Combined Heat and Power Partnership
Calculating Reliability Benefits
Rather than install an unproductive diesel backup generator to provide outage protection, a facility can design capability into a combined heat and power (CHP) system that provides electric and thermal energy to the site on a continuous basis, resulting in daily operating cost savings.
- Reliability Issues and Frequency
- Power Sensitive Loads
- Problems with Traditional Emergency Generators
- CHP as a Reliability Solution
- Estimating Costs of Outages
Power reliability is a critical issue for many customers, representing a quantifiable business, safety, and health risk to their operations. These risks often compel customers to install backup or emergency diesel generator sets, tying up significant capital in rarely used assets that require periodic maintenance and frequent testing. However, even these measures are not foolproof, as shown during the Northeast black-out of 2003, when half of New York City's 58 hospitals suffered failures in their backup power generators.1
CHP can be a reliable and cost-effective alternative to installing unproductive backup generators to provide protection against extended outages. A CHP system is typically selected for a facility due to its ability to reduce operating costs and overall emissions. However, power outage protection can also be designed into a CHP system that efficiently provides electricity and thermal energy to the site on a continuous basis. CHP systems can be configured in a number of ways to meet the specific reliability needs and risk profiles of various customers, and to offset the capital cost investment for traditional backup power measures.
A key step for a customer considering a potential investment in CHP as a solution to reliability concerns is to identify and quantify the value of reliable power to their operations and compare these costs to those associated with configuring CHP to include outage protection.
Reliability Issues and Frequency
Reliability is often defined as how often and how long electric power service is interrupted. Service interruptions and variations in power quality can happen at any time. Although most grid outages are momentary occurrences that are generally brief and do not adversely impact anyone other than the most sensitive operations, an average facility can expect to experience an extended outage (lasting more than five minutes) every other year.2
In disaster situations, power outages can have dramatic effects. During the blackout of 2003, portions of the Midwest, Northeast, and Ontario, Canada were without power for up to four days in some locations. Total losses related to the power outage topped $10 billion, and affected more than 50 million people.
The cost of a service interruption varies by customer and is a function of the impact of the interruption on the customer's operations, revenues, and/or direct health and safety. As an example, Pacific Gas & Electric Company (PG&E) researched the estimated direct costs of outages to their customers (based on a combination of direct cost measures and willingness-to-pay indicators) and showed that the value of service can vary widely by customer class (Table 1).3 PG&E estimated the total annual cost of power outages to its customers at $79 billion per year.4
Table 1. Estimated Direct Costs of Outages for PG&E Customers
|Customer Class||$/kWh unserved|
|Industrial||$12.70 - $424.80|
|Commercial||$40.60 - $68.20|
|Agricultural||$11.50 - $11.70|
|Residential||$5.10 - $8.50|
Power Sensitive Loads
For certain types of customers, reliability is a true business and operations issue, rather than merely an inconvenience. These customers cannot afford to be without power for more than a brief period without significant loss of revenue, critical data/information, operations, or even life.
Some particularly power sensitive customers include:
- Mission-critical computer systems
- Industrial processing companies
- High-tech manufacturing facilities and clean rooms
- Financial institutions
- Digital communication facilities (phone, television, satellite)
- Military operations
- Wastewater treatment facilities
- Hospitals and other health care facilities
Problems with Traditional Emergency Generators
There are at least five notable drawbacks to using diesel gen-sets, which are typically employed as backup generators:
- Backup diesel generators are rarely called to operate and might not start and run when needed. Unless a facility keeps up with maintenance and frequent testing, emergency generators can fail to start on the rare occasions they are needed.
- Diesel fuel deliveries can be difficult or impossible to arrange during a widespread disaster. During a major hurricane or regional blackout when a prolonged outage occurs, a diesel backup system might have to shut down due to lack of fuel.
