03-07-2012, 11:51 AM
ECONOMIC DISPATCH ALGORITHMS FOR THERMAL UNIT SYSTEM
INVOLVING COMBINED CYCLE UNITS
ECONOMIC DISPATCH ALGORITHMS.pdf (Size: 175.17 KB / Downloads: 81)
INTRODUCTION
The economic dispatch (ED) function allocates total
demand among the available generating units to minimize
the total generation cost. This activity is often
executed on a minute-by-minute basis at each control
center or Independent System Operator (ISO) [1]. Cost
curves of conventional thermal units can be modeled as
convex functions. Equal incremental cost is criterial to
solve the traditional ED problem [2]. Combined cycle
units utilize both gas turbines and steam turbines to
produce electrical energy. The waste heat from the combustion
turbines is directed into a boiler just as steam
from the boiler is used to power steam turbines. Combined
cycle units are of relatively high efficiency, have
fast ramp rates and exhibit other beneficial features
compared to conventional thermal units [3]. This has
enabled the combined cycle units to become the technology
of choice for many new power facilities.
COST CURVE ANALYSIS OF COMBINED CYCLE UNITS
Typically, a combined cycle plant consists of several
combustion turbines (CTs) and an HRSG/steam turbine
(ST) set. Based on different combinations of CTs and
STs, a combined cycle unit can operate in multiple configurations.
Each combination of CTs and STs can be
regarded as a state. Each state has its own unique cost
curve characteristic.
COMPARISONS OF SOLUTION METHODS
The condition required to apply the equal incremental
cost criterion is that all curves are monotonous increasing.
With combined cycle units, monotonous decreasing
intervals are typically ignored and only the monotonous
increasing intervals participate in economic dispatch to
apply the equal incremental cost criterion. When cost
curve sections are ignored, the solution may not be optimal.
Thus, methods that incude these segments
should be examined. Such methods include: complete
enumeration, merit order loading, genetic algorithm and
a hybrid technique with convex optimization and complete
enumeration are applied to find the global optimal
solution.
APPLICATION
The case study is a test system with twelve thermal
units and one combined cycle unit.
The CC units consist of two combustion turbines and
one HRSG steam turbine. There are totally four operation
Incremental cost curves of the CC unit are shown in
Figure 4. State 1&2 essentially are thermal unit states.
Correspondingly, incremental cost should be monotonously
increasing. The “One CT” and “Two CTs”
curves are almost monotonously increasing, except a
few points which should be bad data. The “One CT and
One HRGS” and “Two CTs and One HRGS” curves
corresponding to state 3&4 are not monotonously increasing
any more, because state 3&4 are combined
cycle unit states.
CONCLUSION
This paper first explains the conventional solution
and analyzes combined cycle cost curves. Since the cost
curves are not convex and the incremental cost curves
are not monotonously increasing, the equal incremental
criterion can not be applied.