reading / DOE docs / VOL 1 2000 Heat Rate Improvement Guidelines for Indian Power Plants

1.A complete listing of the each measured parameter, for every test. This includes every temperature, pressure, differential pressure (for flows), gas analysis, etc. A table should be provided with the parameter name, the value, and the units for the value.

2.A complete listing of each calculated parameter, including intermediate calculated values. This includes all flow rates, TTDs, DCA, turbine section efficiencies, pump heads, etc. An example of an "intermediate calculated valueyywould be the calculated flow rates such as turbine extractions, determined fiom a measured flow (such as condensate flow) and an energy balance around a feedwater heater. Another "intermediate calculated" parameter would be the "APH no-leakage exit gas temperature corrected to the reference air entering temperature." This value is an intermediate step in determining the boiler efficiency.

Note: much of this data is also typically placed on a heat balance diagram. Preferably two heat balance diagrams should be made for each test, one with the actual data, and the other with "corrected to contract conditions" data.

3.A complete list of all "constantsy' required for calculations, such as data on flow nozzles (pipe & nozzle IDS,etc.).

4.A copy of all curves required for calculations, such as turbine exhaust loss curves, generator loss curves, heat rate and load correction to heat rate curves, flow nozzle/orifice calibration curves, etc.

5.A list of all "assumed" values. Typically, some minor flows, etc. are not measured. All parameters not actually measured should be listed, with their value or curve given.

6.An example of the calculations should be provided, showing all calculations for one test.

7.Copies of the pre test as-lee and post test as-found instrument calibrations should be part of the report.

This information is extremely valuable in allowing the utility to make additional calculations, determine baseline performance (See Section 3), set specificationltarget values for trend charts (see Section 5), etc.

C02 wet: HI = [Q I Fc ] * [%C02/100]

0 2 wet: HI = [Q I Fc ] * [ (20.91100) * (100-%H20) - %02 ] 120.9

In addition to the problem of getting a representative sample of the flue gas (maintaining a leak free system that is usually under vacuum, stratification of the gas due to duct leaks, etc.), there are potential problems with the flow measurement. Enough measurement points must be used to get the true flow as the flow profile is usually non-uniform. Also, the direction of the flow is often not parallel to the duct wall, but is swirling. This must be taken into account to measure the true flow rate. Usually, this involves using a directional pitot tube (such as a 3 or 5 hole Fechheimer) to calibrate the hot wire anemometer(s).

10.3Direct Measurement of Boiler Input

The second method utilizes the ultimate fbel analysis, specificallythe carbon content of the fuel along with the mass of fuel burned. This requires an intensive, continuous sampling of the fbel, followed by an ultimate analysis of the fbel, as well as a method for measuring the quantity of he1 (i.e., scales for coal units). When 1 kilogram of carbon burns, 3.664 (44.01112.01) kilograms of C02 is produced. The equation to calculate CO2 emissions is then:

Kilograms of COz= (Kilograms of Fuel Burned) * (% carbon in as-fired fbeVlO0) * (3.664 kilograms of CO2klogram carbon)

This calculation assumes that all the carbon in the fuel is burned to C02. When firing coal, some of the carbon is not burned and ends up as carbon in the ash. The amount of unburned carbon in the ash is normally small enough that it can be ignored without appreciably affecting the final result. Some of the carbon is not completely burned, but forms CO. The amount of CO is typically very small (on the order of 100 ppm), and will not affect the accuracy of the calculations.

10.4US DOE's Fuel Emissions Factor

The third method is to use US DOE's %el emissions factor (Reference 2). This table lists approximate conversion factors for millions of metric tons of CO2 per quadrillion Btu of heat input, for different fbels. To use this method, the heat input to the boiler (coal burned

and HHV or heat rate and generation), and the fie1 type must be known. The equation to calculate CO2 emissions is then:

English Units

A. Tons of CO2 = (Pounds of Fuel Burned) * (HHV of fbel Btdpound) * (Table 10.2factor lo6Tons c0z/1015Btu)

B. Tons of CO2 = (Generation kwh) * (Heat Rate Btu/kWh) * (Table 10.2 factor lo6Tons ~ 0 ~ 1 1 0Btu)"

Metric Units

A.Tons of CO2 = (kilograms of Fuel Burned) * (HHV of fbel kcal/kilogram)

*(Table 10.2 factor lo6Tons ~ 0 2 / 1 kcal)0 ~ ~

B.Tons of C02= (Generation kWh) * (Heat Rate kcaVkWh) *

(Table 10.2factor lo6Tons C02/10" kcal)

Table 10.2 U.S. DOE Table 11 in Emissions of Greenhouse Gases in the United States

1985-1990. DOE/EIA-0573

10.5 Examples

% Carbon in as-fired k e l fi-omUltimate Analysis = 66.4% Bituminous he1

Calculations:

