1 Introduction
The main factors affecting the performance of an electrostatic precipitator (ESP) are its structure, high-voltage DC power supply, and operating conditions. The structure primarily consists of corona electrodes, collecting electrodes, and a rapping unit, and includes parameters such as the collecting area and electrode spacing. Operating conditions mainly include the properties of the flue gas and dust. Given a fixed ESP structure and a power supply that meets the requirements of the operating conditions, can improvements be made to enhance the overall performance of the ESP? This is a topic worthy of in-depth exploration.
2. Physical Processes of Electrostatic Precipitation
The corona electrode (cathode) of a horizontal ESP is pointed, while the dust collection electrode (anode) is plate-shaped, with a typical distance of 150-200 mm between them. During operation, the corona electrode is connected to the negative terminal of the high-voltage DC power supply, and the dust collection electrode is connected to the positive terminal of the high-voltage DC power supply and grounded as a zero reference potential. The electric field formed between the two electrodes is a highly non-uniform electric field.
Dust removal efficiency formula of Duoyiqi:
In the formula: η is the dust removal efficiency; when the operating conditions are constant, A (dust collection area) / Q (gas flow rate) is a constant; v is the driving speed of charged dust particles moving towards the electrode (mainly referring to the dust collection electrode).
From formula (1), it can be seen that the dust removal efficiency η increases exponentially with the increase of the driving velocity v of the charged dust particles. v is mainly proportional to the charge of the charged dust particles; the larger the charge, the greater the effect of the electric field force. A larger charge requires a larger density of free electrons and negative ions in the outer corona region, a larger density of free electrons, positive ions, and negative ions generated per unit time in the inner corona region, and a stronger self-sustaining corona discharge.
How can a stronger self-sustaining corona discharge be achieved? Townsend theory provides an explanation, suggesting that self-sustaining discharge is mainly formed by the collisional ionization of electrons and the release of secondary electrons by positive ions at the cathode. Continuous collisional ionization forms an electron avalanche corona discharge, i.e., the α (electron ionization coefficient) process. α represents the average number of free electrons (indicating the average number of collisions between an electron and gas particles per 1 cm of travel from the cathode to the anode.)
Mathematical expression:
In the formula: A and B are constants related to the properties of the gas; P is atmospheric pressure; and E is electric field strength.
As can be seen from formula (2) , the α value is very sensitive to changes in the E value, exhibiting an exponential response. The E value is proportional to the output voltage Vd value of the applied high-voltage DC power supply, meaning that the α value increases exponentially with the increase of the DC voltage Vd amplitude . The larger the α value, the greater the density of electrons and ions in the air. However, the Vd amplitude cannot be increased indefinitely. When the gap between the corona electrode and the dust collecting electrode is constant, Vd has a limit value Vdmax . Exceeding this value will cause spark discharge (gas breakdown), resulting in an interruption of the dust removal process and a reduction in dust removal efficiency. The volt-ampere characteristics of air conductivity illustrate this process, as shown in Figure 1.
Figure 1. Current-voltage characteristics of air conduction
When Vd ≥ 5 ( Vdmax ), spark discharge occurs in this stage, causing the ionization current Id to increase sharply and the DC voltage Vd to drop sharply. Usually , the high-voltage power supply will immediately stop supplying power.
When 4 < Vd < 5, this stage is the self-sustaining corona discharge stage, and formula (2) is fully reflected here . Moreover, the closer Vd is to Vdmax , the larger the value of α becomes. This stage is the ideal working stage for electrostatic dust removal.
Townsend elucidated the basic process of gas discharge using the collision theory of discharge, but his explanation of some discharge phenomena when the gap (δS) > 0.26 cm is not yet perfect. For example, how does the change in dust concentration affect spark discharge during ESP operation?
The flow column theory provides an explanation for this, arguing that collisional ionization of electrons and spatial photoionization (secondary electron avalanche) are the main factors in the formation of self-sustaining discharge, and emphasizing the role of the space charge distortion electric field. A flow column is a pile of charged particles under the influence of an electric field between two electrodes; it is a mixed channel and ionization channel filled with positive and negative charged particles. When the flow column connects the two electrodes, it leads to spark discharge. Therefore, as the dust concentration increases, the increased number of charged dust particles accelerates the extension of the flow column, allowing spark discharge to occur even at relatively low Vd amplitudes.
During spark discharge, electrons and ions jump directly from one electrode to the other, so the probability of the dust becoming charged approaches zero. At this time, the AC input current of the power supply is typically about five times its rated value, and the minimum spark delay time is approximately 10ms. Therefore, spark discharge reduces dust removal efficiency and consumes electrical energy, which is detrimental to the operation of electrostatic precipitators.
