Smart Beijing: Correlation of Urban Electrical Energy Consumption with Urban Environmental Sensing for Optimizing Distribution Planning

This paper focuses on consumer-side activities and investigates the environmental impact of the electrical energy consumer (transportation, buildings, street lighting, etc.), in order to improve the operational efficiency of the city as a whole. To achieve the goal we propose a two-layer approach which consists of a sensing layer and an application layer. To begin, circumstances within the city of Beijing are identified which have a large temporal impact on environmental conditions. A taxi-based vehicular ad hoc network is proposed as a low-cost and efficient approach for urban sensing using information collection methods of ”Internet of Things” to capture various environmental parameters (air quality, noise pollution, traffic levels, water quality, etc.) in a distributed manner in real time. Using this data and correlation analysis techniques, global machine learning approaches will be trained to recognize important city events and dynamics which will affect electrical power consumption and create anomolies in pollution levels in specific locations, such as sporting events, rallies and fairs. This event-driven situational information could then indicate a predictive relationship with electrical energy consumption information, which can be exploited in the application layer. Besides, we introduce a middleware above the application layer to proactively plan a certain information management system around the Common Information Model and other information standards, in order to deal with lack of stable integration of the standards. Utilities could then get intelligence and values as expected from the data that will be collected from our system. Keywords-urban sensing; electrical energy consumption; twolayer approach; middleware; correlation analysis.