To solve the increasingly prominent data quality problems about part of bad data in master station and substation in power SCADA system etc., a method for improving the data quality of power grid scheduling system is presented based on analyzing the current data quality in power grid scheduling system. This method obtains the data with optimal quality by utilizing advanced multi-source screening method and data verification technology between different systems applying the CIM/E based on CIM text as carrier and feedback data in substation according to state estimate in master station, improving the overall data quality in power grid scheduling system. New data quality optimization method, comparing with previous artificial remote comparing way, not only improves the degree of automation in data verification to save maintenance time in master station of power grid SCADA system, but also reduces bad data in master station and substation, so overall data quality is better than ever before. Practice has proved that the data quality optimization technology based on CIM/E in power grid scheduling system can effectively improve the data quality in power grid scheduling system, so it could ensure the power grid scheduling system run safely and stably.