【摘要】 大气污染物监测过程中,监测点分布广泛,普遍采用碎片化重组传输方法。但由于大气污染物监测信息规模大、维度高,难以准确划分碎片类别,导致数据传输过程中的误差较大。为此,提出大气污染物监测信息碎片化重组传输方法。使用Lo Ra-RTU技术、TCP~/IP协议栈构建大气污染物监测信息碎片化重组传输平台,采用近似熵、对数均值计算方法挖掘污染物信息变化规律及特征。基于特征并结合K-means算法实现信息碎片化划分,通过余弦相似性函数对大规模、高维度的监测数据相似碎片类进行融合、排序,并上传至传输平台,最终完成信息碎片化重组传输。实验结果表明,所提方法的大气污染物监测信息碎片化重组相对误差小,且传输消耗的内存开销小。更多还原

【Abstract】 In the process of air pollutant monitoring,monitoring points are widely distributed,and fragmented reorganization and transmission methods are commonly used. However,due to the large scale and high dimensionality of air pollutant monitoring information,it is difficult to accurately classify the fragments,resulting in large errors in the data transmission process. To this end,a fragmented reorganization and transmission method for air pollutant monito-/ring information is proposed. Using LoRa-RTU technology and the TCPIP protocol stack,a platform for fragmented reorganization and transmitted air pollutant monitoring information was constructed. Approximate entropy and logarithmic mean calculation methods were used to explore the changing patterns and characteristics of pollutant information.Information fragmentation was partitioned based on features and combined with the K-means algorithm. Similar fragments of large-scale,high-dimensional monitoring data were fused and sorted using the cosine similarity function,and uploaded to the transmission platform, ultimately completing the fragmented reorganization and transmitted information. The experimental results demonstrate that the proposed method achieves low relative error in the fragmented reorganization of air pollutant monitoring information and consumes minimal memory overhead.更多还原