Abstract: Grinding process monitoring is key to assessing the intelligence of grinding processing and ensuring manufacturing quality of the intended product. The grinding force in the grinding process is an important characteristic, and the grinding force signal has a close relationship with the degree of wheel wear, grinding heat, grinding wheel and workpiece contact state, and other entities. Therefore, the analysis of the grinding force signal using the time domain and frequency domain analysis methods can effectively reveal the grinding characteristics and wheel wear evolution mechanism in the gear grinding process. In this study, the microcrystalline corundum grinding wheel was used to conduct the profile grinding test on 20CrMnTi steel gears installed in vertical machining centers. A high-performance general-purpose dynamometer was used to collect the discrete time series data of the grinding force in real time; a confocal laser microscope and scanning electron microscope were used to observe the wear characteristics of the microcrystalline wheel surface; finally, a micrometer was used to measure the thickness of the wheel’s end-top surface to assess the degree of wheel wear in a quick and simple way. To improve the precision, automation, and intelligence level of the gear grinding process considering the microcrystalline corundum grinding wheel, online monitoring of the grinding process and wear mechanism of the grinding wheel was performed. The intrinsic correlation between the time-frequency domain signals of the grinding force, the contact status of the grinding wheel and workpiece, and the wear degree of the grinding wheel was established. Thus, in this study, a new online monitoring method for monitoring the grinding process of the grinding wheel was proposed based on real-time force signals. Meanwhile, the wear mechanism underlying the morphology of microcrystalline corundum grinding wheels was characterized and analyzed using laser confocal microscopy and scanning electron microscopy. The results showed that the distribution pattern of waveform characteristic indicators such as kurtosis, waveform index, peak index, and pulse index in the time domain spectrum of the grinding force showed a distinct trend of varying cutting depth grinding state ≥ steady-state grinding state of a significantly worn grinding wheel ≥ steady-state grinding state of a lightly worn grinding wheel > noncontact state. However, the variation of the two waveform characteristic values of the grinding force, amplitude spectrum and pulse indicator under different grinding states is completely opposite to that of the time domain signals. After assessing the three-dimensional roughness of the working surface of the grinding wheel, it is observed that there is a slight decrease and a subsequent rapid increase in the roughness of the working surface of the wheel with the evolution of wheel wear. The polycrystalline structure of the microcrystalline corundum abrasive particles gives them the ability to update and sharpen along the microcrystalline interface, showing layer-by-layer cleavage peeling.