In this paper, we study the simulation and fault diagnosis of a conventional pumping unit under balanced conditions. As the energy input of sucker-rod pumping (SRP), the motor power contains abundant information about the whole pumping cycle under SRP. It is an important step in oilfield information construction to establish a computer-aided system that is based on motor power diagnosis. Hence, we propose an SRP simulation model for generating motor power. By analyzing the working conditions of six oil wells that contain normal or insufficient liquid supply, gas lock, traveling valve leakage, standing valve leakage, and parting rod, we simulate the motor power of the six wells. In addition, we verify the simulation model using a test well with favorable performance and establish the motor power template set (MPTS) using this simulation model. To allow for computer-aided diagnosis, we propose the use of the area proportion method to extract motor power features. We establish a diagnosis model of oilwell conditions that is based on oblique decision tree and train the diagnosis model using the MPTS. Through the verification of six oil wells in the actual production of the oil field, the diagnosis model shows a favorable response. The test results show that the methods of establishing MPTS and oilwell working-condition diagnosis are feasible.