SUMMARY OF THE INVENTION

[0019] In the electrostatic precipitation of fly ash in flue gas, rapping reentrainment of fly ash particles is a problem. The herein disclosed invention is designed to remedy this problem. The problem is solved by a method in which ammonium bisulfate is made to contact fly ash particles. That is ammonium bisulfate rather than ammonium sulfate is used to contact the fly ash. Ammonium bisulfate will produce a desired sticky product when contacting particulate fly ash. On the other hand, ammonium sulfate will product a fine particulate product which is not sticky. The use of ammonium bisulfate in the process of this invention increases cohesivity of the fly ash. This cohesivity or stickiness causes better plate adhesion.

[0020] The plates, which are generally found in multiple fields, are intended to electrostatically attract and accumulate fly ash particles present in flue gas. To rid the plates of accumulated fly ash, the plates are rapped. In the process of the herein disclosed invention, ammonium bisulfate which is sticky and adhesive causes the fly ash to adhere to the plates. When the plates are rapped, the fly ash drops off as a sheet.

[0021] Another important feature of this invention is the need for the addition of a stoichometric amount or a little less of ammonia. That is, ammonia to SO₃ is ideally in a ratio of 1 to 1 or a little less of ammonia.

[0022] The amount of ammonia to be added is based on the total amount of SO₃ present in the flue gas. This total amount of SO₃ is that amount of SO₃ inherently present due to the burning of the coal and the amount of SO₃ add to optimize resistivity. A method for determining the amount of SO₃ in flue gas can be accomplished by an algorithm. This algorithm method is an important feature of this invention.

[0023] In order to carry out the process under optimum conditions, the amount of SO₃ in the system has to be known so that a stoichiometric amount or a little less of ammonia is to be added. The total amount of ammonia to be added is based on the total amount of SO₃ in the flue gas. The total SO₃ would be the sum of coal combustion and/or the addition of SO₃ to the flue gas.

[0024] It is to be understood that while SO₃ has been referred to throughout this disclosure, in reality SO₃ is reactive with moisture (H₂O) to become sulfuric acid (H₂SO₄) vapor in the flue gas.

[0025] Optimized Ammonia System

[0026] The new ammonia conditioning system control algorithm described in this invention disclosure is based on the assumption that the optimum concentration of flue gas SO₃ is known. The SO₃ control algorithm in the companion disclosure determines this number in one of the intermediate steps. Once the flue gas SO₃ concentration is known, the algorithm in this invention disclosure can be used to calculate the optimum rate of ammonia addition.

[0027] This new method of controlling an ammonia conditioning system is based on the understanding of the ammonia conditioning process developed during numerous EPRI test programs. It starts with the determination of the optimum SO₃ flue gas concentration. Having determined the optimum SO₃ concentration, the rate of ammonia addition is then calculated using the following algorithm.

[0028] Step 1. Determine flue gas SO₃ concentration (assumed to be the optimum concentration for fly ash resistivity control).

[0029] Step 2. Multiply the SO₃ concentration from Step 1 by the number, n, with a value between 0.4 and 1.0. The exact value of n is determined by trial and error.

[0030] Step 3. Output the calculated value to the ammonia conditioning control system.

[0031] The above procedure is based on the finding that the addition of ammonia at a rate that produces an ammonia to SO₃ stoichiometry of one or less will result in the formation of the compound, ammonia bisulfate. It is this compound, which is sticky at typical ESP operating temperatures, that is responsible for its effectiveness as a conditioning agent.

[0032] More specifically, the process of this invention can be carried out by the following steps:

[0033] 1. Complete a combustion calculation to determine the SO₂ concentration in flue gas. The measure could be; for example, parts per million (ppm) by volume in flue gas.

[0034] 2a. Multiply the SO₂ concentration of step 1 by the number 0.004 for Eastern and Western Coals; with the exception, multiply the SO₂ concentration by 0.001 for Powder River Basin Coal. The resulting number is the SO₃ concentration in flue gas from conversion in the boiler, duct work and air heater.

