ABC-BP neural network forecasting and application of sensitive parameter of AERMOD model [ABC-BP神经网络对AERMOD模型敏感参数的预测及应用]

作者：Yang T.; Jiang Y.; Chen J.; Xu Z.

来源：Journal of Liaoning Technical University(Natural Science), 2018, 37(1): 198-204.

DOI：10.11956/j.issn.1008-0562.2018.01.035

• AERMOD model
• Artificial bee colony algorithm
• BP neural network
• On-site assistance
• Short-time wind speed forecasting

In order to increase the prediction accuracy, this paper applied the ABC-BP neural networks algorithm into the short-time wind speed forecasting. Firstly, the validity of Artificial Bee Colony (ABC) algorithm was verified. Then the weights and thresholds of BP neural networks were optimized by ABC algorithm to predict the wind speed. After that, the prediction result was compared with PSO-BP algorithm and GA-BP algorithm. The simulation result shows that the ABC-BP neural network algorithm is superior to other algorithms in the short-time wind speed forecasting. The error of prediction is only 1.82%. Finally, the feasibility of short-time wind speed forecasting in emergency command system was analyzed and applied to the simulation of gas diffusion by the AERMOD model. The result shows that the proposed method can make the AERMOD model provide more accurate decision-making basis for emergency rescue.

• 单位
  中国科学院大学