Performing PV system feasibility studies correctly

Learn how to conduct a feasibility study before specifying a solar photovoltaic (PV) system for a building.
By Javier M. Abarca, PE, Stanley Consultants, Centennial, Colo. November 2, 2017

Learning objectives

  • Know when solar photovoltaic (PV) systems are an option for the building client.
  • Learn how to calculate the feasibility of specifying PV systems for a building.
  • Calculate return on investment for alternative-energy PV systems.

Figure 1: Solar photovoltaic (PV) system feasibility studies can be a great tool if done correctly. Courtesy: Stanley ConsultantsSolar photovoltaic (PV) system feasibility studies can be a great tool if done correctly (see Figure 1). Many clients would like to reduce their overhead by reducing energy consumption, and it is easy to assume that the bigger the solar PV system, the lower the energy cost will be. Unfortunately, this isn’t necessarily true. Sometimes, smaller results can lead to a better outcome.

The right research and fact-finding efforts are important functions that contribute to a well-done study. Typically, the return on investment (ROI) is where the data comes together to give a true picture of the possible outcome. For many clients, the ROI part of the feasibility study is the deciding factor.

Feasibility studies can be broken into three main parts: investigation, mandated studies, and ROI. Although there are exceptions, almost all studies include these three components. The ROI is where all the pieces of the investigative puzzle come together.

Investigation

There are comparable issues that must be addressed with every study, including client energy goals, the energy-usage profile, understanding how the energy bill is calculated, connection agreements, restrictions, and installation locations. While investigating the details, it is important to keep the following question in mind: “Can a solar PV system meet the outcome the client wants to achieve?”

Example 1. An office-building owner has a building where 90% of the office space is leased. The owner would like to add a solar PV system to the rooftop to reduce electrical bills. The building has 34,000 sq ft of available roof space to mount solar panels, and the owner would like to install a system that can use as much of the roof space as possible. Because the goal is to reduce the cost of electricity, the study should be focused on electrical energy savings and not just on the effects of a solar PV system.

Example 2. A military entity has lots of open space and wants to install a solar PV system with battery backup. The base is in the desert, where extreme heat requires significant cooling of building electrical loads. The military wants to supply its entire 25-MW electrical load with a solar PV system.

The preliminary estimated solar PV size for Example 2 is about 125 MW. The solar PV has been around for many years and can easily be calculated and estimated. However, energy storage of this magnitude has not revealed itself to the solar utility market. The main thing that stands out, in this case, is the enormous size of energy storage requested. The feasibility study can lead to researching electrical storage at utility-scale sizes. Many dedicated hours would have to be dedicated to research and estimating the possible cost and lifecycle of an energy-storage solution.

As with these scenarios, client goals and needs should be addressed. These variables will lead to various paths in determining if solar is a viable solution for the client. Not only are the clients’ needs important to understand, but so is the energy demand curve profile.

Understanding the energy demand curve profile

Understanding the energy profile is a critical component of a feasibility study. In many cases, most clients do not have information on their energy usage. Most only have the utility bills, which only give a partial picture of how the load is used. A key component is knowing when and how the heaviest electrical loads occur throughout the day.

Some load curves only track the kilovolt amps or kilowatts. But, just knowing those parameters is not enough. Measurements, such as the kilovolt-ampere reactive, kilovolt amp, and power factor, are critical bits of information. They give a picture and understanding of the electrical energy usage profile. These variables can determine how a solar PV system will react to the current power curve and billing statement. The effect that a solar PV system has on the power factor, kilovolt amps, kilowatts, and kilovolt-ampere reactive can have a negative or positive impact on reducing energy cost. It can sway the viability of installing a solar application. By using this information, we can determine if a larger PV system can make the cost of electricity worse or better. As part of understanding the energy load profile is also understanding how the utilities apply charges to these load profile variables.

Understanding how the utility bill is calculated

Understanding the commercial utility bill is another important part of the investigative process. The key is in knowing how, when, and at what level of load application is best suited to manage the reduction of energy bills. The fees charged by the utility can either make or break the feasibility of a PV system.

Every utility company has its own way of applying electric rates. In the commercial billing world, the bills can be complicated; it takes someone who really has a good grasp of power factor. There are typically two main charges that are found on the commercial billing system: one is the per-kilowatt hour charge and the second is demand.

The kilowatt hours charge is the easiest to understand, as it is a straightforward calculation. Basically, this is the amount of actual electricity that is used during a billing cycle. There are no underlying calculations associated with this variable. It is typically charged on an amount per unit, or something like $0.03411/kWh. However, with demand charges, there are several parts involved with the calculation.

The demand charge is more complicated because the power factor ratio comes into play. Demand charges are set up to cover the cost for running peaker unit generators. When high electrical load is drawn from the power grid, peaker units are used to produce additional power to cover load demand. The cost of running a peaker unit generator is relatively expensive when compared against base load unit generators.

