To improve the robustness and reduce reliance on circuit parameters and engineering experience in control parameters design for voltage source converter-based high voltage direct current transmission (VSC-HVDC) systems

this paper proposes a robust

circuit-parameter-independent

and visualizable optimization method for VSC-HVDC control parameter design based on the combination of Markov transition field (MTF) and deep deterministic policy gradient (DDPG) algorithms. Firstly

the Markov transition field is used to convert one-dimensional time series waveform data of circuit power and voltage into two-dimensional Markov transition field images

and the Markov transition field loss (MTFL) function is used to assess the volatility of the data in the two-dimensional transition domain. Secondly

by combining the MTFL function with the DDPG algorithm

the method takes advantage of the MTFL function’s strong ability to evaluate the dynamic characteristics of system output time series data and the excellent generalization performance of the DDPG algorithm

thereby enabling optimization of VSC-HVDC system control parameters. Finally

MATLAB simulations and experimental results validate the effectiveness of this method.