| 摘要: |
| 根据 2017—2018 年宁波市镇海区负氧离子实况观测资料及其同期观测的天气环境资料，使用偏相关函数分析了各要素因子之间的关联性，并采用回归统计方式建立了负氧离子预报模型。结果表明：负氧离子浓度变化与空气温度和相对湿度呈显著性正相关，相关系数（R）分别为 0.569 和 0.210，与空气质量指数（AQI）呈显著性负相关，相关系数（R）为 -0.578；构建的负氧离子预报模型在预测负氧离子浓度等级方面具有较好的预报能力，预报模型的回归方程决定系数（R2）为 0.677，模型的负氧离子浓度历史回算值与观测值两者之间的相关性（R）达 0.803，浓度等级的历史回算准确率总体为 78%；预报模型的独立性检验进一步证实了本预报模型对负氧离子浓度等级的预报潜力，预报准确率总体可达71%，且预报模型表现出负氧离子浓度等级越高，预测准确性越强。 |
| 关键词: 负氧离子 浓度等级 预测模型 空气质量 |
| DOI： |
| 分类号: |
| 基金项目:浙江省气象科技计划项目（2017ZD13，2020ZD09），浙江省政府投资项目（ZJQX2017166） |
|
| Study on Prediction of Concentration Grade of Negative Oxygen Ion |
|
Li Zhengqun1, Zheng Jian2, Hu Xiao3, Zhu xiaocui4
|
|
1.Zhejiang Climate Center;2.Fenghua Weather Bureau;3.Zhenhai Weather Bureau;4.镇海a Weather Bureau
|
| Abstract: |
| Based on the observation data of negative oxygen ions and the meteorological and environmental data observed in Zhenhai district of Ningbo city during 2017-2018, the partial-correlation function was used to analyze the correlation between air temperature, relative humidity, wind speed, precipitation, air quality index (AQI) and negative oxygen ion. Subsequently, a prediction model of negative oxygen ion was established by using the multivariate regression method. The results showed that there was a signi?cant positive correlation between negative oxygen ion concentration and air temperature and relative humidity, and its correlation coef?cients (R) were 0.569 and 0.210, respectively. Moreover, the concentration changing of negative oxygen ion was negatively correlated with AQI, and the correlation coefficient (R) was-0.578. The prediction model of negative oxygen ion that was constructed by air temperature, relative humidity and AQI in this study, had a good capability for predicting the concentration grade of negative oxygen ion. The coef?cient (R2) of determination of the prediction model was 0.677, the correlation coef?cient (R) between the model recalculation and the observation of negative oxygen ion concentration was 0.803, moreover, the accuracy rate of the model recalculation was 78%, as for the concentration grade of negative oxygen ion. The independent test had further demonstrated that the prediction model has good performance for predicting the concentration grade of negative oxygen ion. Generally, the accuracy rate of the model prediction was up to 71% in the independent test, and the higher the concentration grade of negative oxygen ion was, the more accurate the prediction was. |
| Key words: negative oxygen ion concentration grade prediction model air quality |
