Running in Parallel
For most customers, DG systems are most cost-effective and efficient when they are operated in parallel with the utility grid. In simplest terms, running in parallel with the grid means that both the DG system and the grid supply power to the facility at the same time. Paralleled systems offer added reliability: when the DG system is down for maintenance, the grid meets the full electrical load.

Conversely, DG systems can be designed to keep a facility up and running without an interruption if the grid experiences an outage. Also, grid-paralleled systems can be sized smaller to meet the customer?s base load as opposed to its peak load. Not only is the smaller base load system cheaper, it also runs closer to its rated capacity and, therefore, is more fuel efficient and cost effective.

The Process
Lack of a well-defined interconnect standard and failure to adhere to a standard can add considerably to engineering and equipment costs, making process planning difficult. Many interconnection requirements were drafted and adopted without understanding of the protection capabilities of modern DG equipment. As a result, these requirements often unnecessarily burden projects with redundant studies and hardware.

Barring additional studies or hardware upgrades, the cost of interconnection typically entails an application fee, which can range from a few hundred to a few thousand dollars. Engineering studies to determine the impact on the grid can run from a few thousand to tens of thousands of dollars. As response time is generally slow, multiple review cycles can add months to the schedule for a project. Additional hardware requirements can include upgrades to existing equipment or the installation of new equipment with costs ranging from a few thousand to tens of thousands of dollars.

Utility Power Distribution
Generally speaking, utility distribution systems can be categorized as either radial or networked. Radial distribution refers to a system where power lines extend from a common substation with customer loads coming off at single nodes along the line. In these distribution systems, power can only flow in one direction: from the substation to a load. A disruption to that feed or substation will typically affect all customer loads on that line.

Networked distribution refers to a system where numerous separate lines form a grid so that customer loads can tap off of multiple independent feeds, which are then tied to a common bus on the secondary side of the transformers. These can be separate lines from a common substation or they can be from independent substations. Further, networked distribution systems can either be area networks, in which several sites within a one or two block area are served, or spot networks where a single large building is served.

Networked systems are far more robust because a disruption to any one local element of the distribution system will not disrupt service to customer loads. These systems use network protectors, which quickly isolate faults to protect the grid and shift customer loads onto the remaining feeds. Typically located in downtown metropolitan areas, networked grids serve a utility?s most lucrative customers and are arguably its most valuable assets. Understandably, utilities are reluctant to allow the interconnection of anything that they feel will jeopardize the integrity or safety of this system.

Although networked grids are considered off-limits for interconnection by many utilities, safety and performance issues such as back-feed, islanding, cogging, and asynchronous re-closure can be adequately addressed with existing protection equipment available to even small-scale DG systems. While some upgrading of utility protection equipment may be required, it is typically a manageable cost when compared to the benefits of DG in these environments.

A typical requirement is upgrading electromechanical relays to more easily facilitate programming delays that prevent nuisance tripping or cycling of network protectors. This becomes an issue when on-site generation reduces the loads on the network transformers to the point where regenerative loads in the building and transformer impedance differentials can cause network protectors to trip on back feed. Programming a delay will allow the generator breaker to trip first.

Meeting Utility Requirements
Utilities must be concerned about contributions from a generating asset that will compromise either life safety or the integrity of their distribution systems. For network grids, network protectors have been designed to protect against back feed from the other network feeds and were not intended to be used as secondary protection for on-site generation. It is therefore incumbent on the on-site power system developer to ensure that the network protectors do not see conditions that fall outside of their design intent. In order to meet most utility requirements for protection, a project will need to be able to demonstrate the following capabilities:

• visible break disconnect
• automatic lockout
• anti-islanding protection
• reverse power flow detection
• primary fault detection (line to line and line to ground)
• protection from frequency and voltage excursions
• out of synchronism protection

power quality preservation (harmonics & flicker) Many states have interconnection standards, which can guide the user through a review process. States with either existing or developing interconnection standards include California, New York, Massachusetts, Texas, Delaware, Vermont, Arizona, Illinois, Michigan, Ohio, Pennsylvania, and Wisconsin. In addition, the Institute of Electrical and Electronics Engineers (IEEE) has recently completed work on a national standard for interconnection covering distributed generation up to 10 megawatts. Most utilities also have groups that are specifically charged with handling interconnection applications.

The actual interconnection process and requirements depend greatly on the particular utility. Depending on the size, location, and complexity of the project, the process can either be relatively straightforward or frustratingly complex. It?s often difficult to get a clear definition of time and cost for this process until the project is well defined, which can make the process hard to plan and budget. Schedule delays and added costs resulting from studies and hardware requirements can be much more difficult to deal with if they come late in the project. For this reason, it?s best to start the interconnection process as soon as possible and use the services of a knowledgeable engineer who can interpret the interconnection standard, identify potential problems, and argue the case against unnecessary studies or protection hardware.

Interconnection Options
Inverter-based systems are typically the easiest to interconnect. They disconnect almost instantaneously when the grid signal is lost and are therefore inherently protected against fault currents (islanding). In addition, inverters are current limited so fault currents are typically in the range of one to two times maximum normal output.

By contrast, rotating machines (induction and synchronous generators) are capable of producing instantaneous fault currents of six to ten times maximum normal output. Induction generators are also inherently protected against islanding as any loss of the grid will cause a loss of the induction field on the generator, though they are slower to disconnect than inverter-based systems. Because they can continue to produce fault current (islanding) when the grid goes down, synchronous generators are more capable of creating performance or life-safety problems and therefore require additional protection capabilities.

Modern digital paralleling control systems for synchronous gen sets often have all of the protection capabilities required for interconnection. Lack of familiarity with these systems, however, can make them a hard sell with some utilities. The added investment in a utility grade relay?which is a known quantity to the utility protection engineer?can ultimately save money, as it will greatly speed the review process.

Packaged CHP Systems
Many packaged CHP systems are being designed to comply with state interconnection standards, which can simplify the process to some extent. Still, this does not mean that packaged CHP systems should be viewed as a free pass on the interconnection agreement. Because each area of the grid is different and the impact of a DG asset can never be completed predictably, most utilities will still subject packaged CHP system projects to the same amount of scrutiny and paperwork for interconnection approval.

Understanding the issues surrounding utility grid interconnection is the one of the keys to a successful DG/on-site power system implementation. Working up front with competent DG/on-site power system engineering professionals can make the process easier.

Dan Reicher is executive vice president of Northern Power Systems and head of the company's Renewable Integration business group. Charles (Chach) Curtis is Northern?s vice president of finance and head of the Distributed Generation (DG) business group. Jim McNamara is a sales engineer with the DG business group. Based in Waitsfield, VT, Northern Power Systems designs, builds and installs ultra-reliable electric power system solutions for industrial, commercial, and government customers worldwide. A more detailed white paper on the subject of this article is available at the company?s website (www.northernpower.com).

Return to Archives