2外文翻译

嘉兴学院本科生毕业设计

嘉兴学院毕业设计外文翻译

题目（外文）： New active power filter and control method 题目（中文）： 新型有源电力滤波器及其控制方法 学院名称： 机电学院 专业班级：电气051学生姓名： 刘尚宝 一、外文摘要及关键词

ABSTRACT:A new active power filter with harmonic suppression and adjustable reactive power

compensation is propose. The power converter of the proposed filter can generate a compensating voltage,including a harmonic voltage for compensating the load harmonic current and a fundamental voltage for adjusting the reactive power of the utility side, connected to the power feeder via an impedance. The waveform of the utility current after compensating by the proposed active power filter can be approximated as a sinusoidal waveform,and the input power factor is also improved. A 20KVA prototype has been developed and tested to verify its performance. The test results indicate that the proposed active power filter has the desired performance.

KEY WORDS:active power filter, harmonic, reactive power compensation

二、中文摘要及关键词

摘要：本文提出了一种能抑制谐波和无功补偿可调的新型有源电力滤波器。滤波器中的电源转换

器串联电阻后连接到电力馈线上，能产生补偿电压，包括一个用来补偿负载谐波电流的谐波电压和一个用来调整电源侧无功功率的基波电压。电源电流的波形经过新型有源电力滤波器补偿后可近似为正弦波，并使输入功率因数得到改善。一个20千伏安的样机已经研制成功，并经过测试验证了其性能。试验结果表明，该新型有源电力滤波器具有预期的性能。

关键词：有源电力滤波器，谐波，无功功率补偿

三、外文正文

1 introduction

Power electronics related facilities may generate a significant harmonic current due to the nonlinear input characteristic of such loads. This harmonic current will pollute the power system, resulting in

- 1 -

嘉兴学院本科生毕业设计

problems such as transformer overheating, rotary machine vibration, degrading voltage quality, damage of electic power components, medical facility malfunctions ect[1]. To relieve the problem of harmonic pollution effectively, many harmonic limitation standards, such as IEEE519-1992, IEC1000-3-2, IEC1000-3-4 ect. Have been established. Therefore,solving the problem of harmonics is an important topic for today's power systems. Conventionally,passive power filters have been used to solve this problem, although they have some disadvantages,such as resonance and poor filter performance[2]. Recently,many harmonic suppression facilities based on power electronics techniques have been developed. Some of them can simultaneously suppress the different order harmonic components of nonlinear loads using only the same harmonic suppression equipment referred to as an active power filter [3-9]. Figure 1 shows a single-line system diagram of a conventional active power filter. This system includes a filter inductor, a power converter and a DC capacitor. The power converter is used to generate a compensating current via the filter inductor to inject into a power feeder. The filter inductor is used to suppress high-frequency ripple current caused by the switching behaviour of the power electronics devices in the power converter. The inductance of the filter inductor depends on the switching frequency, the DC voltage and the ripple current limitation. The DC capacitor located in the DC bus of the power converter acts as an energy buffer. Although, a conventional active power filter is capable of suppressing harmonics, it has the following disadvantages:

Fig. 1 System single-line diagram of conventional active power filter

1 To suppress the ripple current generated by the switching operation of power electronic devices in the power converter, a filter inductor with a large inductance is used.

2 A higher DC bus voltage for the power converter is required,which results in high switching power loss and high voltage rating of the DC capacitor and power electronic devices.

3 Using a larger filter inductor will result in a larger power loss, poor energy efficiency, more heat dissipation, bulk dimension and weight.

4 Using a larger filter inductor also results in degrading the high-frequency response.

- 2 -

嘉兴学院本科生毕业设计

Therefore, these above disadvantages limit applications of the conventional active power filter. Another solution for the harmonic problem is to adopt a hybrid power filter consisting of a power converter in series with a passive power filter [10

–15]

. Figure 2 shows a conventional hybrid

power filter comprising a passive power filter serially connected to a power converter.A passive power filter consists of one or many sets of tuned filters, and it is used to filter the dominant harmonic components. Hence,the capacity of the power converter can be reduced. However, it also has some disadvantages:

Fig. 2 System single-line diagram of hybrid power filter 1 The parameters of the passive elements in the passive power filter may have errors in the manufacturing process. Thus, the capacity of the active power filter cannot be significantly reduced if the parameters of the passive elements in the passive power filter are inaccurate.