- Storing large quantities of fuel imposes high costs and risks of fuel leakage or fuel degradation. Diesel fuel begins to chemically break down within 30 to 60 days of delivery and tends to absorb moisture from the air. These fuel quality issues can lead to unreliable engine operations and higher maintenance costs if fuel storage is used to hedge against potential shortages.
- Diesel engines used for backup service typically have high emissions and are permitted for limited use. Having limited permitted hours for operation makes it difficult to keep the engines in the proper state of readiness and prevents generator use for meeting general facility energy needs or reducing operating costs.
CHP as a Reliability Solution
Rather than install a diesel backup generator to provide outage protection, a facility can design capability into a CHP system that provides electric and thermal energy to the site on a continuous basis, resulting in daily operating cost savings (Figure 1). In this type of configuration, the CHP system would be sized, as normal, to meet the base load thermal and electricity needs of the facility. Supplemental power from the grid would serve the facility's peak power needs on a normal basis and would provide the entire facility's power when the CHP system is down for planned or unplanned maintenance. However, the CHP system would also be configured to maintain critical facility loads in the event of an extended grid outage. In order to operate during a utility system outage, the CHP system must have the following features:
Figure 1. CHP System with Backup Responsibility for Critical Loads
- Black start capability. The CHP system must have a battery-powered starting system.
- Generator capable of operating independently of the utility grid. The CHP electric generator must be a synchronous generator, not an induction generator that requires the grid power signal for operation. High frequency generators (microturbines) or direct current (DC) generators (e.g., fuel cells) need to have inverter technology that can operate independently from the grid.
- System integration with load shedding. The facility must match the size of the critical loads to the capacity of the CHP generator. These loads must be isolated from the rest of the facility's noncritical loads, which must be shut down during a grid system outage using appropriate switchgear and control logic. The critical load isolation approach can be manual or automatic and can be configured to incorporate dynamic prioritization of load matching to the CHP system capacity.
The additional costs for switchgear and controls for a CHP system depend on the level of control and the speed with which the facility needs to have the CHP system pick up the critical loads in the case of a utility power outage. Table 2 describes three levels of protection—manual, automatic, and seamless—and site-specific costs for reconfiguring the site wiring and control panels to isolate and serve the critical load. The level of backup capability and control chosen for a CHP system will be directly tied to the value of reliability and risk of outages for the customer.
Manual control requires an operator to isolate the generator to the emergency circuits using manual transfer switches. An automatic transfer switch eliminates the need for operator intervention. The generator is switched to the emergency circuit automatically, a process in which the circuit is open for only a fraction of a second (5-10 cycles). Seamless transfer—most often integrated with a full uninterruptible power supply (UPS)—utilizes a more costly, closed transition, automatic transfer switch with bypass isolation. This switch is a "make-before-break" design that momentarily parallels the two circuits before switching. An isolation bypass switch allows removal of the automatic switching mechanism in the case of failure, with the ability to then manually switch the load.
Table 2. Control Costs for Generator Backup Capability5
|Control Level||Time to Pick Up Load||Equipment Required||Capital Cost|
|Manual||Up to an hour||
||$20-$60 per kW|
|Automatic||5 to 10 cycles when running||
||$25-$105 per kW|
|Seamless||¼ to ½ cycle when running||
||$45-$170 per kW|
|Reconfiguring for Load Shedding||Not applicable||As needed by the site:
||$100-$500 per kW|
Note: Cost range figures represent estimates for a 500 kW CHP system at the high end and a 3,000 kW CHP system at the low end. Cost estimates do not include recircuiting costs, which depend on site needs.
