Productive and commercial demand

Productive and commercial energy markup definition

In the M-LED framework, the electricity demand induced by small-scale productive and commercial activities (such as barber shops, minimarkets, or telecommunication points) is estimated by approximation.

First, a composite index based on the productive activities drivers and energy use is constructed based on road density, employment levels and wealth distribution, and city accessibility proximity is built. The indicators are aggregated using a principal component analysis (PCA). The analysis is carried out in an R scientific computing environment in the .\residual_productive.R file. Input data to the PCA are found in the .\MLED_database\input_folder.

The PCA outcome is rescaled to the 0.3 and 0.6 range (following Moner Girona et al. 2019) to create a bottom-up mark-up factor on top of the residential demand. The 0.3 - 0.6 range is readily varied within the .\residual_productive.R file.

Load curve and seasonality definition

The baseline load curve (share of demand at each hour of the day over the total daily demand) of micro productive activities is assumed to follow the same path of that described in Moner Girona et al. 2019, which in turn relies on ground-metered data from mini-grids in Kenya.

A standard load curve functional form is assumed for crop processing facilities at each cluster. This is reported in the .\MLED_database\input_folder\productive profile.csv file, where each number expresses the share of total daily crop processing electricity consumption at each of the 24 hours of the day.

Finally, a seasonal variation is imposed to the monthly demand loads curves: in particular, the seasonality follows the same monthly mark-up observed in the residential demand across months of the year from the RAMP modelling.

Demand to clusters parsing

Algebraically, the final sectoral demand CommProd_imh (where i, m, and h, identify demand clusters, months of the year, and hours of the day, respectively) is expressed as CommProd_imh=(1+PCA_i^range)×Residential_imh×CommPro〖dCurve〗_mh

where: CommProd_imh is the commercial and productive demand at each cluster i at each month of the year m at each hour h; PCA_i^range is the result of the PCA at each cluster rescaled to the 0.3-0.6 range; Residential_imh is the residential demand at each cluster i at each month of the year m at each hour h; CommPro〖dCurve〗_mh are the twelve month-specific hourly curves for the sectoral demand derived from Moner Girona et al. 2019 and adjusted for the seasonality based on the residential seasonality variation.