Abstract The use of biodiesel as an alternative fuel in the pursuit of renewable and sustainable energy has raised new technological, economic and environmental concerns. Although it has been almost 150 years since the introduction of internal combustion engine, researches seeking engineering solutions still continue. Driving quality, performance and fuel economy have been improved while emissions have been lowered significantly. But there has not been any unified analytical model that can capture the internal combustion engine as a complete system from thermodynamic, mechanical or chemical perspective per se. With experimental research methods usually too involving in terms of engineering costs, computational approaches to deliver numerical solutions have been inevitable as a research methodology – or even sometimes, is left as the only feasible method. In light of these concerns, this article reviews a few trending modelling methods of (1) analytical, (2) regression and (3) artificial neural network methods, and optimisation methods of (1) response surface methodology, (2) Taguchi method and (3) genetic algorithm which have been popularly employed in internal combustion engine research. The review recommends to confluence of advanced statistical methods and emerging popular machine learning algorithms to engine research for deriving comprehensive pragmatic models as empirical compromise.

中文翻译：

---

压燃式内燃机计算与实证研究新进展综述

摘要 在追求可再生和可持续能源的过程中，使用生物柴油作为替代燃料引起了新的技术、经济和环境问题。尽管内燃机问世已近 150 年，但寻求工程解决方案的研究仍在继续。驾驶质量、性能和燃油经济性得到改善，同时排放量已显着降低。但是还没有任何统一的分析模型可以从热力学、机械或化学的角度将内燃机作为一个完整的系统。由于实验研究方法通常太涉及工程成本，因此提供数值解的计算方法作为一种研究方法是不可避免的——甚至有时，被留下作为唯一可行的方法。鉴于这些问题，本文回顾了 (1) 分析、(2) 回归和 (3) 人工神经网络方法的一些趋势建模方法，以及 (1) 响应面方法、(2) 田口方法和(3) 内燃机研究中普遍采用的遗传算法。该评论建议将先进的统计方法和新兴的流行机器学习算法融合到引擎研究中，以推导出全面的语用模型作为经验妥协。(2) 田口方法和 (3) 遗传算法已广泛应用于内燃机研究。该评论建议将先进的统计方法和新兴的流行机器学习算法融合到引擎研究中，以推导出全面的语用模型作为经验妥协。(2) 田口方法和 (3) 遗传算法已广泛应用于内燃机研究。该评论建议将先进的统计方法和新兴的流行机器学习算法融合到引擎研究中，以推导出全面的语用模型作为经验妥协。