配电云平台的决策级数据融合及其并行化

Decision level data fusion and parallelization of power distribution cloud platform

随着传感器技术的不断发展,配电主站中包含的传感器数量不断增加,配电云平台能够接收海量数据。为了提高数据的利用率,同时提高云平台对数据的处理效率,本文提出一种关于配电云平台的决策级数据融合方法及其并行化方案,通过计算传感器的重要程度判断传感器网络中各传感器反映某事项的程度,从而决定是否将数据实时传输到应用层,同时利用改进的基于权重的D-S理论在应用层实现进一步的数据融合,整个过程利用Spark进行并行化计算。本文提出的数据传输及融合方法能够在保证数据传输完整性的前提下大大提高应用层的决策效率,尤其对于需进行实时判断的事件,所提方法能够保证配电云平台实时高效地做出决策。

With the continuous development of sensor technology,the number of sensors included in the power distribution master station is increasing.The power distribution cloud platform can receive massive amounts of data.In order to improve the utilization rate of data and speed up data processing in the cloud platform,this paper proposes a decision-level data fusion method on the distribution cloud platform and its parallelization scheme.By calculating the influence of the sensors,it is possible to determine the degree that each sensor in the sensor network reflects a certain item,thereby deciding whether to transmit the data to the application layer in real time.At the same time,the improved weight-based D-S theory is used for further data fusion at the application layer,and the entire process uses Spark for parallel computing.On the premise of ensuring the integrity of data transmission,the data transmission and fusion method proposed in this paper can greatly improve the decision-making efficiency of the application layer.Especially for events that require real-time judgment,this method can enable the distribution cloud platform to make decisions in real time and efficiently.