2 Generally, the core of the inductor used in the passive filter is made of an iron alloy, such as silicon steel plates, which results in large size, heavy weight and large loss. Also, there is a significant power loss when high-frequency ripple current produced from the power converter is flowing through the inductor made by the silicon steel plates. This results in an increase of heat in the inductor.

3 To reduce the size of the passive power filter, the inductance of the passive power filter must be reduced and the larger capacitance of capacitor must be used. However, when this passive power filter is operated under a light load, the large capacitance may cause a large leading current that degrades the power factor and increases the bus voltage.

This paper proposes a new active power filter and control method. The proposed active power filter has harmonic suppression and adjustable reactive power compensation. The utility current, after compensation by the proposed active power filter,can be approximated as a sinusoidal waveform, and the input power factor under the light load is improved as compared with the conventional hybrid power filter.To demonstrate the performance of the proposed active power filter, a 20 KVA prototype is

- 3 -

嘉兴学院本科生毕业设计

developed and tested.

2 System configuration and operation principle

The system configuration of the proposed active power filter is shown in Fig. 3. It consists of a series-connected inductor and capacitor set, a DC capacitor, a power converter and a high-frequency ripple filter. Voltage-mode control is used to control the power converter. The power converter generates a compensating voltage that is converted into a compensating current flowing through the series-connected inductor and capacitor set, and the compensating current flows into the power feeder in order to filter harmonic currents and compensate for the reactive power generated by nonlinear loads.The configuration of the proposed active power filter is similar to that of the hybrid power filter shown in Fig. 2. However,the function and dimension of the passive elements (L–C) are not the same. In the proposed active power filter, the inductor of the series-connected inductor and capacitor set is very small, and it is used to filter the switching ripple of the power converter. The capacitor in the series-connected inductor and capacitor set is used to supply a basic reactive power.However, the passive elements (L–C) in the hybrid power filter are used to tune the dominant harmonic component of the load current. The inductance in the hybrid power filter is larger than that used in the proposed active power filter, so the size and weight of the inductor used in the hybrid power filter are also larger than those used in the proposed active power filter. The high-frequency ripple filter is configured by a set of capacitors and a resistor, and it is used to filter out further the switching ripples form the power converter.

Fig. 3 System configuration of proposed active power filter

Fig. 4 shows the equivalent circuit of the proposed active power filter. It consists of two voltage

- 4 -

嘉兴学院本科生毕业设计

sources: one is the utility and the other is the compensating voltage source generated by the power converter.The compensating voltage generated by the power converter is a dependent voltage source whose voltage depends on both the harmonic components and the fundamental reactive component of the load current. The equivalent circuit shown in Fig.4 can be further divided into the fundamental frequency equivalent circuit and the harmonic frequency equivalent circuit shown in Fig.5. Figure 5a shows the equivalent circuit under the fundamental frequency where the impedance of the series-connected inductor and capacitor set is capacitive under the fundamental frequency. To adjust the generated fundamental reactive current, the power converter must generate a fundamental voltage whose phase is the same as that of the utility voltage. The fundamental voltage Vc1 (RMS) across the capacitor can be represented as:

Vc1=Vs1-Va1 (1)

Fig. 4 Equivalent circuit of proposed active power filter

Fig. 5 Equivalent circuit of proposed active power filter

a Equivalent circuit at fundamental frequency b Equivalent circuit at harmonic frequency

where Vs1 and Va1 are the phase RMS values of the fundamental component for utility voltage and power converter output voltage. Then,the reactive power Qa supplied from the active power filter can be derived as

Qa=Qc(1-Va1/Va1) (2)

- 5 -

嘉兴学院本科生毕业设计

where Qc is the basic reactive power generated by the power capacitor of the series-connected inductor and capacitor set under the utility fundamental voltage. From (2), it can be found that the proposed active power filter can supply an adjustable reactive power by adjusting the fundamental component of power converter’s output voltage.

