Abstract: The high-temperature and high-pressure low-permeability gas reservoirs in Dongfang A Area of Yinggehai Basin have not yet achieved effective large-scale development, making the search for favorable reservoirs crucial. Utilizing rock core testing and analysis methods such as laser granulometry, thin section petrography, scanning electron microscopy, and high-pressure mercury injection, we first investigated the microscopic differences in reservoir characteristics, and classified and evaluated reservoir quality differences. Then, by analyzing the relationship between reservoir quality differences and well logging features, we selected well logging parameters suitable for evaluating reservoir quality. Using the principal component analysis method, we constructed sensitivity factor curves that reflect the quality of reservoirs. Finally, based on the co-simulation method of sensitivity factor curves and seismic waveform indicators, we predicted the reservoir quality. Results indicate that reservoir quality is mainly determined by mudstone production characteristics. When the mud content and the sediment grain size of the reservoirs are similar, the reservoir quality in the study area could be divided into three grade levels based on mudstone production characteristics. For Grade Ⅰ reservoirs, the mudstone production state is characterized by orderly distributed mudstone bands, with deposits of coarse-grained siltstone, and strong dissolution effects. Grade Ⅱ reservoirs are characterized by mixed distribution of mudstone, with medium to coarse-grained siltstone deposits, and moderate to strong dissolution intensity. Grade Ⅲ reservoirs exhibit scattered mudstone production in a dispersed state, with the deposition of fine-grained siltstone in weak dissolution intensity. Grade Ⅰ and Grade Ⅱ reservoirs are favorable reservoirs. The established reservoir quality grading model has a cumulative variance contribution rate of 98.1%, capable of reflecting reservoir quality differences in the study area. A reservoir quality prediction method based on the synergy of reservoir quality sensitivity factors and seismic waveform indicators was proposed. The predicted results are highly consistent with actual drilling data, revealing the spatial distribution of favorable reservoirs. This approach has significant implications for guiding development decisions and well deployment in gas fields.