基于机器学习算法的切削加工过程建模.docxVIP

基于机器学习算法的切削加工 摘 要 近几年来，诸如“中国制造2025”，“工业4.0”概念已逐渐成为制造业未来的导向，又伴随着人工智能这几年的火热兴起，智能与制造行业的关系不断的紧密了起来。在保证产品质量的前提下，如何使得生产效率提高，切削参数如何合理的选择，许多研究者已经开始运用机器学习来寻找答案。机器学习包含的一些重要的算法对于提高生产效率，减少制造成本是一种极优的途径。所以基于机器学习算法的切削加工过程的研究对于制造智能化有着巨大的意义。 本文对目前常用的切削参数优化的算法进行了详细的介绍，其中包括人工神经网络构成及其原理进行了阐述，并对其实现过程进行说明。遗传算法的原理以及步骤也给出详尽的介绍。最后针对两种算法的优缺点进行比较，选择神经网络算法作为本文的算法对象。 其次，在了解了目前的切削加工原理后，对它进行了探讨，其中包括探讨加工变形的主要原因，以及探究了影响切削力的主要因素，对本文的研究内容建立起到了良好的参考作用。学习了工件数值的取值规律，并利用该规律进行模拟数据的生成。 最后选取了输入输出层数和节点数，包括设置合适训练函数，学习速率，目标精度等等网络参数，利用MATLAB进行神经网络模型的建立，并将训练样本放入模型训练，分析训练结果。 关键词：机器学习；切削加工；神经网络；钛合金车削 Machining based on machine learning algorithm Abstract In recent years, such as made in China 2025, the concept of industry 4.0 has gradually become the future direction of manufacturing industry, and with the hot rise of artificial intelligence in recent years, the relationship between intelligence and manufacturing industry has become increasingly close. Under the premise of ensuring product quality, many researchers have begun to use machine learning to find out how to improve production efficiency and how to choose cutting parameters reasonably. Machine learning contains some important algorithms for improving production efficiency and reducing manufacturing costs is an excellent way. In this paper, the commonly used algorithm of cutting parameter optimization is introduced in detail, including the structure and principle of artificial neural network, and its implementation process is described. The principle and steps of genetic algorithm are also introduced in detail. Secondly, after understanding the current cutting principle, this paper discusses it, including the main causes of machining deformation, as well as the main factors affecting the cutting force, which has a good reference for the establishment of the research content of this paper. Finally, the input and output layers and nodes are selected, including setting appropriate training function, learning rate, target accuracy and other network par