Chinese Journal of Lasers, Volume. 47, Issue 9, 901002(2020)

Energy Control of Excimer Laser Based on Reinforcement Learning

Sun Zexu1,2, Feng Zebin1,2, Zhou Yi1,2, Liu Guangyi1,2, and Han Xiaoquan1,2、*
Author Affiliations
  • 1Optoelectronics Research and Development Center, Institute of Microelectronics of the Chinese Academy of Sciences,Beijing 100029, China
  • 2University of Chinese Academy of Sciences, Beijing 100049, China
  • show less
    Figures & Tables(12)
    Single pulse energy change under constant high voltage working mode of excimer laser
    Schematic of single pulse energy simulation model of excimer laser
    Comparison of Burst's pulse energy change measured and model under the same working mode. (a) Measured Burst pulse energy; (b) model Burst pulse energy
    Control system of excimer laser based on reinforcement learning
    Comparison of light energy values. (a) Z-N parameter tuning PI algorithm; (b) PSO tuning PI algorithm; (c) algorithm based on reinforcement learning
    Comparison of algorithm single pulse energy histograms. (a) Z-N parameter tuning PI algorithm; (b) PSO tuning PI algorithm; (c) algorithm based on reinforcement learning
    Comparison of Burst energy stability. (a) Z-N parameter tuning PI algorithm; (b) PSO tuning PI algorithm; (c) algorithm based on reinforcement learning
    Comparison of Burst dose stability. (a) Z-N parameter tuning PI algorithm; (b) PSO tuning PI algorithm; (c) algorithm based on reinforcement learning
    Comparison of dynamic stability. (a) Z-N parameter tuning PI algorithm; (b) PSO tuning PI algorithm; (c) algorithm based on reinforcement learning
    • Table 1. Parameter range of PI controller

      View table

      Table 1. Parameter range of PI controller

      ParameterKp1Ki1Kp2Ki2Kp3Ki3
      Range[90,110][240,260][40,60][30,50][40,60][10,30]
      TableQ-table1Q-table2Q-table3Q-table4Q-table5Q-table6
    • Table 2. Algorithm energy data information

      View table

      Table 2. Algorithm energy data information

      AlgorithmMinMaxAverageMedianRangeσ
      Z-N9.26310.6610101.3970.1645
      PSO9.19610.729.999101.5240.1889
      RL9.05510.6210101.5650.1293
    • Table 3. Dynamic stability data

      View table

      Table 3. Dynamic stability data

      AlgorithmPulseEnergy /mJTime /s
      Z-N300910.890.101
      PSO240410.800.165
      RL150910.890.068
    Tools

    Get Citation

    Copy Citation Text

    Sun Zexu, Feng Zebin, Zhou Yi, Liu Guangyi, Han Xiaoquan. Energy Control of Excimer Laser Based on Reinforcement Learning[J]. Chinese Journal of Lasers, 2020, 47(9): 901002

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: laser devices and laser physics

    Received: Feb. 21, 2020

    Accepted: --

    Published Online: Sep. 16, 2020

    The Author Email: Xiaoquan Han (hanxiaoquan@ime.ac.cn)

    DOI:10.3788/CJL202047.0901002

    Topics