This paper is mainly about the calculation of reservoir parameters and the interpretation method for identifying oil/water beds in Ke82 well areas of Junggar basin. It is difficult to determine the reservoir parameters with common logging methods such as core calibration log because of the diversity of minerals and rocks and the complexity of pore structures in the conglomerate reservoir of Junggar basin. Optimization logging exploration is a good method to determine the porosity by establishing the multi-mineral model with logging curves based on the integration of geological, core and well testing data. Permeability is identified by BP algorithm of neural network. Hydrocarbon saturation is determined by correlating Archie’s and Simandoux formulas. Comparing the exploratory result and core data, we can see that these methods are effective for conglomerate logging exploration. We processed and explained six wells in the Ke82 well areas. And actual interpretation has had very good results, 85 % of which conform to well testing data. Therefore, this technique will be effective for identifying conglomerate parameters. This paper is mainly about the calculation of reservoir parameters and the interpretation method for identifying oil / water beds in Ke82 well areas of Junggar basin. It is difficult to determine the reservoir parameters with common logging methods such as core calibration log because of diversity minerals and rocks and the complexity of pore structures in the conglomerate reservoir of Junggar basin. Optimization logging exploration is a good method to determine the porosity by establishing the multi-mineral model with logging curves based on the integration of geological, core and well testing data Permeability is identified by BP algorithm of neural network. Hydrocarbon saturation is determined by correlating Archie’s and Simandoux formulas. Comparing the exploratory result and core data, we can see that these methods are effective for conglomerate logging exploration. We processed and explained six wells in the Ke82 well areas. And actual interpretation has had very good resu lts, 85% of which conform to well testing data. Thus, this technique will be effective for identifying conglomerate parameters.