In the context of globalization, with the continuous advancement of artificial intelligence technology, machine translation technology has developed rapidly. Although the quality of machine translation has improved, there are still limitations in ensuring accuracy, especially in texts requiring high levels of professionalism and precision. Therefore, it is particularly important to study the translation model that combines machine translation with human post-editing. This study takes the article Synthetic polymers: A review of applications in drilling fluids, the most-cited article in Petroleum Science in 2024, as an example. Based on the manual comparative review, the study deeply analyzes problems such as mistranslated vocabulary, incorrect sentence structure segmentation, and poor discourse cohesion in machine translations of petroleum science and technology texts. It also explores human post-editing methods such as verification, sentence structure clarification, and cohesive techniques to address these problems and improve the translation quality of petroleum science and technology texts. This study finds that machine translation demonstrates high efficiency in the translation of petroleum science and technology texts, capable of rapidly generating roughly accurate translations. However, there are still deficiencies at the lexical, syntactic, textual, symbolic, and graphical levels. The author hopes that this study can provide valuable insights for the application and development of the human-machine collaboration model in the translation of specialized texts.