Misalignment in the rotating machine reduces the life of the shafting component and increases the power consumption, which is a type of failure that greatly damages the rotating machinery. Detecting and repairing misalignment at an early stage is important for rotating machines such as pumps. In this study high-precision state determination is carried out based on the symptom parameters of current, voltage, flow and pressure signals. Moreover, efficiency-based trend management of the pump is proposed by accurately diagnosing abnormal degree (misalignment, cavitation, etc.) through vibration signals, to solve the inefficiency problem of pump based on theoretical research and experimental research.