[0035] 2b. When an SCR (Selective Catalytic Reactor) for NO_x control is present, compute the conversion of SO₂ to SO₃ from the measured or the catalysis manufacturer's supplied data. This is the additional SO₃ in the flue gas from catalytic oxidation of SO₂.

[0036] 2c. Add the SO₃ concentration from step 2a to the SO₃ from step 2b. The resulting number is the SO₃ concentration in the flue gas.

[0037] 3. Add to the number of step (2c) any SO₃ added to the flue gas by the SO₃ flue gas conditioning systems.

[0038] 4. Multiply the number in step (3) by a number in the range of 0.4-0.7. This is an empirical number that compensates for the SO₃ adsorbed on the fly ash. Some of the SO₃ molecules are adsorbed on the surface of the fly ash particles. Because of this, the number of step (3) is too large by an amount depending on the chemistry of the fly ash.

[0039] Expressed another way, the new method to control an ammonia conditioning system is as follows:

[0040] Step 1. Measure certain physical and chemical parameters that characterize the operation of a process.

[0041] In this case:

[0042] a. Measure or calculate the flue gas SO₃ concentration. For example, the SO₃ concentration is determined to be 8 ppm.

[0043] b. Observe and record the instantaneous opacity of the flue gas leaving the electrostatic precipitator. This record will undoubtedly include short-term increases that are from several seconds to minutes in duration (called opacity excursions or spikes). The objective of the optimization method disclosed in this application is the minimization of the number, intensity and duration of these short-term excursions or spikes.

[0044] Regarding the opacity measurement (opacity excursions or spikes) of step (1) is a measure of fly ash particles in the flue gas. The fly ash particles in the flue gas may be due to the combination of fly ash particles in the flue gas resulting from the burning of coal plus the particles which may result from reentrainment. Spiking (indicating fly ash reentrainment) is not desired and accordingly the minimization of the spikes is desired. Minimization is accomplished by the addition of ammonia. For example the baseline opacity is at a minimum of 5%-10%, but periodically is elevated to 40%-50%. When opacity goes up somewhat, more ammonia is added.

[0045] Step 2. Mathematically adjust the value of the single parameter controlled in the calculations until a value is found that produces the desired process operation.

[0046] In this case:

[0047] Multiply the SO₃ concentration from Step 1 by the number, n, with a value between 0.4 and 1.0. The exact value of n is determined by an iterative process (see Step 5).

[0048] For example, choose 0.4 as the multiplier with the result that 3.2 ppm of ammonia should be added to the flue gas.

[0049] Step 3. If an SCR is present, reduce the calculated ammonia feed requirement by any ammonia slip that gets through the SCR and air preheater.

[0050] Step 4. Using the measured or calculated gas flow, convert ammonia concentration to an ammonia feed rate.

[0051] Step 5. Physically set the controlled parameter to the value identified in Step 4.

[0052] In this case:

[0053] Output the calculated value to the ammonia conditioning control system

[0054] For example, a control signal would be sent to the ammonia conditioning system that would cause this system to add ammonia to the flue gas at a rate that would add ammonia at a rate that would produce a concentration of 3.2 ppm.

[0055] Step 4. Repeat the Step 1 measurements to verify that the process is operating in the desired manner.

[0056] In this case:

[0057] Observe and record the opacity of the flue gas leaving the ESP. The number of opacity excursions or spikes should have diminished. If significant spikes still persist, increase that factor used to determine the rate of ammonia addition.

[0058] For example, use the factor 0.6 in Step 2 to determine that 4.8 ppm should be added to the flue gas. Repeat Step 3 with this new value and then repeat Step 4 until a satisfactory opacity record in produced.

