Adequate capacity planning of substations
and feeders primarily depends on an accurate estimation
of the future peak electricity demand. Traditional coinci-
dent peak demand estimation is carried out based on the
empirical metric, after diversity maximum demand (ADMD),
indicating individual peak consumption levels and of de-
mand diversification across multiple residents. With the
privilege of smart meters in smart cities, this paper pro-
poses a data-driven probabilistic peak demand estima-
tion framework using fine-grained smart meter data and
socio-demographic data of the consumers, which drive
fundamental electricity consumptions across different cat-
egories. In particular, four main stages are integrated in
the proposed approach: load modeling and sampling via
the proposed variable truncated R-vine copulas (VTRC)
method; correlation-based customer grouping; probabilis-
tic normalized maximum diversified demand (NMDD) esti-
mation; and probabilistic peak demand estimation for new
customers. Numerical experiments have been conducted
on real demand measurements across 2,639 households in
London, collected from Low Carbon London (LCL) projects
smart-metering trial. The mean absolute percentage error
(MAPE) and pinball loss function are used to quantitatively
demonstrate the superiority of the proposed approach in
terms of the point estimate value and the probabilistic
result, respectively.