Bituminous he1 3 from Table 10.2 3 365 lo6tons C02/ 10" kcal

10.5.1 Direct Measurement of Boiler Input

kilograms of C02= &lograms of Fuel Burned) * (% carbon in as-fired fbeVlO0) * (3.664 kilograms of C02/kilogramscarbon)

kilograms of CO2 = (2.532 10' kilograms) * (66.4/100) * (3.664 kilograms of C02/kilogramscarbon)

kilograms of C02= 6.16 10' kilograms = 6.16 lo6tons of C02

10.5.2 US DOE's Fuel Emissions Factor (Using Coal Burned and HHV)

Tons of CO2 = (Kilograms of Fuel Burned) * (HHV of &el kcalMogram) * (U.S. DOE factor 1o6tons c02/1015kcal)

Tons of C02= (2.532 10' kilograms) * (6668 kcaVkg) * (365 lo6tons C02/ 1015kcal)

Tons of C02= 6.16 1o6tons of C02

10.5.3 US DOE's Fuel Emissions Factor (Using Heat Rate and Generation)

Tons of CO2 = (Generation kwh) * (Heat Rate kcaYkWh) * (U.S. DOE factor 1o6 tons ~ 0 2 / 1 0kcal)' ~

Tons of C02 = (6.7 10' kwh) * (2520 kcaVkWh) * (365 1o6tons CO2 / 1015kcal)

Tons of C02= 6.16 lo6tons of CO2

10.6Conclusions & References

For immediate implementation of C02 emissions tracking, the third method, detailed in Section 10.4, could be used. Since the type of i3el is known, and most Indian plants have a relatively good measurement of the amount of &el burned, estimation of CO2 emissions using U.S. DOE'S Fuel Emissions Factor could be started immediately. For fbture installations, installation of stack sensors to use the direct measurement method should be considered.

United StatesFederal Register, Volume 64, Number 101

40 CFR Parts 72 & 75 Acid Rain Program; Continuous Emission Monitoring Rule Revisions; Final Rule

40 CFR Part 60 Appendix - Test Methods

Method 1 - Sample and velocity traverses for stationary sources

Method 2 - Determination of stack gas velocity and volumetric flowrate (type S pitot tube)

Method 3 - Gas analysis for carbon dioxide, oxygen, excess air, and dry molecular weight

Method 4 - Determination of moisture content in stack gases

United StatesDepartment of Energy, Energy Information Administration. 1993. Table 11 in Emissions of Greenhouse Gases in the United States 19851990. DOEEIA-0573

SECTION 11 ECONOMIC DISPATCH OF MULTIPLE UNITS

11.1Introduction & History

Economic Dispatch (also known as Merit Order Dispatch) is the process by which each unit is loaded, so as to minimize the total cost of power production while providing the required power to the system. This process can be applied to two or more units at a single plant, to two or more units in a region, or to dozens of units at multiple sites. The result of economic dispatch is that the total cost of providing the required amount of electricity is minimized. The cost of production at any individual unit may not be minimized, but the total cost of the required energy will be minimized.

Prior to 1930, there were primarily two methods used to dispatch units in a system:

1."Base Load Method" where the most efficient unit is loaded to its maximum capability, then the second most efficient unit is loaded to its maximum capacity, etc.

2."Best Point Loading" where units are successively loaded to their lowest heat rate point beginning with the most efficient unit, and working down to the least efficient unit, etc.

These methods result in the more efficient units operating at a low cost, but the less efficient units are left producing power at a very high cost. The net result on the overall system is a production cost that is higher than it should be.

It was recognized by 1930 that the "Equal Incremental Cost Method" yielded the most economic results. The idea was for the next increment in load to be picked up by the unit whose incremental cost was the lowest; it was recognized that the net effect would be an equalizing of incremental costs. By 1943, the incremental cost characteristics were represented by straight line segments, so that the equal incremental criterion was conveniently applied.

The only cost that is considered is the variable cost of power production. The variable cost is made up of two components, fie1 cost and variable "operations and maintenanceyycost.

The fuel cost is the cost of the required energy to meet an assigned load. It is usually derived from the heat rate curve for each unit. Multiplying the heat rate curve by the unit output gives an Input versus Output curve (energy input rate versus net output). The derivative of the input versus output curve is the curve of incremental heat rate (kJkWh or kcaVkWh) versus net output (kW). Multiplying this curve by the he1 cost results in an incremental fuel cost curve.

While the total rupee amount of the variable operations and maintenance cost (O&M) for a specified period varies with the amount of generation, its increment value in Rs/MWh is considered constant over the entire operating range (not a fbnction of load). It includes the O&M cost of pulverizer parts and labor, he1 handling, demineralized water production, etc. These costs vary with the amount of generation. Examples of fixed O&M costs that would not be included

Substituting ( 6 ) into (7) gives

Since I(x), dVdx, and d21/dx2are continuous,

will yield a minimum or maximum I,. Differentiating (8) with respect to X,yields

Since d21a/dx; and d21dds2are both greater than zero,

Therefore, dIJdXa = 0 yields a minimum I,.

In other words, when dIa/dXa= dIddXb,I, is a minimum value.

11.3Mathematical Expressions

The first line of the following table shows three possible forms of Input versus Output curve, I(x). The first is a 1" order (a straight line), next is a 2"* order curve, and last is a 3rdorder curve. The second line of the table shows each Input versus Output equation divided by the Output, which results in the equation for the heat rate, I/x. The third line shows the first derivative of the heat