| Industry is not only the main source of China’s real economic growth,but also the main body of energy consumption.Industries(steel,electrolytic aluminum,etc.)engaged in the combustion of fossil fuels(coal,oil,natural gas,etc.)are generally considered to be the main industries of industrial energy consumption.They release a lot of heat in the production process and form industrial heat sources.Shandong Province is a major industrial province in China.Investigating the temporal and spatial distribution of industrial heat sources in Shandong Province can reveal the spatial distribution pattern and development trend of industry in Shandong Province to a certain extent,which has important reference value and guiding significance for improving people’s understanding of the effectiveness of industrial transformation and upgrading in Shandong Province and formulating reasonable industrial policies.Thermal infrared remote sensing can determine the target temperature remotely.It has the advantages of wide coverage and is not limited by surface observation conditions.It provides the possibility for continuous,continuous and rapid monitoring of industrial thermal anomalies.However,due to the lack of precision of biomass thermal source extraction and infrared classification.Taking Shandong Province as the research area,this paper classifies and extracts industrial heat sources from the dimensions of time,space and thermal radiation,and deeply excavates the distribution characteristics of industrial heat sources.The main research contents are as follows:(1)In view of the accuracy of industrial heat source identification and the inaccurate classification of industrial subclass heat sources,taking Shandong Province as the research area,using the VIIRS nightfire data with spatial resolution of 750 m,the industrial heat source objects are extracted by DBSCAN algorithm and time series clustering method,and the K-nearest neighbor classification algorithm is used to classify the industrial subclass heat sources,Finally,combined with Google high score image and other data,the extracted industrial heat source object is visually interpreted to determine the extraction accuracy of industrial heat source object.The experimental results show that the number of correct industrial heat source objects extracted by this method in Shandong Province is 194,the extraction accuracy is 99.48%,and the overall accuracy of industrial subclass heat source object classification is 84.54%.(2)Aiming at the problem of insufficient spatial refinement in the extraction of industrial heat sources,based on the VIIRS active fire data with spatial resolution of 375 m,the OPTICS algorithm is used to divide the heat source objects,and the contour coefficient is introduced to evaluate the advantages and disadvantages of the clustering results.The temperature feature templates of industrial heat sources and non-industrial heat sources are constructed,and the accurate extraction of industrial heat source objects is realized by combining with the BP neural network.Finally,the extracted industrial heat source objects are visually interpreted by combining with the data such as Google high-resolution images,and the extraction accuracy of industrial heat source objects is determined.The experimental results show that the number of industrial heat source objects in Shandong Province is 445,and the extraction accuracy is 96.16%.Under the condition of high extraction accuracy of industrial heat source,the number of industrial heat source objects extracted is relatively large,which is conducive to analyzing the spatial and temporal characteristics of industrial heat source in Shandong Province.(3)The industrial heat sources in Shandong Province are mainly distributed at the junction of Zibo and Binzhou,southeast and central Jinan,eastern Heze,southern Linyi,southern Rizhao,central Jining and central Zaozhuang,showing an obvious spatial aggregation trend.The number of industrial heat source objects and thermal anomaly points in Shandong Province showed a decreasing trend from 2013 to 2020.Industrial transformation and upgrading have a great impact on industrial heat sources and have achieved remarkable results.According to the industrial heat anomaly points and thermal radiation energy grid,Jinan,Heze,Rizhao and Linyi have higher industrial heat source aggregation and energy consumption than other cities.By analyzing the changes in the number of industrial heat source objects and heat anomaly points in Shandong Province and Wuhan City during the period of new corona pneumonia(COVID-19)epidemic(January to May 2020),it can be seen that the new corona pneumonia epidemic has no significant influence on the change of industrial heat source in Shandong Province.It can be inferred that the production activities of the industries engaged in fossil fuel combustion in Shandong Province during the period of new corona pneumonia epidemic(January to May 2020)are not significantly affected. |
PDF Full Text Request