1: \begin{abstract}

2: We present a multi-objective evolutionary optimization algorithm that uses 

3: Gaussian process (GP) regression-based models to select trial solutions 

4: in a multi-generation iterative procedure. 

5: In each generation, a surrogate model is constructed for each objective function 

6: with the sample data. 

7: The models are used to evaluate solutions 

8: and to select the ones with a high potential before they are 

9: evaluated on the actual system. 

10: Since the  trial solutions selected by the GP models tend to have better performance 

11: than other methods that only rely on random operations, the 

12: new algorithm has much higher efficiency in exploring the parameter space. 

13: %\zhesnote

14: {Simulations with multiple test cases 

15: show that the new algorithm has a substantially higher convergence speed and stability than  

16: NSGA-II, MOPSO, and some other more recent algorithms.} 

17: \end{abstract}

18: