The detection and monitoring of oil-gas-water three-phase flow is a hot and difficult topic in oil-gas well industry. Concentrating on combining the three sub-states of oil, gas and water to form a whole flow state, a monitoring framework based on partial representation is proposed. In this framework, a differential pressure (DP) -pulse wave ultrasonic Doppler (PWUD) system is designed for obtaining information of three-phase structure and the intrinsic relationships among different phases, so as to accurately characterize the sub-states of oil, gas and water. Some information approaches such as Variational Mode Decomposition (VMD) and statistics method are applied to detection signal analysis to further realize the conversion of data to features. On the basis of obtained state features, the parts-based Nonnegative Matrix Factorization (NMF) is adopted to model and monitor the flow states in oil-gas-water flow process. The proposed novel scheme about partial representation-based horizontal oil-gas-water flow state monitoring via dual-modal detection is verified by dynamic experiments on three-phase flow loop at Tianjin University.