Peak shaving: the solutionto reduce the peak demand charge component of energy bills

Peak shaving: the solution
to reduce the peak demand charge component of energy bills

Management of peak demand through an automatic load-activation system enables better planning and management for the entire electricity grid, generating an immediate benefit in the form of lower costs on bills and reliability of electricity supplies.

Peak shaving analysis enables assessment of the impact of peak demand charges on company energy spending, calculation of energy spending due to specific peaks in consumption and the proposal of a series of strategies to level out peaks and reduce costs.

Peak-demand charges on bills represent 5–20% of the total and are calculated based on the maximum power peak recorded within quarter-hour slots of monthly consumption. This means that it only takes one quarter-hour period in the month (1 out of 2,976) with a figure above the average value to end up with a higher bill

The annual cost for the peak demand charge component varies depending on the incentive category of power-intensive users:

MT – Users with P>500 kW50,40642,63940,45238,26534,913

How can peak-demand charges be reduced while maintaining production levels high?

With over 25 years of experience, Energy Team offers a comprehensive solution aimed at reducing at least 8–15% of the power peak

Reducing the power peak by just 200 kW corresponds to a saving of approximately € 10,000 per year

The 4-steps solution:


Creation of company power-usage models and individual loads that contribute to generating the peak. The model is created using software specially developed to correlate different parameters contributing to overall company consumption. This utilises a data science approach to simulate real effects.


The model created, with the addition of the parameters of maximum disconnection time for each user, enables assessment of the behaviour of the load-management system based on the power targets, simulating and evaluating all possible scenarios with complete security.


Once all of the possible activation scenarios have been created, the model is used to analyse possible power thresholds. A cost-benefit analysis is carried out for each scenario in order to generate a significant saving with minimal impacts on the process.


Optimised configuration of the load-management system to control loads selected for activation and triggered by the management software. Control logics identified as optimal during modelling are entered into the software, and are updated monthly based on load forecasts.

Here is an example

We took the example of a FAT.1 iron foundry with a monthly power peak of approximately 3 MW and peak-demand charges totalling € 128,000.

Focusing on the greater loads present, the choice is made to analyse the two induction melting furnaces present at the facility, developing an optimised activation logic so that disconnection of the furnaces exploits the power increments and does not impact production, setting the maximum disconnection times agreed with the foundry.

Different activation scenarios are created, which undergo cost-benefit analysis, identifying a configuration that generates a saving of approximately € 20,000 per year, altering overlapping loads by a few minutes, avoiding unjustified charges. This outcome was achieved by managing loads for a few minutes in only 50 quarter-hour slots per year (50 out of 35,040 = 0.14% quarter-hour slots/year).

Having defined the implementation architecture, Energy Team supports the customer with a consulting service aimed at identifying the optimal activation threshold for each month of operation.

This threshold is calculated monthly on the basis of historical performance and forecasted production for the following month, in order to optimise the effectiveness of the system and maximise savings.

The system is engineered entirely by Energy Team and enables maximum flexibility in the implementation of load-control logics.

  • Instant reading of power usage through synchronisation with the distributor’s meter.
  • Creation of planned power-usage curve.
  • Management of loads if the available energy is to be exceeded in the integration period analysed.
  • Constant fine-tuning of the optimal power threshold and load-modulation logics.