Demonstrating Electrical Model-Free Voltage Calculations with Real Smart Meter Data

Voltage calculations are key for most distribution network analyses. However, the challenge in low voltage (LV) networks is that electrical models are not readily available and, therefore, accurate voltage calculations (via power flows) are not possible. Alternatively, regression methods can be used to capture the relationships among the historical smart meter data of customers (P, Q, and V) and the corresponding LV network. This paper improves a previously proposed methodology to carry out electrical model-free voltage calculations based on Neural Networks (NNs) so it can be used with real smart meter data. The effectiveness of the improved methodology is demonstrated using data from 6 Australian LV networks. The results show that the improved methodology significantly reduces the computing time required to produce the NN, making it an accurate and extremely fast alternative for distribution companies to calculate voltages without electrical models.