The work aims to objectively evaluate the use effectiveness of wheelchair by using surface electromyographic equipment to test the electromyographic signals during wheelchair use, and to construct a human-machine evaluation model by integrating posture signals from an inertial measurement unit (IMU). Folding, braking, and slope of the wheelchair were tested to evaluate its performance, and the fatigue condition between muscle groups was compared to determine the folding and braking methods that were beneficial to the wheelchair. The experimental subjects were required to use two folding methods and three braking methods of different wheelchairs to perform the wheelchair tasks and to collect sEMG and IMU signals during the operational task process. After the experimental task, the NASA-TLX scale was filled out. Based on the evaluation indicators of the model, the experimental data were compared and analyzed. The wheelchair subject to horizontally folded folding and concave brake (hand brake in front) had less fatigue, and the user's force was minimal at a 3~4° slope range, which was more comfortable. This verified the feasibility of the model in wheelchair human-machine evaluation and provided a reference for optimizing wheelchair design. The human-machine evaluation model is suitable for evaluating product performance. Meanwhile, the wheelchair performance evaluation method proposed in this study, which integrates surface electromyography and posture information, has higher accuracy and can effectively evaluate the performance level of wheelchair users.