In this article, a new kind of neural network model named multi-scale convolutional echo state network (MCESN) is proposed for solar irradiance prediction, which integrates the strong feature extraction capability of convolutional neural network (CNN) and the fast yet efficient prediction ability of echo state network (ESN). Firstly, the feature information at different time scales of solar irradiance (one dimensional series) data are extracted and selected by multi-scale CNN (MCNN) in the pre-training stage. Then, the trained features extracted above are concatenated and passed to ESN module as the input signal, which can be further encoded into high-dimensional state space; Meanwhile, the target solar irradiance value is fitted and predicted by ESN in the prediction phase. Finally, the effectiveness of MCESN is evaluated by hourly solar irradiance prediction. In experiment, RMSE, MAE, MAPE and R are chosen as four metrics to evaluate the performance of the proposed model. Simulation results demonstrate that the proposed MCESN perform better than classical ESN, MCNN, backpropagation (BP) random forest (RF), long short time memory (LSTM) and deep ESN (DESN) algorithms. INDEX TERMS Echo state network, multi-scale convolution, pre-processing, solar irradiance prediction. I. INTRODUCTION 1 In recent years, solar irradiance prediction plays signifi-2 cant roles in renewable energy systems. Solar energy is a 3 sustainable, environmental friendly, and, more importantly, 4 inexhaustible resource, and its potential cannot be matched 5 by other energy sources [1]. However, solar energy presents 6 highly nonlinear, naturally volatile and intermittent [2], due 7 to the fluidity of cloud and other variable meteorological 8 factors [3]. Undoubtedly, the uncertainty will have a certain 9 impact on solar energy systems [4]. Consequently, it is of 10 importance and urgency to predict solar energy accurately, 11 to ensure the security, efficiency and reliability of renewable 12 energy systems [5]. 13 Various techniques have been successfully proposed for 14 solar irradiance prediction, which can be roughly divided into 15 three categories: statistical models [6], [7], machine learning 16 algorithms [8]-[11] and artificial neural networks [12], [13]. 17 Among these techniques, artificial neural networks (ANN) 18 have become the most mainstream methods, because of their 19 strong nonlinear fitting ability and accurate prediction results. 20 This article has been accepted for publication in IEEE Access.