We consider the problem of estimating the number of significant components in high-dimensional principal component analysis (PCA). We propose a new penalized approach using the explained variance ratio and the ratio of the adjacent eigenvalues of sample covariance matrices of p variables. We show the consistency of the estimator for different types of data matrices including independent data and some times series data when the dimension p and the sample size n both tend to infinity with their ratio being a positive constant. Hence it works under weaker conditions than some existing approaches such as AIC and BIC. Simulation studies are also conducted to illustrate its performance.  It's a joint work with Zhixiang Zhang and Guangming Pan.