A Case for Solar and BESS for Utility Savings

Published:

By Jeff MacKinnon, P.Eng.

I have been thinking about behind-the-meter (BTM) solar generation for years, basically since I was at a cement plant in Kansas. The image above was taken when I was on-site, there are acres and acres of limestone mine that isn't being used, and wasn't scheduled for reclamation for decades.

Two years ago I drafted a behind-the-meter solar landing page for developing these type of projects. However, since then I haven't promoted the service, or developed any background information on what this would look like.

This is the first post to rectify some of that.

BTM Project Setup

This project is based on a load profile for a modern manufacturing facility in a warm climate. Like most large manufacturing there is ample roof space for solar and we have assumed there is plenty of physical space for any modifications needed, including BESS.

Solar-Bess Power System

Existing Power System configuration

The existing power system has two 10MVA trains for powering two 12.47kV primary voltage circuits that feed three main areas. They are configured both in loops and primary selective depending on the phase that they were built.

I put together a composite demand for the facility for 2022 and using that information prepared the following load profile.

Power load for 2022

Annual Power Load

From this load profile I can see two specific trends. One that the load follows the temperature, starting in April and ending around October. The other is there is a obvious weekly trend.

Basis of Design

The basis of this design is the premise that with on site generation and storage can reliably reduce the utility expenses of the operation and provide energy for critical power systems.

For this design I've assumed that the critical power load is a constant 500kW, and that we will require at least 24h of storage for the situations where there is extended utility outages, coupled with low PV solar.

The first step is to figure out the correct size range for the solar system such that is meets the primary and secondary goals.

Limitations

The plant can't export into the utility. There are two reasons for this limitation:

  1. This arrangement makes the integration more complicated,
  2. The low load period is likely low for the entire system meaning that the rates that will be available to the plant will not be great.

Sample Project

For a project that includes PV Solar, we first need a site. For the first pass of these example projects I'm going to use Arizona. Its weather matches the natural seasonal load profile of the plant load which means that I don't have to do any shifting of the load data to make it work.

For our first pass I'm setting up the solar system to have 8MWdc of panels with a 30deg tilt, and to help flatten the production, and account for the temperature de-rating of the panels, 6.3MW of inverter capability. This is a drastic inverter under-size for most applications, but for this I think it makes more sense to ensure that the critical has capacity needed.

8000kWdc, 6300kWac

Expected PV Solar Generation

With the undersized inverter the generation profile is extremely flat over the year, with only a dip during the winter months.

PV Solar Generation vs Plant Demand

PV Solar Generation vs Plant Demand

There are two interesting areas, the winter lows and the summer highs. Looking at them a little closer we can get an idea about how much PV Solar generation will compare to the plant load, and from that how much storage will be needed to shift that generation.

PV Solar Generation vs Plant Demand - Winter

PV Solar Generation vs Plant Demand - Winter

PV Solar Generation vs Plant Demand - Summer

PV Solar Generation vs Plant Demand - Summer

Sizing the PCS

There are three roles for the storage system:

  1. The export power peak
  2. The excess demand peak
  3. The critical power load

This means that the PCS, the portion of the BESS that charges and dis-charges the batteries needs to be sized to the maximum export peak first. From the graphs above we can see that is in the winter months. To assist in saving on the utility bills, the storage should also be able to generate limiting the monthly peak to the optimal.

Note

I've calculated the optimal peak on a daily basis.

This has been defined as the demand load in kW where the total daily energy is the same as that of the recorded load.

Export Power Peak

This is relatively easy to calculate, its the minimum net plant load (PV Solar Generation - Plant Load). In this case the minimum net plant load is -5000kW.

The PCS will have to be able to absorb this generation.

Excess Demand

The maximum excess demand, on any month, is 2.2MW.

Critical Power Load

Both the export power peak and excess demand are an order of magnitude higher than the critical power load.

PV Solar Generation vs Plant Demand - Energy

PV Solar Generation vs Plant Demand - Energy

Sizing the BESS

The C-Rate is going to be the limiting factor for the speed at which BESS will be able to absorb or discharge. A lot of storage systems today are rated for 1C, however, for this example I want to assume a C-Rate of 0.5C. This example is very early in the total design process and we don't want to design something with really sharp pencils.

So, assuming a C-Rate of 0.5 the minimum kWh will be 2x the PCS size of 5MW, or 10MWh.

The size needed for the critical power system is 500kW for a 24h period, making this portion of 12MWh.

Next we need to confirm that the there is enough storage to shift any excess generation to the next day. This is 14MWh on the highest excess day.

PV/BESS and Potential Revenue

We now have the solar size, PCS and BESS specification.

  • Solar - 8000kWdc
  • PCS - 2.2MW
  • BESS - 14MWh

With this information we can calculate the utility savings. Since this example project is in Arizona I have used Arizona Public Service (APS) tariffs. This gives me the following costs to work with:

  • Demand Charge = $20 /kW
  • Energy Charger = $0.04421 /kwH

Excess Demand Savings

The excess demand charges have been calculated by picking the day each month with the maximum demand versus the optimal demand for that day. This will be the maximum demand charge that can be saved.

For this example, this results in a total savings of $327,303 per year.

Energy Savings

Since this project assumes that all of the generated energy will be absorbed by the system, this means that the total generation will offset the energy bought from the utility.

The total generation is expected to be 15GWh resulting in a saving of $669,410

The total revenue savings is expected to be approximately $990,000 per year.

Capital Cost

The expected capital cost for this system is likely to be in the range of $18m - $20m [*]. This total is only for the system to the plant primary voltage power system, no power system configuration or integration costs.

Smaller System may be better

It doesn't leave a great payback, so I think I have over-sized the solar system. I think a smaller PV Solar system will provide all the benefits, with a smaller capital expense, the cost of the BESS is driving the capital cost.

I will write up this smaller system next.

Footnotes
[*]We have calculated the capital cost using a total solar cost of 1300 USD/kW this is based on the numbers from fastercapital.com and verified by a couple of PV developer friends. The BESS cost is likely high at 500 USD/kWh according to a couple suppliers, but by using this capital cost this high will likely cover the power system integration, and other costs associated with a project like this. This cost was referenced from exencell.com.