The issue of uncoordinated regional development and unequal income distribution is an important content in the field of regional economic research, and it is also an influential factor in formulating a suitable regional development strategy. Based on the economic convergence hypothesis theory, this paper expands the inspection analysis by period, region, and industry by establishing an absolute convergence model and an extended conditional convergence model. At the same time, a contiguous space weight matrix, a distance space weight matrix, and an economic space weight matrix were introduced in order to build a spatial convergence model. Using panel data from 30 provinces and cities in China from 2004 to 2016, we explored whether China 's economy is being "flattened" and whether The impact of innovation on China's economic convergence. The research results show that: from the perspective of time, China's regional economy has significant convergence characteristics during the study period, especially from 2005 to 2010. According to this convergence trend, it will take at least more than 100 for the backward regions to catch up with the developed regions. From the perspective of space, China 's regional economic convergence pattern is roughly represented by Northeast> Eastern> West and Central, that is, the more backward the Northeast region, the faster the convergence, it is difficult to catch up with other more developed regions; from the perspective of the industrial dimension Although innovation can accelerate the rate of economic convergence as a whole, the impact on industries one, two, and three is different. Innovation mainly promotes economic convergence by acting on the secondary industry. The research conclusions in this article can provide some theoretical guidance for China to better formulate regional development policies and achieve coordination between innovation-driven and industrial policies.