Figure 5b shows the equivalent circuit under harmonic frequency.If the frequency is lower than the resonant frequency, then the series-connected inductor and capacitor set is capacitive. However, the series-connected inductor and capacitor set is inductive if the frequency is higher than the resonant frequency. The switching frequency of the power converter is significantly higher than the resonant frequency of the series-connected inductor and capacitor set. As a result, the series-connected inductor and capacitor set acts as an inductor to filter the switching frequency of the power converter.To suppress the load harmonic current, the desired compensating voltage can be derived as: Vah=ZLCILH (3)

where Ilh is the harmonic component of the load current, and ZLC is the impedance of the series-connected inductor and capacitor set. If the power converter can generate a voltage as shown in (3), this voltage is converted into a compensating current that is opposite to the load harmonic current. Hence, the load harmonic current can be suppressed. Equation (3) shows that the desired compensation voltage is dependent on the load harmonic current and the impedance of the series-connected inductor and capacitor set.

From the operation theory of the bridge power converter, the DC bus voltage of a power converter must be higher than the peak value of the utility voltage in a conventional active power filter.However,a series-connected inductor and capacitor set is used to connect in series with the power converter, and it can block the most fundamental component of utility voltage. Hence the DC bus voltage in the proposed active power filter can be smaller than the peak value of the utility voltage, and it is only dependent on the amplitude of the compensating voltage, which is smaller than the peak value of the utility voltage. This means that the DC bus voltage in the proposed active power filter can be significantly reduced as compared with the conventional active power filter. Consequently, the switching power loss and the voltage rating of DC capacitor and power electronics devices can also be reduced. Furthermore, the ripple current of the power converter is dependent on the DC bus voltage and filter inductance. This implies that, the lower the DC bus voltage, the smaller filter inductance required for specified ripple current limitation. Therefore, the filter inductance used in the series-connected inductor and capacitor set is smaller due to the lower DC bus voltage.

- 6 -

嘉兴学院本科生毕业设计

Compared to the conventional active power filter, it can be seen that the proposed active power filter uses three additional AC capacitors to reduce the inductance of the three-phase filter inductor,the voltage rating of the DC capacitor, the voltage rating of the power electronics devices and the size of the heatsink. In practice,the core of an inductor with large inductance is made from the iron alloy, which results in large size, heavy weight and large power loss. In contrast, the core of an inductor with small inductance can be made from ferrite materials, which have small volume, light weight and low eddy current loss [15]. The electromagnetic interference (EMI) generated by the switching of the power converter, is also dependent on the DC bus voltage. Therefore, the salient advantages of the proposed active power filter are low voltage rating of the DC capacitor and power electronic devices, smaller filter inductance, smaller dimension, light weight, good filter performance and low EMI.In addition, the smaller filter inductance can improve the high-frequency response performance of this active power filter. Since the capacity of the DC bus voltage is dependent on the amount of compensation current, rather than the utility voltage, the application of the proposed active power filter can be extended to a wider voltage range. In a limited variable voltage application, such as in the range 220–480 V, the change in the main components is only the voltage rating of the series-connected inductor and capacitor set. However, the voltage rating of both active and passive components must be changed in the conventional active power filter. In addition, the proposed active power filter can be applied in 50/60 Hz power systems by adjusting only the parameters of the control circuit. However, the L-C parameters of the passive power filter must be changed in the hybrid power filter since the dominant harmonic frequency is different in 50/60 Hz power systems.