[0059] All current ammonia conditioning system controls inject ammonia at a preset rate that is moderated only by unit load. The algorithm in this invention disclosure improves on this process by selecting a rate of ammonia addition that will optimize ESP performance even if the SO₃ concentration in the flue gas changes. The SO₃ concentration at plants that burn variable coal supplies, a common occurrence, does vary considerably; hence, the new algorithm is superior to current control processes which are insensitive to the flue gas SO₃ concentration.

[0060] This new control process can be applied to any ammonia conditioning system. These systems are in use at approximately twenty domestic utility plants.

[0061] The technical support for this control process comes from field tests and laboratory studies of ammonia conditioning conducted by EPRI. Data from these tests demonstrated that ammonia injected at a rate that produces a molar concentration of ammonia that is equal to, or less than, the molar concentration of sulfuric acid vapor in the flue gas will result in the formation of ammonium bisulfate. It was further demonstrated that ammonium bisulfate co-precipitates with fly ash onto ESP collection plates, increases the cohesivity of fly ash on the plates and thereby reduces re-entrainment.

[0062] Sulfur Trioxide (SO₃) and Ammonia (NH₃) System

[0063] As part of the disclosed invention, there is a process depending on an algorithm wherein optimum amounts of SO₃ as well as ammonia are added to the flue gas stream in order to enhance the removal by the electrostatic precipitator of flue gas particles. As illustrated in Appendix A. Broadly considered, the algorithm is used to calculate the optimum amount of SO₃ to be added to the flue gas after which the derived value for optimum SO₃ is inputted to derive by an algorithm the optimum amount of ammonia to be added. Stated another way, once the optimum concentration of SO₃ has been determined by algorithm an optimum amount of ammonia is determined by algorithm. In this way, optimum amounts of SO₃ and ammonia can be added to the flue gas stream to properly condition fly ash particles. The algorithm for determining the amount of SO₃ to be added is set forth in detail in the accompanying Appendix A, attached as part of the specification disclosure.

[0064] In this invention optimized injection of SO₃ is obtained by continually monitoring the boiler load, boiler heat rate or the coal feed rate, the temperature of the flue gas, SO₂ content of the flue gas, and precipitator transformer/rectifier secondary currents and then using these variables to compute an optimum SO₃ injection rate. See Addition A.

[0065] The procedure is to first calculate the actual flow of flue gas to the precipitator based on the burn rate of coal in the boiler, which is either calculated from the boiler load and boiler heat rate or measured by the coal feed rate. Alternatively, where a stack gas flow measurement is available, the measured flue gas flow rate can be used and converted to actual cubic feet per minute of flue gas

Addition A

[0066] For example, representative numbers for these parameters are as follows: 1

[0067] flow for the temperature and pressure conditions at the inlet of the precipitator. The SO₂ to concentration in the flue gas stream is then determined by either a coal combustion calculation or by direct measurement at the stack by means of a continuous emission monitoring system (CEMS). A conversion factor is then used to estimate the SO₃ concentration in the flue gas based on the SO₂ concentration. The conversion factor is a function of the ash chemistry, the boiler configuration and composition of boiler components. In this invention the conversion factor can either by manually inputted or automatically selected based on the chemistry of fly ash. The chemistry the fly ash is either manually provided to the computation program or where available, measured online in real-time and provided to the computation program. See Addition B. When an SCR is present, a second conversion factor is used to compute the conversion of SO₂ to SO₃ in the SCR. The SO₃ from normal boiler conversion is added to the SO₃ from the SCR conversion to get the total SO₃ in the flue gas.

