The plunger pump is one of the important power conversion components of the hydraulic system, and its performance directly affects the safety and stability of the hydraulic system. In order to accurately evaluate the operating status of the plunger pump, a plunger pump health status assessment method based on a combination of convolutional neural network(CNN) and long short-term memory network(LSTM) was proposed, and a genetic algorithm was introduced to optimize the parameters of the neural network. The vibration signals of the plunger pump at different operating moments were collected. The energy characteristics of the vibration signals were extracted by using wavelet packets. Combined with the time-frequency domain characteristics of the signals, a dataset of the health status characteristics of the plunger pump was constructed. The health status was identified and classified by the CNN-LSTM method, and the classification results were evaluated by sample entropy. To verify the effectiveness of this health assessment method, it was applied to the experimental test of the plunger pump. The results show that the recognition accuracy of this method reaches 99%, which can effectively improve the accuracy of the health status assessment of the plunger pump.