Stresses due to blood flow on a blood vessel wall (hemodynamic stresses) are closely related to development and progression of circulatory diseases such as atherosclerosis and aneurysm. Therefore, for advanced diagnosis of circulatory diseases, accurate and detailed information of hemodynamics is necessary. To reproduce blood flow field, we have proposed ultrasonic- measurement-integrated (UMI) simulation, in which feedback signals are applied to the governing equations based on errors between ultrasonic measurement and numerical simulation at feedback points. Efficiency of the UMI simulation was shown by our previous numerical experiment dealing with a three-dimensional unsteady blood flow field in the descending aorta with an aneurysm. However, real ultrasonic measurement data inherently includes some errors. In this study, the effects of four major measurement errors, namely, errors due to Gaussian noise, aliasing, wall filter and lack of data, on computational accuracy of the UMI simulation were examined by a numerical experiment dealing with the blood flow field in an aortic aneurysm, the same as in our previous study. While solving the governing equations in UMI simulation, Gaussian noise did not work as an effective feedback signal, and, therefore, hardly influenced the computational result. In contrast, aliasing caused significant errors in the UMI simulation. By detecting significantly large feedback signals as a sign of aliasing and by replacing the measured Doppler velocity with the computational one, the computational accuracy of the UMI simulation was substantially improved. Effects of wall filter and lack of data especially appeared in diastole and in systole, respectively, but they were alleviated by not adding feedback signals where measured Doppler velocities were zero. Hence, UMI simulation can be performed with suppression of measurement errors.