A typical demand charge is applied by multiplying the peak measured kilowatts by a fee. Most commercial buildings’ demand fees can be as high as 50% of the electrical bill. So, let’s say that a typical bill might charge $13.02 per kilowatts of measured demand. Typically, the demand is determined by the maximum kilowatts in a 15-minute period of average actual peak kilowatts measured during a billing cycle.

For example, if during the entire billing cycle, the peak was below 100 kW and there were only one 15-minute peak of 200 kW, the demand charge would be 200 kW, doubling the demand charge. As an example, if a building’s electrical load curve profile peaks out in the evening instead of high noon, the solar PV system might not be as effective at reducing the energy cost.

There is also a power factor penalty that is tabulated into the demand charge. Utilities will typically charge additional fees for having a low power factor. Power factor is a ratio of kilowatts divided by kilovolt amps. What does all this mean? Without getting into writing a long dissertation, there are two forms of power. One is kilowatts, known as real power, and the other is known reactive power, or kilovolt-ampere reactive. Kilovolt-ampere reactive is power loss that is typically caused by inductive loads found in things such as electric motors and transformers.

In simple terms, power factor basically tells us how much power is real power, kilowatts. Ideally, the closer the power factor is to 1, the more real power and less reactive power is being used. A power factor of 1 means that all the power being used is efficient. A smaller power factor equates to less efficient use of power. This is because the smaller power factor requires more current to run the same load. Thus, it is best to try and keep the power factor as close to 1 as possible. The utilities are very aware of this and must provide opposing reactive power to counteract the inefficiency. The utilities incur extra cost to correct the power factor and pass this cost to the customer. Most utilities set up a power factor threshold. Any time the average power factor drops below this threshold, power factor penalties will apply.

The power triangle on the left shows the facility had a starting power factor of 93% with a reactive power of 124 kVAR and real power of 309 kW. How does this relate to solar PV? Solar PV produces real power kilowatts with a power factor close to 1. On the customer side of the meter, solar PV systems displace the amount of real power that the utility is supplying. Keep in mind that power factor is a ratio of real power being used. So, if we look at the power factor from the utility side of the meter, the power factor would decrease. If the decrease is enough to pull the power factor below the threshold, a penalty demand charge would be added to the bill. The more real power the PV system displaces, the lower the power factor could be, thus increasing the power factor penalties (see Figure 2).

From the utility side of the meter, the utility would be suppling less real power. And the kilovolt-ampere reactive would remain constant. The meter now measures higher reactive power relative to the real power. From the perspective of the utility, extra measures have to be taken to counteract the lower power factor. So, depending on the power factor, before installing a PV system, the building could have enough power factor margin to avoid power factor penalties.

There is another power factor threshold below the penalty threshold. Utilities also have a minimum allowable tolerable power factor that must be met. If any connected electrical system drops below the minimum power factor, the utility could potentially disconnect power until the power factor problem is corrected. The solar PV system may not pull the power factor down to this level; it is just some information that must be considered. Armed with this information, solutions could be used, such as installing a capacitive bank to help raise the power factor.

Knowing the way utilities charge for electricity can help in producing a better study and assist in determining the best solution that fits the client’s needs. These are not the only factors that contribute to the feasibility study; they are just a few key points. There are other factors that also have a high influence on the bottom line.

Figure 3: The graph shows the relationship between the size of the PV system versus the ROI payback. It plots PV system sizes from 5 kW to 80 kW. Courtesy: Stanley ConsultantsConnection agreements

Solar PV interconnection agreements can drive up the cost of solar installations. Some utilities have extensive testing requirements under IEEE 1547.1-2015: IEEE Standard Conformance Test Procedures for Equipment Interconnecting Distributed Resources with Electric Power Systems. Some of these testing requirements can add substantial cost to the solar PV system. Thus, it is important to consider the costs of required tests into the cost of installing a commercial solar system. This can impact the ROI and payback period.

Connection agreements also describe how the utility will handle charges or credits for excess electricity produced. These factors can add up to a good feasibility study. Other crucial factors are the unique circumstances of each site.

Site-specific requirements

Many site-specific requirements also must be addressed in feasibility studies. In Example 2 mentioned earlier in this article, the military base also has restrictive areas that must be addressed. Military bases and air fields have required clear zones, which restrict the kind of structures that can be built in these areas. In this example, the military has plenty of space to install a utility-scale system. However, all of the available open spaces have some extensive low-lying flood-susceptible areas. In this situation, the cost of installing a PV system is extensive, due to the amounts of fill required to raise the level of the land area.

