With emission control standards becoming more stringent, coal fired power plant operators must consider costly ESP upgrades. Advanced modelling techniques can help to evaluate upgrade scenarios, and are currently being applied to a number of plants in Russia, the Ukraine and Kazakhstan.
Ken Parker, Consultant, UK, Norman Plaks, Consultant, USA
Of the 15 trillion kWh of electrical energy generated worldwide, almost one third is derived from coal fired installations. This will continue into the foreseeable future, and likely expand as developing countries increasingly make use of their abundant indigenous coal supplies.
With the combustion of this ready source of fuel come environmental concerns. The primary pollutant is fine particle matter or ash, for which the major control method is the electrostatic precipitator (ESP), which is capable of collection efficiencies in excess of 99.8 per cent. Typically, for a large boiler installation, these units cost upwards of several million dollars, therefore, decisions involving them are not trivial. Other pollutants are acid gases and, more recently, mercury, which require additional control technologies.
At the same time more stringent international emission control standards are developing which make it important that countries improve the particulate collection from their coal fired power stations which will require costly upgrading of their precipitators. What is needed is a quick, economical method for evaluating the available upgrade scenarios and to provide the basis for choosing the lowest-cost method that will achieve the desired emission level.
Figure 1. Basic principle of the ESP and the model
The authors are pioneers in the application of precipitator modelling beyond its traditional application, which was as a tool to determine the expected performance of an existing unit. The unique relationship between the full-scale precipitator and the computer model allows rapid and economical evaluation of different design and upgrade scenarios thereby allowing their optimization. This proactive approach makes use of the model’s ability for allowing changes to be made within the computer, to the precipitator’s mechanical design and hence electrical operating parameters and in determining the efficiency resulting from these changes. This proactive approach is centred on the readily available ESPVI 4.0 models, which can be used in combination with the requisite skills and knowledge to evaluate poorly performing precipitators.
The approach, under the auspices of the US Environmental Protection Agency (EPA), is currently underway in the Newly Independent States (NIS) to train a group of engineers in ESP modelling and its proactive application. With support from the authors, a number of precipitators in Russia, the Ukraine and Kazakhstan, where particulate emission reductions are required to meet forthcoming regulations, have been fully evaluated and for which implementation of relatively low-cost upgrade modifications are underway. The management of these now private sector electric energy companies is fully supportive of the approach.
The ESPVI 4.0 series model precisely tracks the full scale ESP whose parameters have been inputted by the use of algorithms involving the well recognised mathematical physics of precipitation. The analysis commences by dividing the particle size distributions into a number of smaller increments, each of which represents a discrete size for computing its charging and collection. The model then divides the precipitator length in a number of shorter increments, two for each of the corona discharge electrodes or elements, one upstream and the other downstream, as shown in Figure 1.
For each particle size increment and for each length increment the model computes the particle charge taking into consideration diffusion charging for the finer ones and field charging for the larger ones; as the particles travel downstream. For each length increment the model recomputes the charge for each particle size. The current for each element is computed as a function of its diameter and the surrounding particle and ionic space charge.
With the electrical conditions for each electrode element and its two length increments determined, the model computes the electrical migration velocity for each particle size. It then goes on to rigorously calculate the amount of each particle size collected using the Deutsch Equation. Finally the model summates the data to determine the particle collection efficiency for the whole precipitator. With the addition of empirically used non ideal inputs, such as rapping re-entrainment and non-uniform gas distribution, the model’s computations can very accurately duplicate the performance of any full scale precipitator.
To enable the model to precisely track the full scale precipitator it is necessary to input all of its essential data, including physical dimensions plus electrical, gas and particle properties. The most difficult to accurately determine are the electrical properties of the corona discharge electrodes and the particle sizing, especially the small micron and submicron sizes that establish the space charge and consequently the precipitator’s electrical operation. Techniques have been developed by the authors to establish both the corona discharge electrode’s electrical properties as well as the particle characteristics.
The model assumes that all corona discharge electrodes are of circular cross section, which is true for some electrode formats, however, most complex electrodes in commercial usage can be represented in the model, either by a single circular cross-section element or as an array of elements, as indicated in Figure 2. The array is especially useful for simulating the controlled emission type of electrode used to provide high corona current while maintaining good mechanical properties.
Figure 2. Array representing the tape and needle corona discharge electrode
The array shown in Figure 2 represents the tape and needle electrode, which has been widely used in Eastern Europe. It consists of a metal strip with rolled beaded edges with tabs or needles alternately punched out in opposite directions. The two 0.4 mm elements represent the tips of the needles where the corona forms and the 9 mm element, which has no corona, represents the tape. The single element or array’s properties are determined by matching its model generated VI curves to the actual electrode’s clean air load VI curves at ambient conditions as seen in Figure 3 for the tape and needle electrode. The clean gas eliminates any space charge effects. Once the air load VI curves have been computed, the model’s inbuilt capability is used to match the gas conditions to those of the precipitator.
After the corona discharge electrode properties have been characterized, the particle properties, especially the small micron and submicron species, can be identified. VI curves under actual precipitator operating conditions are compared and matched to the VI curves generated by the model as seen in Figure 4. This is an ideal practical approach, since the fine particles, through their space charge helps establish the precipitator’s electrical conditions, which the model considers explicitly. For a well operating precipitator, it is the difficult fines that constitute a significant portion of the particle matter that evades collection; the larger particles are readily collected and are therefore of little consequence.
