To address the phase lag issue in bus voltage control of DC microgrids
a deep reinforcement learning-based active disturbance rejection control strategy incorporating a phase compensation network is developed for bidirectional DC-DC converters in energy storage systems. In this strategy
the order of the linear extended state observer is reduced
and a phase compensation network is connected in series with its total disturbance channel to provide the required lead phase angle. Subsequently
the mechanism by which the compensation link enhances disturbance suppression performance is analyzed
and the compensation parameter ranges are configured. Besides
a control parameter optimization model is established based on deep reinforcement learning algorithms. By training an intelligent agent to interact with the DC microgrid environment
the optimal combination of policy parameters is explored to achieve adaptive adjustment of observer bandwidth and correction parameters. Finally
simulations are conducted to compare the tracking ability and control accuracy of different control methods for bus voltage disturbances under typical operating conditions
and the results confirm the effectiveness of the proposed control strategy.
THOMAS T, MISHRA M K, KUMAR C, et al.Control of a PV-wind based DC microgrid with hybrid energy storage system using Lyapunov approach and sliding mode control[J]. IEEE transactions on industry applications, 2024, 60(2): 3746-3758.
GAO Z Q.Scaling and bandwidth-parameterization based controller tuning[C]//Proceedings of the 2003 American Control Conference, 2003. Denver, CO, USA, 2003: 4989-4996.
ZHENG M H, CHEN X, TOMIZUKA M.Extended state observer with phase compensation to estimate and suppress high-frequency disturbances[C]//2016 American Control Conference (ACC). Boston, MA, USA, 2016: 3521-3526.
WEI W, ZHANG Z Y, ZUO M.Phase leading active disturbance rejection control for a nanopositioning stage[J]. ISA transactions, 2021, 116: 218-231.
GHEISARNEJAD M, FARSIZADEH H, KHOOBAN M H.A novel nonlinear deep reinforcement learning controller for DC-DC power buck converters[J]. IEEE transactions on industrial electronics, 2021, 68(8): 6849-6858.