Estimating Costs of Outages
Traditionally, facilities have perceived that it is difficult to quantify the value of reliability to their operations; however, to justify the added costs of configuring a CHP system to provide stand-alone power, the value of reliability must be determined as a factor in the feasibility analysis. At least two different approaches can be used to estimate the value of reliability:
1. Estimate the direct costs of service interruptions based on experience. Customers pay for electricity based on the utility cost of service. While the cost of service determines the electric rates, the value of that service is different for each customer. When power delivery is disrupted, customers generally experience losses to their operations that are much greater than the cost of the electricity not delivered. The value of these losses can be referred to as the customer's value of service (VOS). VOS can be measured in terms of the direct costs of an outage. Power outages or service interruptions can impose direct costs on customers in a number of ways:
- Damaged plant equipment
- Spoiled or off-spec product
- Extra maintenance costs
- Cost for replacement or repair of failed components
- Loss of revenue due to downtime that cannot be made up
- Costs for idle labor
- Liability for safety/health
Some customers can determine their VOS—the direct costs of outages—by reviewing recent outage history and estimating an annualized cost of outages to their operations. One approach is to quantify the direct cost impacts of momentary outages (less than 10 seconds) on either a dollars per incident or dollars per minute basis if the momentary outage results in an extended disruption at the facility, and to similarly quantify the direct cost impacts of extended outages (greater than 10 seconds) on a dollars per minute or dollars per hour basis. Estimates of typical annual values for the number of momentary outages and total time of extended outages can be determined by reviewing utility bills and/or facility records. The resulting cost value represents an annual direct operating cost that could be avoided with a properly configured CHP system and would be treated as operating savings in a CHP feasibility analysis. Dividing this total cost value by the number of unserved kWh (average power demand in kW times total annual outage time in hours) produces a value of service estimate similar to those included in Table 1.
Table 3 presents an example of how to quantify the cost of facility disruptions due to both momentary and long-term outages. The number of occurrences in this example is based on electric industry survey data6. The disruption caused by a particular type of outage is customer specific. In this example, even momentary outages cause extended disruption to plant operations (30 minutes), as would be the case where production is controlled by programmable logic controllers that need to be manually reset after an outage. The cost of an outage for this customer is estimated at $45,000 per hour of disruption based on operating history. Assuming an average plant power demand of 1,500 kW, the value of service is estimated to be $30/unserved kWh; this is towards the lower range of outage costs for industrial customers as shown in Table 1.
Table 3. Value of Service Direct Cost Estimation and CHP Value
|Facility Outage Impacts||Annual Outages||Annual Cost|
|Power Quality Disruptions||Outage Duration per Occurrence||Facility Disruption per Occurrence||Occurrences per Year||Total Annual Facility Disruption||Outage Cost per Hour*||Total Annual Costs|
|Momentary Interruptions||5.3 Seconds||0.5 Hours||2.5||1.3 Hours||$45,000||$56,250|
|Long-Duration Interruptions||60 Minutes||5.0 Hours||0.5||2.5 Hours||$45,000||$112,500|
|Unserved kWh per hour (based on 1,500 kW average demand)||1,500 kWh|
|Customer's Estimated Value of Service (VOS), $/unserved kWh||$30 /unserved KWh|
|Normalized Annual Outage Costs, $/kW-year||$113 $/kW-year|
* Outage costs per hour estimated based on facility data and include production losses, increased labor, product spoilage, etc.
2. Estimate the willingness-to-pay to avoid loss of service. Because outages occur infrequently at different times and last for different durations, it is sometimes difficult to determine the annualized cost of outages. In this situation, customers can use an alternative measure to estimate the value of reliability—their willingness-to-pay to avoid loss of service. If customers invest in backup power generators, a second utility feed, power conditioning equipment, or UPS, these costs represent their willingness-to-pay to avoid an outage and therefore, represent an approximation of how much they value reliable electric service. The costs of these measures (e.g., the capital and maintenance costs of backup generators) can be quantified and are important to consider as cost offsets in a CHP feasibility analysis.
