Chiu, M.-S. Cheng, Y.-S. Felix, C.P. Luo, G.-H. Tsai, H.-J. Tseng, F.H. JACoW Publishing Geneva, Switzerland 2673-7035 978-3-95450-224-0 10.18429/JACoW-FLS2023-WE4P19 English 188-191 network storage-ring lattice synchrotron photon Contribution to a conference proceedings 2024 2024-01 https://doi.org/10.18429/JACoW-FLS2023-WE4P19 http://jacow.org/fls2023/papers/we4p19.pdf Machine learning has been applied in many fields in re-cent decades. Many research articles also presented re-markable achievements in either operation or designing of the particle accelerator. This paper focuses on the simulated orbit correction by neural networks, a subset of machine learning, in Taiwan Photon Source. The training data for the neural network is generated by accelerator toolbox (AT).