Principal Component Analysis-Improved Fuzzy Genetic Algorithm

Since traditional intelligent algorithms composed of air quality prediction are commonly used, but such algorithms still have shortcomings for the validity of data, especially the problem of time-series prediction data. In order to investigate the problem of traditional intelligent algorithms for effective time-series data, this paper proposes a principal component analysis -improved fuzzy genetic algorithm (PCA-IFGA), in order to get more effective data in predicting air quality. the PCA-IFGA algorithm just divide into two modules. The first is PCA (Principal Component Analysis, PCA) which solves the problem of"dimensionality reduction" of air quality data by analyzing the largest individual differences revealed by taking the principal components and discovering the characteristics that affect the data in air quality prediction. The IFGA (Improved Fuzzy Genetic Algorithm) improves the traditional FGA (Fuzzy Genetic Algorithm) by increasing the variation rate of the algorithm to enhance the population diversity and facilitate the algorithm to jump out of the local optimum, while preserving the superior population diversity, increasing the crossover rate and reducing the variation rate of the algorithm to accelerate the convergence of the algorithm, thus the convergence efficiency and correct rate of the algorithm are improved. The experimental results show that PCA-IFGA is significantly better than BP algorithm, LSTM and SVM algorithms in terms of stability and full correctness.