The proactive approach
The heart of the ESPVI 4.0 model, as well as the theory that explains particle precipitation, is the well known Deutsch equation. This in its rigorous form computes the probability that a particle travelling in the turbulent interior of an ESP will enter and be retained within the collecting surface’s laminar layer. The rigorous application of Deutsch requires that the charge and migration velocity be specific to the particle and collecting area. The model does this by computing the charging and migration velocity for each particle size in each length increment.
The Deutsch relationship,
Collection = 1 ” exp (-A/Q*w)has as its parameters, the total collection area, A, the gas flow rate, Q, and the migration velocity, w. It tells us that if the area and migration velocity increases and the gas flow rate decreases, the collection of the precipitator improves. It further tells us that if, for example, the area is doubled it would provide the same order of improvement achieved by doubling the migration velocity or halving the gas flow rate.
Possible upgrade scenarios for proactive modelling to an ESP include, but are not limited to:
The collection area, A:
- Increase section heights
- Add additional sections.
The gas flow rate, Q:
- Decrease excess air
- Eliminate air inleakage.
The migration velocity, w:
- Increase high voltage
- Improve transformer rectifier sets, controllers, etc.
- Use different energization approach
- Increase high voltage sectionalization
- Improve electrode formats
- Flue gas conditioning
- Decrease gas temperature.
All of these, plus other upgrade options can be speedily evaluated by the model, which, once set to the physical and operating conditions, will accurately track performance. These upgrade options, which have different costs, have been shown to work during many years of ESP operation. There are other factors which must be addressed when considering any change, such as the precipitator’s age, mechanical and electrical condition and the particle characteristics, e.g. resistivity, etc.; which is where knowledge and experience comes to the fore.
Difficult fly ash
The proactive application of ESP modelling, combined with knowledge and experience, has special value when applied to precipitators collecting difficult fly ashes having high resistivity leading to reverse ionization or back corona operation. The problem is especially troublesome when the high resistivity ash is also present in high concentrations.
Figure 3. Model/actual VI curves
For example an old ESP in poor mechanical and electrical condition, collecting high resistivity ash operating under severe reverse ionization had a collection efficiency of about 88 per cent. Proactive modelling indicated that with a relatively low cost upgrade the efficiency could be ultimately raised to the 99 per cent level. This involves a combination of techniques including improved controllers to maintain the electrical conditions below the knee of the VI curve, where it goes into severe reverse ionization, plus the use of lower emission electrodes to allow operation at higher voltages while maintaining the same current density to prevent the re-emergence of reverse ionization.
Some of these principles are shown in Figure 5. Curve A is the VI curve for the original high current controlled emission electrode. The normal operating point, V3,I2, has the precipitator in deep reverse ionization. The current density I1 is the value, below the knee of the curve, above which the particular precipitator section goes into reverse ionization, therefore, V1, I1 is a good operating point where optimum efficiency would actually be attained. With flue gas conditioning the VI curve would change to A’, which would provide improved performance. Curve B is attained by using a reduced emission corona discharge electrode that has a higher onset voltage. Operation at V2, I1, would again be below the knee of the curve minimizing reverse ionization. Curve B’ would result from flue gas conditioning. Hence changing conditions to increase the operating voltage provides higher electric fields for charging, thereby enhancing the migration velocity and hence the collection efficiency.
Figure 4. VI curves for particle sizing
The proactive approach provides the ability to easily explore the effects of other proven techniques such as intermittent energization, which might allow the use of somewhat lower voltage corona discharge electrodes. One of the model capabilities is in providing, as an output, the electric field in the ash layer under each electrode element. This is used as an indicator of the presence of reverse ionization, which is especially important when there is a very large space charge due to a high ash concentration in the coal; a large change in the space charge between the inlet and outlet of a section could result in sufficiently high currents at the outlet to cause reverse ionization.
The proactive methodology using a combination of ESP modelling combined with knowledge and experience has matured into a viable and usable approach for evaluating performance upgrade scenarios. Furthermore, the use of this approach provides the ability to rapidly and economically optimise precipitator performance by providing low cost solutions for achieving desired levels of particulate emissions. It is applicable to precipitators collecting ash from the combustion of all types of coals giving both favourable and difficult fly ashes. It is especially useful for evaluating precipitator upgrades for the collection of difficult high resistivity, high concentration ashes arising in many countries, including India, China and Russia.
Figure 5. VI curves showing effects of improved electrode formats
The success of the approach to date has encouraged the wider application of the proactive modelling approach to the whole coal based electricity generation industry. The examples included here and in the references touch on what can be done by the proactive approach. The ESPVI 4.0 series of models are freely available; a number of individuals in the ESP community have already made use of it to varying degrees, depending upon their skills and knowledge. It is expected that as more engineers become exposed to the proactive approach for evaluating lowest-cost ESP upgrade scenarios its use will increase.
Parker, K., Plaks, N. Electrostatic Precipitator (ESP) Training Manual, U.S. Environmental Protection Agency, Washington, DC, EPA-600/R-04-072, July 2004
Parker K.R, Plaks N., Zykov A.M; Kolchin K.I; Konovalov, V.K ESP Performance Enhancement Analysis by Advanced Modelling Techniques, Electric Power 2004 Conference. Baltimore, Apr 27-29, 2004.
Zykov, A. M, Kolchin, K. I., Tumanovsky, A. G, Jozewicz, W. Joint Russian-American Project to Enhance Performance of Electrostatic Precipitators Used in Power Plants in the Newly Independent States, Mega Symposium 2004, Washington, DC, Aug 30-Sep 2, 2004.
Parker, K. R., Electrical Operation of Electrostatic Precipitators, Institute of Electrical Engineers, UK, 2003. ISBN 0-85296-137-5.