It is of great significance to realize transient stability evaluation accurately and quickly for stable and safe operation of power system. Aiming at the problem of low prediction accuracy in the current transient stability evaluation of grid-connected photovoltaic energy storage systems, a transient stability evaluation method based on integrated deep Random forest algorithm （gcForest-AdaBoost） is proposed by combining deep learning and ensemble learning techniques. Firstly, the initial input features are constructed according to the steady-state components of the grid-connected photovoltaic energy storage system during operation. Secondly, in order to reduce the overfit of the model and ensure that the model still has strong learning ability after the number of cascades increases, a transient stability evaluation method based on gcForest-AdaBoost is proposed by combining deep random forest algorithm and AdaBoost algorithm. Thirdly, the gcForest-AdaBoost model is trained with input feature set, and the transient stability evaluation model of grid-connected PV energy storage system is established. Finally, the IEEE 39-node system is connected to the photovoltaic energy storage unit to build a case system for simulation analysis and data acquisition, and the evaluation results are obtained. Numerical examples show that the proposed model can effectively analyze the transient stability of grid-connected photovoltaic energy storage systems, and it is found that the proposed model has good robustness and generalization ability.

solar power generation  /  energy storage  /  transient stability  /  random forests  /  gcForest-AdaBoost  /  input characteristics