Mandated studies

In a few cases, special studies are required before a solar system can be installed. The Federal Aviation Administration requires a glint-and-glare study to be performed on any solar project within an airfield property. The analysis determines how the glare from the solar panels could affect aircraft and control towers along the landing and departure flight paths. The study should determine the intensity and time of day the glare can be most disruptive. Sometimes, a simple degree of change in tilt angle or compass direction can solve a potential problem before installation. Identifying this issue upfront can prevent additional cost from having to changing the angles after the fact.

Return on investment

This is where it all comes together. All the information that has been discovered from the investigation can now be put together and develop a big-picture scenario through the ROI.

When calculating a return on investment, several points must be considered. Many times, an ROI is shown without a time frame or payback period. Without the payback period, it is difficult for the client to make an evaluation and determine the financial risks. So, one of the best ways to present this is to show a spreadsheet that depicts the present value as time goes on.

As an example, several ROI scenarios are typically created to demonstrate two things. First, how a graph derived from a spreadsheet with the right information can serve a good ROI analysis. Second, how the utility billing can be affected by the size of the solar PV system. Below are examples of two identical buildings that are located across the street from each other. Both have the same electrical loads except for power factor (PF).

ROI examples

Case 1 building parameters:

  • PF = 91%.
  • Monthly = 72,000 kWh.
  • Peak demand load = 300 kW.
  •  Roof space can support a 200-kW PV system.

Case 2 building parameters:

  • PF = 93%.
  • Monthly = 72,000 kWh.
  • Peak demand load = 300 kW.
  • Roof space can support a 200-kW PV system.

ROI specifications

Many things were considered in this ROI. We considered installation cost, maintenance cost, electrical cost increase, solar panel degradation over time, and many other factors. One of the most important factors was the effect of demand charges and power factor. As you look at the plots, you will see how significant a power factor can be. In many cases, however, power factor is left out of a study.

Figure 3: The graph shows the relationship between the size of the PV system versus the ROI payback. It plots PV system sizes from 5 kW to 80 kW. Courtesy: Stanley Consultants

Various solar PV size plots were developed from this spreadsheet after running several scenarios. The first thing plotted was the relationship between the size of the PV system versus the ROI payback (see Figure 3, which plots PV system sizes from 5 kW to 80 kW). The first issue that emerges is that any size below 20 kW has a payback period of about 7 years. Anything greater than 20 kW has a payback jump to 18 years or more. The cause is a power factor penalty. In both scenarios, the utility charges a penalty for anything that is below a PF of 90%.

Here, the building is sitting on an average power factor of 91% before solar. This building only has a 1% margin between the power factor penalty and its current power factor. As we discussed in the billing section, the more real power from the solar is applied, the lower the power factor will be. In this case, a solar PV smaller than 20 kW did not generate enough real power to pull the power factor below the PF penalty. Thus, it can operate within the power factor margin. To keep the payback period to 7 years, the largest PV system that could be installed is a 20-kW system, otherwise the payback could reach 18 years. Now let’s compare this to Case 2 where the initial power factor is at 93% before the solar PV is installed (see Figure 4).

In Case 2, the building has a PF margin of 3%. In Figure 4, a PV system of up to 55 kW can be installed while maintaining a 7-year payback. Any PV size above 55 kW could have a payback of 12 years. The extra PF margin can handle a larger PV system without dipping into the PF penalty (see Figure 5).

In Figure 5, a 20-kW system is installed on both buildings. Each building starts with a different power factor margin. The same system installed on the twin building will have different payback periods. The Case 1 building’s payback is 7 years as compared with Case 2’s 20 years. In Figure 5, notice how in Case 1, the 20-kW system pulled down the power factor below the threshold. While in Case 2, the power factor stayed above the threshold.

Figure 5: The graph shows that the extra power factor margin can handle a larger PV system without dipping into the power factor penalty. Also, in Case 2 the ROI starts getting more expensive. It is only until the 20-year point is reached that the ROI begins to improve. Why is this? As the PV panels degrade over time, the PV system produces less real power. This, in turn, causes the power factor to begin to increase. Eventually, after about 12 years, the panels degrade enough to allow the power factor to increase above the PF penalty point. At this point, penalty fees are no longer charged and the PV system can begin to recoup the installed cost. For these reasons, a bigger PV system is not always the answer. It just depends on the circumstances.

In general, it is easy to see how the right information can produce the right ROI and show a much clearer overall picture.

When performing a solar system feasibility study, pay attention to the client’s goals and gather the right kinds of information including unique aspects of a facility. Understand how the utility charges are assessed. Lay out a good ROI spreadsheet with charts, which adds to the quality and overall picture portrayed by the ROI. And, of course, ask and answer the ultimate question, “Does this meet the needs of the client?”


Javier M. Abarca is a control system engineer at Stanley Consultants. He has 13 years of experience designing and developing power plant control systems and conducting feasibility studies for photovoltaic systems.