[0068] The next step is to input a desired operating resistivity or current density. The current density inputted is converted to an effective resistivity using correlations of current density and resistivity developed under EPRI project WO629 and included in EPRI report CS-5040. Precipitator Performance Estimation Procedure. The required SO₃ concentration in ppm is computed by using an inversion of the Roy Bickelhaupt resistivity calculation procedure for modification of ash resistivity by SO₃. The Beckelhaupt equation (calculation) is included in EPRI Report CS-4145. A Manual on the Use of the Gas Conditioning for ESP Performance Enhancement. A discussion of the computer code for performing this procedure is attached. Once the required SO₃ is determined for the selected ash resistivity, it is compared with the estimated flue gas SO₃ computed above. If the required SO₃

Addition B

[0069] For example, the coal and ash chemistry for the example case is as follows: 2

[0070] For this coal and ash, the appropriate SO₂ to SO₃ conversion factor is 0.004, so the estimated SO₃ concentration is:

0.00088×0.004=3.52 ppm

Addition C

[0071] For the example, the inherent SO₃ concentration is estimated to be 3.52 ppm and the needed or required concentration is estimated using the Bechlehaupt equation to be 7.33 ppm. Hence, it would be necessary to add 3.81 ppm to the flue gas.

[0072] is less than the estimated SO₃, no SO₃ injection is needed. If the required SO₃ is greater than the estimated flue gas SO₃ from the coal, then SO₃ injection is needed. The amount of SO₃ injection needed is the equal to difference between the amount of SO₃ in the flue gas from the conversion of coal sulfur to SO₃ and that required to condition the fly ash to achieve a high current density in the precipitator and agglomeration. See Addition C.

[0073] The additional SO₃ in ppm is converted to SO₃ injection in pounds per hour based on the measured or calculated flue gas flow rate. Procedures for this step are attached. The SO₃ injection rate in pounds per hour is then converted to a sulfur feed rate adjusted to include the conversion efficiency of sulfur to sulfur trioxide (SO₃) in the SO₃ supply system. The conversion rate is often a function of the sulfur feed rate. In this procedure, adjustments can be made to correct for the changes in conversion efficiency. Heat balance measurements for the SO₃ supply system can be used to calculate the conversion efficiency. The sulfur feed rate is then converted to a desired output to control the SO₃ supply system. Typical outputs are in units of sulfur gallons per hour, pump rate gallons per hour, or pump percent of maximum rating.

[0074] Optimization of the sulfur feed rate is obtained by adjustments of either the selected resistivity set point or the selected operating current density set point. See Addition D.

[0075] A refinement to procedure that can adjust for variations in coal chemistry is to compute an effective dynamic resistivity based on the precipitator measured secondary currents and voltages. If the dynamic resistivity is less or greater than the desired set point, a proportional adjustment is made to create a new set point to be used for the SO₃ injection calculation. Because of the delayed response time of precipitators to the injection of SO₃, a time delay that can be adjusted to select an appropriate delay for making the correction is used. Using this procedure, the set point is periodically adjusted to correct for variations in ash chemistry. Typically, ash chemistry will vary at

Addition D

[0076] In this example case, the optimum resistivity is chosen to 1.0×10¹⁰ ohm-cm. The corresponding current densities would be: 3

Addition E

[0077] In this example case, the ESP transformer current given earlier can be converted to current densities by dividing by the plate area energized by each transformer. The result of this procedure yield the following numbers: 4

[0078] These current densities correspond to the following estimated fly ash resistivities: 5

[0079] The resistivities in this example were calculated using the inverted form of the equations found in EPRI report CS-5040. Precipitator Performance Estimation Procedure.

[0080] a much slower rate than the fluctuations in coal sulfur content, boiler load, and gas temperature. In this invention the correction procedure is an optional calculation. See Addition E.

