Evaluation and prediction of the resources and environmental carrying capacity based on NMF-KELM
Due to the important guiding significance for regional sustainable development of the early warning of resource and environmental carrying capacity,a new evaluation and prediction method of the resource and environmental carrying capacity based on non-negative matrix factorization(NMF)and kernel extreme learning machine(KELM)is proposed.The evaluation index system of resource and environmental carrying capacity of Jiangxi Province is constructed,and the NMF method is introduced to quantitatively measure and systematically analyze the resource and environmental carrying capacity of Jiangxi Province from 2005 to 2020.Verification and analysis of resource and environmental carrying capacity results is used to compare with grey relational analysis and classified-array polygon method.Then,resource and environmental carrying capacity prediction model based on NMF and KELM is established to forecast the evolution trend.The study results show:① from 2005 to 2020,the resource and environmental carrying capacity index increased from 0.096 3 to 0.797 5,showing an overall trend of fluctuating increase,with the rapidly developing socio-economy being the most direct driving force of carrying capacity.② The trends and conclusions calculated by NMF,grey correlation method and full arrangement polygon graphic method were consistent,but NMF evaluation results were more objective.③ Environment subsystem was the main factor to restrain the improvement of resource and environmental carrying capacity,of which the industrial waste gas emission per 10 000 yuan GDP was the most important influence factor.④ Compared with BP neuron network and grey model,the proposed model based on NMF and KELM has a better prediction and generalization,thus it can better predict the evolution trend of resource and environmental carrying capacity in Jiangxi province.
Resource and environmental carrying capacityNon-negative matrix factorization(NMF)Grey relational degreeKernel extreme learning machine(KELM)Jiangxi Province
万方数据
维普
