The electro-hydraulic servo pump control system has the advantages of high power to weight ratio and fast response,and is widely used in various fields.But how to conduct more effective health assessments for the system and further ensure its safety and reliability has become a necessary issue to face.Research on health assessment methods was conducted according to clearing principles,establishing mathematical models,establishing simulation models,and making simulation experiments.Three health assessment indicators as oil volume gas content,air gap magnetic density,and leakage coefficient,were proposed and their thresholds were determined.Then a LGA(LSTM-GRNN-ANN) deep neural network health assessment method was constructed and simulation analysis was conducted.The results show that the accuracy of this method is about 97.48%,which is higher than LSTM and GRNN health assessment methods.This provides theoretical support for further research on health assessment of electro-hydraulic servo pump control systems.