This effort attempts to study the change point problem in the area of non-linear profiles. A method based on Artificial Neural Networks (ANN) is proposed for estimating the real time of a single step change. The feature vector of the proposed Multi-Layer Perceptron (MLP) is based on Z and control chart statistics for nonlinear profiles. The merits of the proposed estimator are evaluated through simulation experiments. The results show that the estimator provides an accurate estimate of the single step change point in non-linear profiles in the selected case problem.

中文翻译：

---

使用人工神经网络检测和估计非线性轮廓中单步变化的时间

这项工作试图研究非线性轮廓区域中的变化点问题。提出了一种基于人工神经网络（ANN）的估计单步变化实时性的方法。拟议的多层感知器（MLP）的特征向量基于Z和非线性轮廓的控制图统计信息。通过仿真实验对提出的估计器的优缺点进行了评估。结果表明，在选定的案例问题中，估计器可提供非线性轮廓中单步变化点的准确估计。