A SAFSA- and Metabolism-Based Nonlinear Grey Bernoulli Model for Annual Water Consumption Prediction

摘要

Although the traditional grey model is an effective method in predicting the water consumption, it is difficult to reflect the sequence of randomness, fluctuation and discreteness. In this study, a prediction model of urban annual domestic water consumption is proposed. AF-MNGBM(1,1) model is established as an optimized nonlinear grey Bernoulli model to select the optimal parameters, by combining a self-adaptive artificial fish swarm algorithm (SAFSA) and the metabolic method. The time series data of Wuhan's residential water consumption between 1994 and 2017 are used to verify the effectiveness of AF-MNGBM(1,1) in predicting annual water consumption. Meanwhile, the prediction results are compared with those of common NGBM(1,1) model, traditional GM(1,1) model and grey Verhulst model. The results show that the AF-MNGBM(1,1) model has higher prediction accuracy. The optimized model provides a new method in predicting the mid- and long-term annual water consumption with the data of randomness, fluctuation and discreteness in different industries. The model has been applied in the new round water quota modification of Wuhan.