In summary, the following control strategy is proposed: In order to achieve the goal of improving the efficiency and saving energy of electrostatic dust removal, the operating point of the high voltage DC voltage Vd amplitude should be controlled in the self-sustaining corona discharge stage, and should be as close as possible to the Vdmax value, i.e. the critical spark point, so that α (electron ionization coefficient) is stably maintained at the maximum value, but spark discharge conditions should be avoided as much as possible.
This control strategy allows for the analysis of waveform parameters and different adjustment rules of the high-voltage DC output voltage under different main circuit topologies. Here, minimizing the ratio (ripple coefficient) of the average output voltage Vd to the peak voltage Vp is fundamental to ensuring stable operation at the critical spark point.
3. Waveform parameters and power supply of DC output voltage for different main circuit schemes
DC power supplies are classified into linear power supplies and switching power supplies according to their working principles. Currently, the high-voltage DC power supply used for electrostatic precipitators is mainly a linear thyristor AC phase-controlled voltage regulation power supply, which is divided into single-phase thyristor AC phase-controlled voltage regulation scheme and three-phase thyristor AC phase-controlled voltage regulation scheme in terms of main circuit structure.
3.1 Single-phase thyristor AC phase-controlled voltage regulation
The main circuit block diagram is shown in Figure 2. ~Va and ~Vb are single-phase AC input voltages (380V/50Hz); SCR1~SCR2 are thyristors; TM is a single-phase step-up transformer; V1 is a single-phase high-voltage silicon stack rectifier bridge.
Figure 2. Block diagram of single-phase thyristor AC phase-controlled voltage regulation principle
When the firing angle α = 0°, the relationship between the average DC voltage Vd at the output of the rectifier bridge and the effective AC voltage V2 (TM secondary side) at the input of the rectifier bridge is as follows:
If the voltage Vd applied to the corona electrode and the dust collecting electrode is 72 (KV), according to formula (3), V2 = 80 (KV), the relationship between its peak voltage Vp (see Figure 3) and V2 is as follows:
Figure 3. Voltage waveform of single-phase bridge rectifier
Vp = 113 (KV), indicating that Vp is much larger than Vd , and is the main factor affecting the electron ionization coefficient α in formula (2). Figure 3 shows that the envelope of the Vp waveform is a combination of half-sine waves, making it difficult to maintain the electron ionization coefficient α at its maximum value. To improve this, the DC output voltage Vd must be increased, and the peak voltage Vp will be even higher. When Vp ≥ Vdmax persists for a certain period, spark discharge is very likely to occur. This makes it difficult to meet the requirements of the control strategy.
3.2 Three-phase thyristor AC phase-controlled voltage regulation
The main circuit block diagram is shown in Figure 4. ~ Va , ~ Vb , and ~ Vc are the three-phase AC input line voltages (380V/50Hz); SCR1 ~ SCR6 are thyristors; TM is a three-phase step-up transformer; and V1 is a three-phase high-voltage silicon stack rectifier bridge.
When the firing angle α = 0°, the relationship between the average DC voltage Vd at the output of the rectifier bridge and the effective AC voltage V2 (TM secondary side) at the input of the rectifier bridge is as follows:
If the voltage Vd applied to the corona electrode and the dust collecting electrode is 72 (KV), according to formula (5), V2 = 53.3 (KV). The relationship between the peak voltage Vp (see Figure 5) and V2 is as follows: as in formula (4), Vp = 75.4 (KV). Therefore, the value of Vd is approximately equal to the value of Vp , which is the main factor affecting the electron ionization coefficient α in formula (2). As shown in Figure 5, the waveform of Vd is approximately the waveform of a pure DC voltage, which makes it easy for the electron ionization coefficient α to be stably maintained at its maximum value. When Vd is changed, the peak voltage Vp is less likely to cause spark discharge. It is easy to meet the requirements of the control strategy.
Figure 5. Voltage waveform of three-phase bridge rectifier
3.3 Power Supply
In actual engineering, when the rated high voltage DC output voltage Vd is 72 (KV) and the rated high voltage DC output current Id is 1.2 (A), typically: the rated current Iab on the input side of a single-phase step-up transformer™ is 338 (A) and the rated voltage Vab is 380 (V); the rated current Iab on the input side of a three-phase step-up transformer™ is 138 (A) and the rated line voltage Vab is 380 (V).
Therefore, the rated input apparent power S_single -phase for the single-phase scheme is 128.44 (KVA); the rated input apparent power S_three -phase for the three-phase scheme is 90.83 (KVA).
In addition, compared with the single-phase scheme, the three-phase power supply is balanced, with equal voltage, current and magnetic flux in each phase and phases differing by 120°, thus avoiding phase pollution to the power grid.