As an example of how these cost offsets can impact CHP economics, Table 4 provides an economic comparison of a hypothetical 1500 kW natural gas-fueled CHP system with and without the capability to provide backup power to a site during grid power outages. The impact of enhanced reliability is calculated two different ways. The first method is based on a customer's specific calculations for the value of service and expected number of hours per year of facility disruption that could be avoided (Table 3) with a CHP system that includes backup capability. For a customer with a VOS of $30/unserved kWh and an expected decrease in downtime of 3.75 hours/year, the internal rate of return for the CHP project example increases from 12.2 percent for the standard CHP system to 17.5 percent for the system with backup capabilities, and the net present value increases by a factor of four. The second approach, based on willingness to pay, is simply to take a capital cost credit for avoiding the cost of a diesel backup generator. A capital credit is taken for the backup gen-set, controls, and switchgear that would not need to be installed at the site because backup capability is integrated into the CHP system (note that the CHP system includes an additional capital cost for this capability, but the incremental capital cost is more than offset by credit from the displaced backup gen-set). With the second method, the simple payback for the CHP system is reduced from 6.8 to 5.3 years and the internal rate of return is increased to 16.9 percent.
Table 4. CHP Value Comparison With and Without Backup Power Capability7
|CHP System Components||Standard CHP (no off-grid reliability benefit)||CHP With Backup Capabilities – Direct Cost Measure||CHP With Backup Capabilities – Avoided Diesel Generator Measure|
|Generator Capacity (kW)||1500||1500||1500|
|CHP System Installed Cost, ($/kW)||$1,800||$1,800||$1,800|
|Added Controls and Switchgear Cost, ($/kW)||N/A||$175||$175|
|Typical Backup Gen-Set, Controls, and Switchgear, ($/Kw)||N/A||Not valued directly||($550)|
|Total CHP System Capital Cost, ($/kW)||$1,800||$1,975||$1,425|
|Total CHP System Capital Cost, ($)||$2,700,000||$2,962,500||$2,137,500|
|Net Annual Energy Savings, ($)||$400,000||$400,000||$400,000|
|Decrease in Annual Outrage Time (hours/year)||0||3.8 hours||Not valued directly|
|Customer Value of Service ($/kW-year)||N/A||$113/kW-year||Not valued directly|
|Annual Decrease in Outage Costs||N/A||$168,750||Not valued directly|
|Total Annual Savings||$400,000||$568,750||$400,000|
|Payback||6.8 Years||5.2 Years||5.3 Years|
|Internal Rate of Return||12.2%||17.5%||16.9%|
|Net Present Value (at 10% discount)||$311,302||$1,239,507||$822,665|
It should also be noted that a properly configured CHP system can provide better protection than a backup generator because the CHP system reduces the time to pick up load (when it is running), and it provides a measure of voltage support that helps to protect the facility from momentary, as well as extended outages.
1 New York Times, August 16, 2003.
2 Electric Power Research Institute (EPRI), An Assessment of Distribution System Power/Quality: Volumes 1-3, TR-106294 (V1, V2, V3), EPRI, Palo Alto, CA, 1995.
3 California Energy Commission (CEC), The Cost of Wildlife-Caused Power Outages to California's Economy, Energy and Environmental Economics, CEC Report CEC-500-2005-030, February 2005.
4 Kristina H. LaCommare and Joseph H. Eto, Understanding the Cost of Power Interruptions to U.S. Electricity Consumers, Lawrence Berkeley National Laboratory, September 2004.
5 Adapted from: K. Darrow and M. Koplow, Dual Fuel Retrofit Market Assessment, Onsite Energy Corporation for Gas Research Institute, 1998. (Costs escalated at 3 percent per year for equipment and 6 percent per year for labor.)
6 An Assessment of Distribution System Power Quality, Volumes 1-3, TR-106294-V1, V2, V3, EPRI, Palo Alto, CA, 1995.
7 Adapted from The Role of Distributed Generation in Power Quality and Reliability, Energy and Environmental Analysis, Inc. for New York State Energy Research & Development Administration. June 2004.