materials, and policies encouraging biodiesel production from waste. Computational techniques are applied to catalyst design and deactivation, reaction and reactor optimization, stability assessment, waste feedstock analysis, process scale-up, reaction mechanims, and molecular dynamics simulation. Waste feedstock comprise cooking oil, animal fat, vegetable oil, algae, fish waste, municipal solid waste and sewage sludge. Waste cooking oil represents about 10% of global biodiesel production, and restaurants alone produce over 1,000,000 m³ of waste vegetable oil annual. Microalgae produces 250 times more oil per acre than soybeans and 7–31 times more oil than palm oil. Transesterification of food waste lipids can produce biodiesel with a 100% yield. Sewage sludge represents a significant biomass waste that can contribute to renewable energy production.

对化石燃料的过度依赖导致了能源危机、环境污染和健康问题，需要生物柴油等替代燃料。在这里，我们回顾了用于优化废物生物柴油生产的计算化学和机器学习。本文介绍了计算和机器学习技术、生物柴油特性、酯交换反应、废料以及鼓励利用废物生产生物柴油的政策。计算技术应用于催化剂设计和失活、反应和反应器优化、稳定性评估、废物原料分析、工艺放大、反应机制和分子动力学模拟。废弃原料包括食用油、动物脂肪、植物油、藻类、鱼类废物、城市固体废物和污水污泥。废弃食用油约占全球生物柴油产量的 10%，仅餐馆每年就产生超过 1,000,000 m ³废弃植物油。每英亩微藻的产油量是大豆的 250 倍，是棕榈油的 7-31 倍。食物垃圾脂质的酯交换反应可生产生物柴油，收率 100%。污水污泥是一种重要的生物质废物，有助于可再生能源的生产。