A Data-Driven Machine Learning Approach for Consumer Modeling with Load Disaggregation
This work addresses the need for accurate load disaggregation in applications like distribution planning and demand response, but it is incremental as it builds on existing semiparametric models with a hybrid method.
The paper tackled the problem of load disaggregation into fixed and shiftable components for consumer modeling, proposing a two-stage machine learning approach that uses NMF and GMM for disaggregation and L2-norm regression for parameter estimation, with validation on actual residential energy data.
While non-parametric models, such as neural networks, are sufficient in the load forecasting, separate estimates of fixed and shiftable loads are beneficial to a wide range of applications such as distribution system operational planning, load scheduling, energy trading, and utility demand response programs. A semi-parametric estimation model is usually required, where cost sensitivities of demands must be known. Existing research work consistently uses somewhat arbitrary parameters that seem to work best. In this paper, we propose a generic class of data-driven semiparametric models derived from consumption data of residential consumers. A two-stage machine learning approach is developed. In the first stage, disaggregation of the load into fixed and shiftable components is accomplished by means of a hybrid algorithm consisting of non-negative matrix factorization (NMF) and Gaussian mixture models (GMM), with the latter trained by an expectation-maximization (EM) algorithm. The fixed and shiftable loads are subject to analytic treatment with economic considerations. In the second stage, the model parameters are estimated using an L2-norm, epsilon-insensitive regression approach. Actual energy usage data of two residential customers show the validity of the proposed method.