What is peak shaving?


In the energy industry, peak shaving refers to leveling out peaks in electricity use by industrial and commercial power consumers. Power consumption peaks are important in terms of grid stability, but they also affect power procurement costs: In many countries, electricity prices for large-scale consumers are set with reference to their maximum peak-load. The reason is simple: the grid load and the necessary amount of power production need to be designed to accommodate these peak loads.

Peak shaving vs. Load shifting

With peak shaving, a consumer reduces power consumption ("load shedding") quickly and for a short period of time to avoid a spike in consumption. This is either possible by temporarily scaling down production, activating an on-site power generation system, or relying on a battery.

In contrast, load shifting refers to a short-term reduction in electricity consumption followed by an increase in production at a later time when power prices or grid demand is lower. Dedicated generators or electricity storage facilities owned by the power consumer can be used to bridge high-price or high-load phases, but play less of a role if production will eventually catch up again.

Peak loads and grid usage fees

Peak loads are not popular with grid operators; they must design the grid based on the maximum amount of power that will be needed. Nevertheless, everyday operation at many industrial companies – such as powering up or increasing a production process – can cause fluctuating loads on the grid. It is possible to reliably detect the source of a sudden load increase by monitoring power consumption. Depending on the grid operator, these peaks are used to calculate grid usage fees assessed to certain power consumers. The following example illustrates how these additional grid fees are calculated for a medium-sized company in Germany.

Calculation example

A company has a constant load of 4,000 kW throughout the year without peak loads. The company pays a fixed annual grid fee, which is assessed per kilowatt. In this example, this is 50 € per kW: 4,000 kW x 50 € = 200,000 € per year in grid charges. A special production order causes an exceptional peak load of an additional 500 kW, which lasts for just 30 minutes. The grid fee increases immediately, with additional costs of 25,000 € based on 4,500 kW of annual consumption. This is just to cover grid usage and does not include the cost of electricity utilized by the company.

Practical application of peak shaving

As the example shows, in some cases peak shaving makes sense to avoid peak loads and to keep grid usage costs low.

Peak shaving load control (demand-side management), power storage, and generation

When it comes to managing peak loads, there are a few different approaches. “Peak load monitors” track and regulate a pre-defined peak load every quarter of an hour. If the monitor predicts that the accumulated peak load will exceed a certain threshold in the next quarter-hour interval, certain power consumption processes are reduced. This allows a company to determine and influence its maximum power consumption.
If reducing load is not desired or possible, a company can provide its own supplemental power to avoid peak loads. Additional power could come from sources such as the company’s own electricity storage facilities or CHP plants. This creates a time-limited provision of power from the electricity storage facilities and/or a generator within the company’s grid, which absorbs the additional peak load at the transfer station before it reaches the public grid.

Peak shaving, energy turnaround, and flexibility

For distribution network operators, peak shaving is a good way to keep the costs of network expansion low. An efficiently-operating network requires less copper installation in the form of power lines and distribution points. Uniform power generation and consumption is the ideal scenario, leading grid operators to create an incentive for reducing peak loads, especially in light of increasingly volatile feed-in from wind and photovoltaics.