On June 12-13, 2016, under the background of the same cold vortex circulation, strong convective weather occurred in Shanxi for two consecutive days. The 12th was a scattered convective weather, and the 13th was a large-scale strong wind. Based on multi-source data, the influence of sub-synoptic scale system and its evolution on the storm environment and storm structure is discussed through comparison. The results show that: (1) On June 12, the convective was isolated and dispersed, and the echo intensity was less than 55 dBZ. On the 13th, the strong convection developed from quasi-linear convective storm to larger-scale bow echo, with central echo intensity of 55~60 dBZ. (2) On the 12th, a low vorticity on the west side of the 500 hPa cold vortex began to develop far away from Shanxi. On the 13th, a sub-synoptic-scale cut-off vortex was formed. The jet at the bottom of the low vortex had cyclonic shear, and the low-level matched shear line and the cold front moved eastward to affect Shanxi. (3) On the 12th, the low-level humidity was low, and the vertical wind shear at 0-3 km was only 1.6×10-3 s-1, the scattered convection was triggered by the weak convergence line on the ground. On the 13, the southward water vapor transport before shear made the low layer high humid, superimposed the dry and cold air in the middle layer to form unstable stratification. Under the influence of low vortex, the vertical shear of the 0-3 km wind increased to 5.0 ×10-3 s-1. The initial convection was triggered by the low-level denatured cold front, and then the gust front organized the storm to form a squall line. The interaction between the cold pool and 0-3 km wind shear maintained the squall line. (4) The sub-synoptic-scale cut-off vortex system on the west side of the cold vortex affected the cold front and the low-level shear line, humidity and stratification stability, vertical wind shear, cold pool and gust front, which was the key factors for triggering, organizing and developing convection to form squall line and bow echoes, and were the key systems for short-term forecasting.