Research On The Relationship Between Air Quality And Meteorological Conditions And It's Forecasting In Beijing

In recent years,foggy and hazy weather is becoming more frequent in Beijing.The public began to pay attention to the cause of haze,the sources of pollutants,the impacts of air pollution and hazards,and so on.Therefore,a forecast model of air quality based on meteorological conditions should be established in a bid to conduct research into the distribution characteristics of meteorological conditions and air quality in Beijing,and study the effects of meteorological conditions on air quality carefully.Firstly,this paper analyzes the trend of climate change in Beijing?1951-2016?,and then particularly analyzed the temporal distribution of meteorological elements in Beijing during2013-2016,and studied the basic meteorological conditions of Beijing in recent years.Secondly,the temporal variation characteristics of air quality in Beijing?including interannual,annual and intra-annual variations?and spatial distribution characteristics are analyzed.Thirdly,the correlation between meteorological elements,pollutant concentrations,meteorological elements and pollutant concentrations is studied.Finally,two kinds of air quality prediction models based on BP neural network and GA-BP neural network are established.The main conclusions are as follows:From 1951 to 2016,the annual average wind speed of Beijing was 2.4m/s,the annual relative humidity of 56.6%,and the annual average precipitation of 589.8mm,which all showed decreasing trends.The annual average temperature was 12.3?,and the annual average air pressure of 1011.6 hPa,which both showed increasing trends.Except the fact that the trend of annual mean wind speed changed slowly,the trend of other meteorological elements changed more obviously.The annual variation of wind speed,temperature,relative humidity,air pressure and precipitation in Beijing from 2013 to 2016 were more regular.The wind speed was the fastest in spring,the smallest in autumn,and close to annual average in summer and winter.The temperaturewas the highest in summer,the lowest in winter,and close to the annual average in spring and autumn.The relative humidity was the lowest in spring,the higher in summer and autumn,and the lower in winter but slightly higher than that in spring.The atmospheric pressure was the lowest in summer,the highest in winter,and close to the annual average in spring and autumn.The precipitation was the highest in summer,close to the annual average in autumn,and lower in spring and winter.The air quality showed an improvement trend in Beijing from 2013 to 2016.Specially,in2016,air quality was significantly better than that in other years.In addition,the annual average concentration of PM₁₀,SO₂,CO,NO₂ during 2013-2016 showed decreased trend.But the average concentration of PM_2.5,O₃ in 2013 was significantly lower than that of the other years,and the increase of concertation in 2014 was more significant than in 2013.The results of the spatial distribution of air quality showed that the air quality in mountainous areas of Beijing is obviously superior to that in plain areas,with a decreasing trend from south to north in the whole city,the urban area was characterized by high southeast and low northwest,and the air quality of urban area was superior to that of southern suburb,but worse than in the northern suburbs.By studying the correlation between meteorological elements and air quality elements,a strong correlation between meteorological elements and air quality concentration was found,and a significant test was made.There was a negative correlation between wind speed and air quality factors,that is,the greater the wind speed,the smaller the pollutant concentration,the better the air quality.The same as the correlation between temperature and air quality factors.In addition,the more the relative humidity?air pressure?was,the higher the pollutant concentration was,which meaned the worse the air quality.Then,the negative correlation also between precipitation and most air quality elements,indicating that rainfall had the effect of clean air quality.That means the higher the rainfall,the smaller the pollutant concentration and the better the air quality.The air quality prediction models of Beijing based on BP neural network and GA-BP neural network are designed respectively.The results showed that the accuracy of air quality prediction model based on BP neural network was 73.73%,and that based on GA-BP neural network was74.80%.The air quality prediction models based on BP neural network and GA-BP neural network have reference value in accuracy and can provide data reference and theoretical support for air quality prediction and decision-making in Beijing.