The energy efficiency of a self-powered wireless sensing system for pressure monitoring in injection molding is analyzed using Bond graph models. The sensing system, located within the mold cavity, consists of an energy converter, an energy modulator, and a ultrasonic signal transmitter. Pressure variation in the mold cavity is extracted by the energy converter and transmitted through the mold steel to a signal receiver located outside of the mold, in the form of ultrasound pulse trains. Through Bond graph models, the energy efficiency of the sensing system is characterized as a function of the configuration of a piezoceramic stack within the energy converter, the pulsing cycle of the energy modulator, and the thicknesses of the various layers that make up the ultrasonic signal transmitter. The obtained energy models are subsequently utilized to identify the minimum level of signal intensity required to ensure successful detection of the ultrasound pulse trains by the signal receiver. The Bond graph models established have shown to be useful in optimizing the design of the various constituent components within the sensing system to achieve high energy conversion efficiency under a compact size, which are critical to successful embedment within the mold structure.