Baseline Calculation of Industrial Factories for Demand Response Application

Baseline calculation is one of the most important issues in demand response program. Accurate baseline calculation means fair payment to the customers. This paper proposes a method that is based on the Levenberg-Marquardt algorithm of neural networks for such calculations. Its performance is compared with two well-known methods: (1) Linear Regression Analysis, and (2) Polynomial Regression Analysis. We have shown that the most accurate result of the baseline in the sampled industrial factory is neural networks using the Levenberg-Marquardt algorithm.