[0081] An additional refinement to the procedure is to use the total flue gas SO₃ (SO₃ from flue gas SO₃ from conversion of SO₂ in a SCR, and SO₃ from injection) as the control basis for the injection of NH₃ for precipitators using dual conditioning with SO₃ and NH₃. In this case, the ammonia concentration in ppm is set at a ratio of the total SO₃, for example 0.5 times the total SO₃. The ratio factor can be programmed to be selected by an operator and can be set to scale as a function of load and gas temperature. The required factor is reduced if there is any ammonia slip to the precipitator from ammonia injection to control NO_x in a SCR. Often additional NH₃ is needed with higher temperatures to cause the SO₃ to form a conductive layer on the fly ash. The NH₃ concentration is converted to a feed rate by multiplying the flue gas flow rate determined by combustion calculations or by flue gas flow measurements. A control output in mass per time or as a percent of maximum supply is computed and provided to the ammonia supply system to adjust the ammonia feed rate.

[0082] Often the flue gas flow rate and temperature vary across the face of the inlet duct to a precipitator. In particular, there are temperature variations due to the air pre-heater. By either measuring the flow and temperature or by using ratios that reflect the variations in the duct flow and temperature, the SO₃ required for each section of the duct can be computed using the same procedure as above. In this case, each section is treated separately and the SO₃ and NH₃ feed for each section are computed. The total required feed for each component is determined by summing the feeds for each section. The ratio of the feeds for each section to the total feed for each component are then computed and used to control valves to ratio the feed to each section of the duct. By using this procedure, over and under conditioning can be avoided resulting in a reduction in precipitator emissions.

[0083] Referring to Appendix A:

[0084] Invention I (Procedure with no ESP Feedback)

[0085] 1. “Typical” Starting Conditions:

[0086] low flue gas SO₃ concentration measured at the ESP inlet—0 to 4 ppm SO₃—example number=3.5.

[0087] moderate to high fly ash resistivity—8×10¹⁰ (8×10E10) ohm-cm to 5×10¹² (5×10E12) ohm-cm.

[0088] low ESP power level characterized by low average current densities—see Table 1. Appendix A.

[0089] B. Desired “End” Conditions:

[0090] increased flue gas SO₃ measured at ESP inlet—from 2 to 12 ppm, depending on flue gas temperature and fly ash composition.

[0091] optimum fly ash resistivity—8×10⁹ (8×10E9) ohm-cm to 4×10¹⁰ (4×10E10) ohm-cm, depending on ESP collection and reentrainment characteristics, example 1×10¹⁰ ohm-cm.

[0092] high ESP power levels as indicated by current density levels—again, see Table 1. Appendix A.

[0093] C. Calculation (invention) to determine the level of SO₃ injection needed to produce optimum fly ash resistivity and, hence, optimum level of flue gas SO₃.

[0094] Step 1.

[0095] 1. Determine the temperature of the flue gas entering the ESP from plant instrumentation, example=291° F.

[0096] 2. Obtain coal proximate the ultimate analysis and fly ash mineral analysis, see Table 2.

[0097] Step 2.

[0098] Estimate SO₃ background level of SO₃ in the flue gas using correlation relating flue gas SO₃ to coal type and coal sulfur content. The SO₃ concentration is calculated as a percentage of SO₂ in flue gas which can be determined from a combustion calculation using the coal analysis and flue gas O₂ or CO₂ or if the flue gas SO₂ is available from plant instruments, this number can be used in the SO₃ calculation.

[0099] See Calculation 1 and Calculation 2.

[0100] Step 3.

[0101] Calculate the base ash resistivity using empirical equations relating ash resistivity to ash composition, flue gas moisture and flue gas temperature. The Bickelhaupt equations are an example of relationships that can be used for this calculation.

[0102] See Calculation 3 and Calculation 4.

[0103] Step 4.

[0104] Use a correlation relating the base fly ash resistivity and flue gas SO₃ concentration to determine the flue gas SO₃ concentration needed to produce the optimum fly ash resistivity. Examples of correlation relating base fly ash resistivity and flue gas SO₃ to actual resistivity are procedures given by Bickelhaupt and by McCain. See Calculation 5.

[0105] Step 5.

[0106] Subtract the background SO₃ concentration from the needed SO₃ concentration from the needed SO₃ that must be added to the flue gas to produce the optimum fly ash resistivity. See Calculation 6.