4 Control schemes with different regulation laws
Commonly used control principles in engineering include: spark tracking (spark rate setting); maximum average voltage tracking; and constant power control. Among these, spark tracking (spark rate setting) is the most basic control principle, and currently, almost 95% of power supply equipment in industrial sites mainly uses this control method.
Based on the above control strategy, the following adjustment law is proposed: tracking the critical spark (no spark) point.
4.1 Spark Tracking (Spark Rate Tuning) Scheme
The system tracks changes in the high-voltage DC output current Id value based on feedback signals, automatically adjusting the firing angle of the thyristor to control the spark frequency (number of sparks per minute) between the corona electrode and the dust collector electrode, maintaining a constant operating range within the set spark frequency range. This ensures that the operating point of the high-voltage DC output voltage Vd always tracks the spark discharge voltage point, achieving the goal of maximizing Vd .
This scheme aims to maintain the maximum value of Vd by increasing the frequency of spark discharge, and therefore should not be considered an ideal control method. It is especially undesirable when the main circuit uses single-phase thyristor AC phase-controlled voltage regulation.
4.2 Scheme for Tracking Critical Spark (No Spark) Point
The system tracks the changes in the high-voltage DC output current Id and the high-voltage DC output voltage Vd based on the changing trend of the current-voltage characteristics of the dust concentration. The firing angle of the thyristor is adjusted accordingly to ensure that the operating point of the high-voltage DC output voltage Vd always tracks the critical spark discharge voltage point, thereby achieving the maximum value of Vd , while minimizing the generation of spark discharge.
This scheme also aims to maintain the maximum value of Vd , but it does not achieve this by increasing the frequency of spark discharge, making it an ideal control scheme. Especially when the main circuit uses three-phase thyristor AC phase-controlled voltage regulation, it ensures that the operating point of the high-voltage DC output voltage Vd always tracks the deep critical spark discharge voltage point, making it the recommended scheme.
5. Dust Removal Effect Industrial Test
To demonstrate the correctness of the control strategy, two comparative experiments were conducted. Three high-voltage DC power supplies with different characteristics were used in the experiments. The first type: The regulation law used a tracking critical spark (no spark) point control mode, and the main circuit used a three-phase thyristor AC phase-controlled voltage regulation scheme; this was called power supply A. The second type: The regulation law used a spark tracking (spark rate setting) control mode, and the main circuit used a single-phase thyristor AC phase-controlled voltage regulation scheme; this was called power supply B. The third type: The regulation law used a spark tracking (spark rate setting) control mode, and the main circuit used a three-phase thyristor AC phase-controlled voltage regulation scheme; this was called power supply C.
5.1 Experiment 1
Electrostatic precipitator load characteristics: flue gas emissions from a thermal power plant boiler; Electrostatic precipitator (ESP) layout and power zoning: 300mm electrode spacing, 1 chamber with 3 electric fields. Test procedure: Under the same operating conditions, all 3 electric fields used power supply A for 24 hours of continuous operation, and relevant data were measured; all 3 electric fields used power supply B for 24 hours of continuous operation, and relevant data were measured. Dust removal efficiency is shown in Table 1.
Table 1 Monitoring Report Form
Table 1 shows that using power source A results in an emission dust concentration of 22.8 mg/ Ndm³ , while using power source B results in an emission dust concentration of 91.0 mg/ Ndm³ . Therefore, power source A exhibits superior dust removal performance compared to power source B.
5.2 Experiment 2
Electrostatic precipitator load characteristics: Emissions from alumina clinker rotary kiln tail gas, with recovery of raw materials from the tail gas; Electrostatic precipitator (ESP) layout and power zoning: 400mm electrode spacing, 2 chambers, 3 electric fields. Test procedure: Under essentially the same operating conditions, chamber 1 (3 electric fields) of clinker kiln #1 was continuously operated using C-type power supply (already in operation), and relevant data were measured; chamber 1 (3 electric fields) of clinker kiln #2 was continuously operated using A-type power supply (already in operation), and relevant data were measured. Dust removal efficiency is shown in Table 2.
Table 2 Monitoring Report
Table 2 shows that using power supply A results in a dust concentration of 24 mg/ Ndm³ , while using power supply C results in a dust concentration of 32 mg/ Ndm³ . Therefore, power supply A demonstrates superior dust removal performance compared to power supply C.
6. Conclusion
This paper theoretically analyzes the physical process and power supply technology of electrostatic precipitators, and proposes a control strategy for the power supply technology to improve the efficiency and energy saving of electrostatic precipitators. The control strategy and power supply scheme were verified through field industrial tests. The verification results show that the control strategy of the power supply technology is correct, and the adjustment law adopting the critical spark (no-spark) point tracking mode and the main circuit using a three-phase thyristor AC phase-controlled voltage regulation scheme (A power supply) is a relatively ideal high-voltage power supply scheme for electrostatic precipitators.
