Although energy harvesting embedded systems can work permanently when the chance to harvest energy is steady, it is necessary to adopt an appropriate operation strategy due to the limitation of harvested energy. In this paper, we propose an operation scenario model for energy harvesting embedded systems. The proposed model is defined as a set of operation points selected from periodical candidate points to execute the application. In order to ensure the stability of system operation, the application has to be executed at least once within a given maximum interval. In this work, an operation quality in the model is defined as execution frequency. We further propose an efficient algorithm to maximize the operation quality, i.e., to find a maximum set of operation points. The proposed algorithm selects as many execution points as possible while satisfying the constraints of the battery capacity and the system stability. We prove the optimality of proposed algorithm. Additionally, the experiments demonstrate that proposed algorithm can find one of the optimum solution quickly.