[0107] Step 6.

[0108] Send rate of addition signal to the controls that operate the SO₃ conditioning system.

[0109] Referring to Appendix A:

[0110] Invention II (Procedure with ESP Feedback)

[0111] 1. “Typical” Starting Conditions:

[0112] low flue gas SO₃ concentration measured at the ESP inlet—0 to 4 ppm SO₃—example number=3.5.

[0113] moderate to high fly ash resistivity—8×10¹⁰ (8×10E10) ohm-cm to 5×10¹² (5×10E12) ohm-cm.

[0114] low ESP power level characterized by low average current densities—see Table 1. Appendix A.

[0115] B. Desired “End” Conditions:

[0116] increased flue gas SO₃ measured at ESP inlet—from 2 to 12 ppm, depending on flue gas temperature and fly ash composition.

[0117] optimum fly ash resistivity—8×10⁹ ohm-cm to 4×10¹⁰ ohm-cm, depending on ESP collection and reentrainment characteristics—example number 1×10¹⁰ ohm-cm.

[0118] high ESP power levels as indicated by current density levels—again, see Table 1. Appendix A.

[0119] 3. Calculation (invention) to determine the level of SO₃ injection needed to produce optimum fly ash resistivity and, hence, optimum level of flue gas SO₃.

[0120] Step 1.

[0121] Obtain coal proximate and ultimate analysis and fly ash mineral analyses—example case, see Table 2.

[0122] Step 2.

[0123] Determine the temperature of the flue gas entering the ESP from plant instrumentation, example number 291° F.

[0124] Step 3.

[0125] Estimate SO₃ background level of SO₃ in the flue gas using correlation relating flue gas SO₃ to coal type and coal sulfur content. See Calculations 1 and 2.

[0126] Step 4.

[0127] Measure the current levels in each field of the ESP and calculate the corresponding current densities. See Calculation 3.

[0128] Step 5.

[0129] Determine effective fly ash resistivity level in the ESP using a correlation that relates fly ash resistivity to ESP current density for each electrical field in the direction of gas flow. Average the results to produce an average resistivity for the ESP. If this resistivity is close to or lower than the optimum range, go to Step 10, otherwise proceed to Step 6. See Calculation 4.

[0130] Step 6.

[0131] Use a correlation relating fly ash composition and flue gas temperature and SO₃ concentration to fly ash resistivity to determine the flue gas SO₃ concentration needed to produce the optimum fly ash resistivity. See Calculations 5, 6 and 7.

[0132] Step 7.

[0133] Subtract the background SO₃ concentration from Step 3 from the needed SO₃ concentration from Step 6 to determine the amount of SO₃ that must be added to the flue gas to produce the optimum fly ash resistivity. See Calculation 8.

[0134] Step 8.

[0135] Send rate of addition signal to the controls that operate the SO₃ conditioning system.

[0136] Step 9.

[0137] Repeat Steps 4 and 5.

[0138] Step 10.

[0139] 1. If indicated ash resistivity is equal to or less than optimum resistivity, decrease rate of rejection by x percent where x is between 5 and 25.

[0140] Or

[0141] b. If indicated ash resistivity is greater than optimum resistivity, increase rate of injection of x percent where x is between 5 and 25.

[0142] Step 11.

[0143] Repeat Step 10 until indicated fly ash resistivity passes through optimum resistivity point and then set rate of injection at a point in the range bounded by the levels calculated in the last two interactions—for example, at a point that is halfway between the two levels.

[0144] Step 12.

[0145] Every y minutes, where y is number between 5 and 30, restart the process beginning at

[0146] Step 2

[0147] Obviously, many modifications may be made without departing from the basic spirit of the present invention. Accordingly, it will be appreciated by those skilled in the art that within the scope of the appended claims, the invention may be practiced other than has been specifically described herein.