Liu Zuhan, Hong Xianping
School of Information Engineering, Nanchang Institute of Technology, Nanchang 330099, China.
Jiangxi Province Key Laboratory of Smart Water Conservancy, Nanchang 330099, China.
Toxics. 2025 Apr 23;13(5):327. doi: 10.3390/toxics13050327.
To address the performance degradation in existing PM prediction models caused by excessive complexity, poor spatiotemporal efficiency, and suboptimal parameter optimization, we employ stacking ensemble learning for feature weighting analysis and integrate the ant colony optimization (ACO) algorithm for model parameter optimization. Combining meteorological and collaborative pollutant data, a model (namely the stacking-ACO-LSTM model) with a much shorter consuming time than that of only long short-term memory (LSTM) networks suitable for PM concentration prediction is established. It can effectively filter out feature variables with higher weights, thereby reducing the predictive power of the model. The prediction of hourly PM concentration of the model is trained and tested using real-time monitoring data in Nanchang City from 2017 to 2019. The results show that the established stacking-ACO-LSTM model has high accuracy in predicting PM concentration, and compared to the same model without considering time and space efficiency and defective parameter optimization, the mean square error (MSE) decreases by about 99.88%, and the coefficient of determination (R) increases by about 2.39%. This study provides a new idea for predicting PM concentration in cities.
为解决现有PM预测模型因过度复杂、时空效率低下和参数优化欠佳而导致的性能退化问题，我们采用堆叠集成学习进行特征加权分析，并集成蚁群优化（ACO）算法进行模型参数优化。结合气象和协同污染物数据，建立了一个比仅适用于PM浓度预测的长短期记忆（LSTM）网络耗时短得多的模型（即堆叠-ACO-LSTM模型）。它可以有效滤除权重较高的特征变量，从而降低模型的预测能力。利用南昌市2017年至2019年的实时监测数据对该模型的小时PM浓度预测进行训练和测试。结果表明，所建立的堆叠-ACO-LSTM模型在预测PM浓度方面具有较高的准确性，与未考虑时空效率和参数优化存在缺陷的相同模型相比，均方误差（MSE）降低了约99.88%，决定系数（R）提高了约2.39%。本研究为城市PM浓度预测提供了新思路。
