Doctoral Defense

Spatial-Temporal Frameworks for Renewable Energy and Power Grid Forecasting: From Research to Application

Jin Xu

December 17, 2018
11:00 AM
Light Engineering room 250
Advisor: Prof. Yue Zhao

As the integration of renewable energy into the electricity network becomes increasingly prevalent, utilities and grid operators confront major challenges in maintenance and regulation stemming from the intermittent nature largely due to atmospheric interference. The significant improvements and falling costs of photovoltaic (PV) technology make solar energy a promising resource, yet the cloud induced variability of surface solar irradiance inhibits its effective use in grid-tied PV generation. Short-term solar forecasting is critically important for grid system stability and auxiliary power source management. Machine learning based framework is developed in this work for solar forecasting, from employing the use of sky imaging devices to irradiance sensor network. Image processing pipeline is designed, stochastic forecasting models and adaptive spatio-temporal forecasting model are developed and tested on large dataset from PV plants. The systematic study and validation shows significant improvement.

The online spatio-temporal forecasting framework is adapted and applied to electricity load forecasting in power grid as well, which allows for a heightened enhancement in grid operations, energy management, and planning. Load forecasting is historically based on aggregated spatial and temporal consumption data; with the deployment of Advanced Metering Infrastructure (AMI) systems, it can be achieved not only at a system level but also down to the consumer level. With this new increase in data, novel approaches and methods to load forecasting at a ‘big-data’ level is explored in this work. To deal with such colossal amounts of data, a stream processing forecasting framework is designed to spatio-temporally analyze these data while they are still on their way towards the data centers and provide the “on the wire” prediction.