Prediction of China's coal production-environmental pollution based on a hybrid genetic algorithm-system dynamics model

This paper proposes a hybrid model based on genetic algorithm (GA) and system dynamics (SD) for coal production–environmental pollution load in China. GA has been utilized in the optimization of the parameters of the SD model to reduce implementation subjectivity. The chain of “Economic development–coal demand–coal production–environmental pollution load” of China in 2030 was predicted, and scenarios were analyzed. Results show that: (1) GA performs well in optimizing the parameters of the SD model objectively and in simulating the historical data; (2) The demand for coal energy continuously increases, although the coal intensity has actually decreased because of China's persistent economic development. Furthermore, instead of reaching a turning point by 2030, the environmental pollution load continuously increases each year even under the scenario where coal intensity decreased by 20% and investment in pollution abatement increased by 20%; (3) For abating the amount of “three types of wastes”, reducing the coal intensity is more effective than reducing the polluted production per tonne of coal and increasing investment in pollution control.

► We propos a GA-SD model for China's coal production-pollution prediction.
► Genetic algorithm (GA) can objectively and accurately optimize parameters of system dynamics (SD) model.
► Environmental pollution in China is projected to grow in our scenarios by 2030.
► The mechanism of reducing waste production per tonne of coal mining is more effective than others.