DOI: 10.21276/sjet

Pages: 359-364

With the rapid development of urbanization process, the rise of smart city has become a new direction for building the future city. However, the congestion, pollution and waste problems have being happened in various cities, which are in urgent need of a scientific evaluation system to weigh. This paper has studied a number of cities in different continents, while making a detailed analysis about the impact indicators, and through the principal component analysis to select the indicators further, so the establishment of useful output data is accomplish. Then in the construction of comprehensive evaluation system, we choose the combination of heuristic algorithm—SA and BP neural network, which can improve the speed of operation and avoid falling into the local minimum. In the 79 large and medium-sized cities we randomly calculate 2 cities, which are used to do the simulation. The results show that the predicted value and the actual value of the error there are meet the requirements, thus it proves that comprehensive model established is effective.