Area Control

Concept Link IconSee Also

 

One of the most important aspects of interconnected power system operation is the requirement that each operating area changes its total generation to match changes in the sum of its load plus losses plus power transactions with other areas. This requirement is normally met by Automatic Generation Control (AGC). The purpose of AGC is to ensure that the actual MW output of an area is equal to the scheduled MW output of the area. The AGC system accomplishes this by first calculating the Area Control Error (ACE), which is defined as

 ACE = Pactual - Pscheduled

In Simulator, Pscheduled for an area is made up of the Area’s MW Transactions and the Area’s Unspecified MW Export. MW Transactions represent the transfer of power between two areas in the power system. This transaction is done presumably under a contract between the two areas. The advantage of using MW transactions to describe these is that it ensures that the total export of all areas is consistent: if one area is exporting 100 then another area is automatically importing that power. However, each area also specifies a value called the Unspecified MW Export which can be entered on the Area Information Dialog. The unspecified MW export represents an export of power from the area that goes to an unspecified other area. When using unspecified MW exports, it is important that the total unspecified MW exports in the system sum to zero. PowerWorld highly recommends that care be taken when using unspecified exports.

Whenever the ACE is greater than zero, it means that the area is over generating and thus needs either to decrease generation or to sell more. Likewise, whenever the ACE is less than zero, the area is under generating and thus needs either to increase generation or to buy more. AGC works to keep the ACE close to zero.

 

In Simulator, there are six options for implementing AGC:

No area control

The output of the generators does not change automatically. You must manually change the generation to match system load/losses/transaction variation. All change in load/losses/transaction in this area will be made up at the islands slack bus.

Participation Factor Control

The output of all the area’s generators who have their AGC field set to "YES" change automatically to drive the area control error (ACE) to zero. Each generator’s output is changed in proportion to its participation factor. Checking this option enables the Set Factors button on the Area MW Control Options tab of the Area Information Dialog, which, when pressed, opens the Generator Participation Factors Dialog. Participation Factor Control only adjusts generation when a change to the system has taken place, such as changing the amount of load in the case, or defining new area to area transactions.

In Participation Factor Control, the ACE is allocated to each AGC generator in the area in proportion to that generator’s participation factor divided by the total of the participation factors for all AGC generators in the area. A generator’s participation factor cannot be negative. By default, a generator’s participation factor equals its current MW setpoint value, but individual participation factors can be changed.

Economic Dispatch Control

The output of all AGC generators in the area changes automatically to drive the area control error (ACE) to zero. Each generator’s output is changed so that the system is dispatched economically, based on cost information entered for the generators in the case. Note that cost data is not generally included in standard load flow data. Without realistic cost data entered into Simulator, the use of the economic dispatch algorithm may not be very useful. Cost data must be obtained from another source and entered into a case in Simulator, either manually or through the use of Simulator Auxiliary Files.

With Economic Dispatch (ED) Control, Simulator tries to change the output of the area’s AGC generators economically so that the area’s operating cost is minimized. ED control recognizes that some generators are less expensive than others and tries to use the least expensive generators to the largest extent possible.

To do economic dispatch, we need to know how much it would cost to generate one more MW at a particular generator. This is known as the incremental or marginal cost. For example, for the cubic cost curve model, the incremental cost for each generator is modeled using the formula:

 li = ICi (Pgi) = ( bi + 2ci Pgi + 3di (Pgi) 2 ) * fuelcost   $/MWH

The plot of ICi(Pgi) as a function of Pgi is know as the incremental-cost curve. The economic dispatch for a system occurs when the incremental costs for all the generators (li) are equal. This value is known as the system l (lambda) or system incremental cost. Its value tells you how much it would cost to generate one more MW for one hour. The system lambda becomes important when trying to determine whether or not an area should buy or sell power. For example, if an area can buy power for cheaper than it can generate it, it might be a good idea for the area to buy power.

Generators that are allowed to participate in economic dispatch control will have their MW limits enforced regardless of how any of the options specifying that MW limits be enforced are set.

Area Slack Bus Control

Only the output of the area’s slack bus changes automatically to drive the area control error (ACE) to zero. This type of generation control is usually only good for small disturbances to the injections and/or transactions in a case, and can often fail to find a solution when larger disturbances are examined.

Injection Group Area Slack Control

Only the Injection Group specified as the Injection Group Area Slack will change to automatically drive the area control error (ACE) to zero. This allows you to be very specific with each area as to which generation (or load) should vary to maintain ACE.

Optimal Power Flow (OPF)

The OPF option will only be available if you have the OPF add-on for PowerWorld Simulator. The OPF control is very similar to the Economic Dispatch control in that it attempts to dispatch generation to minimize costs. The additional function of the OPF is to minimize the costs while also obeying line, transformer, and interface limit constraints. This option is also not useful without realistic generator cost information, which usually must be obtained from another source and entered into Simulator to augment a load flow case.

The OPF control also relies on the cost curve in order to perform an economically optimal power flow. However, the OPF routine makes use of piecewise linear curves in its solution algorithm. This does not prevent you from entering the cost information as cubic cost models, described by the equation above. Rather Simulator’s OPF routine allows you to specify how to break up the cubic curve and model it as a piecewise linear curve for the OPF algorithm.

Generators that are allowed to participate in OPF control will have their MW limits enforced regardless of how any of the options specifying that MW limits be enforced are set.

 

In addition, you can also enter piecewise linear curves directly instead of the cubic cost curve models. In fact, a mixture of piecewise linear and cubic models is acceptable. For the economic dispatch routine, whichever type of model is entered will be used directly for each generator. For the OPF routine, all piecewise linear curves entered directly will be used as is, and any cubic models entered will be converted to piecewise linear curves internally during the processing of the OPF algorithm.