The above indicates that the proposed active power filter has the performance of suppressing harmonic current and providing an adjustable reactive current. When the active power filter is operated under light load, the harmonic load current is small. The power converter is used mainly to generate a fundamental voltage to reduce the reactive current supplied by the proposed active power filter. When the proposed active power filter is operated under heavy load, the power converter is not used to generate the fundamental voltage but a harmonic voltage. Thus, the entire fundamental component of the utility voltage will drop on the series-connected inductor and capacitor set that produces a maximum reactive power current. To improve the input power factor at the utility current side, the active power filter is able to produce an adjustable reactive current according to either the light or heavy load conditions. Consequently, the proposed active power filter can improve the disadvantage that the reactive power generated from the conventional hybrid power filter is constant. Compared to the

- 7 -

嘉兴学院本科生毕业设计

conventional active power filter, the proposed active power filter has cost advantage due to the lower voltage rating of the DC capacitor, lower voltage rating of power electronic devices and the smaller size of the heat-sink. Hence, the hardware cost of the proposed active power filter is very competitive for nonlinear loads whose input is a diode-rectifier or phase-controlled rectifier with a low-level voltage below 480V.

3 Control method

Conventionally, the active power filter has been controlled by the current mode. However, this is very difficult to implement under low filter inductance because of the high switching ripple, and it may generate multiple crossings during a carrier period of the pulse-width modulator. This phenomenon of multiple crossing will result in more than one switching operation during a carrier period. To resolve this, the proposed active power filter uses voltage-mode control. The three-phase power converter controlled by voltage-mode control acts as a voltage amplifier with the gain represented by:

Kcon=Vdc/2Vtri (4) where Vdc is the DC bus voltage and Vtri is the amplitude of the carrier signal of the pulse-width modulator. Hence, the control circuit of the voltage-mode controlled power converter is used to determine a reference voltage by dividing the desired compensating voltage by the gain shown in(4). From the above, it can be found that the desired compensating voltage generated by the power converter for filtering the load harmonic current is derived in(3). Hence, the first control signal v1(t) can be further derived from (3), and it is represented as:

(5)

Where L and C are the inductance and capacitance of the series-connected inductor and capacitor set,and R is the component stray loss of the active power filter.

If the power converter can generate a harmonic voltage equal to the first control signal v1(t) and convert it into a compensating current by the series-connected inductor and capacitor set, then theoretically the harmonic components of the load current can be compensated. However, the filter performance is degraded due to the parameters of the series-connected inductor and capacitor set, which may vary due to age, variable frequency, production and temperature in practice. To improve the compensating performance, a second control loop must be used to modify the error of the compensating

- 8 -

嘉兴学院本科生毕业设计

result. The concept of the second control loop is based on the theory of the conventional hybrid power filter[10

–12]

. The second control signal v2(t) is obtained by detecting the harmonic components of the

utility current and then amplifying with a gain (k1), and it can be represented as: v2(t)=k1ish(t)

(6)

where ish(t) is the harmonic component of the utility current.If the power converter can generate a voltage equal to the second control signal v2(t), the utility harmonic current can be derived from Fig.4b and represented as:

(7)

where ILh is the harmonic component of the load current.From (7), it can be found that a term k1 is added to the denominator when the power converter generates a voltage as the second control signal v2(t). Hence, the second control loop is used to control the power converter to act as a virtual harmonic resistor k1. The virtual harmonic resistor k1 is in series with the utility to block the uncompensated harmonic components of load current flowing back to the utility[12]. In the proposed active power filter, the first control loop acts as the rough tuning, and the second control loop is used for the fine tuning. Owing to the use of voltage-mode control in the proposed active power filter, the series-connected inductor and capacitor set may result in high-frequency oscillation between the utility and the active power filter. Hence,a third control loop is applied to avoid high-frequency oscillation.The third control loop is used to generate a virtual harmonic resistor to be connected in series with the series-connected inductor and capacitor set.The virtual harmonic resistor acts as a harmonic damper. This can be realised by using the power converter to generate a harmonic voltage that is proportional to the harmonic components of the active power filter current.Hence, the third control signal v3(t) can be represented as:

v3(t)=k2iah(t) (8)

where iah(t) is the harmonic current component of the active power filter. Hence, the power converter can act as a virtual harmonic resistor (k2).

A DC capacitor located at the DC bus of the voltage-source power converter is used to supply a DC voltage to the power converter and act as an energy buffer. The DC bus voltage is expected to be a constant voltage. However, the virtual harmonic resistor in the second and third control loops and the switching loss of power converter will consume the real power. Then, the voltage variation in the DC bus cannot be avoided. To maintain a constant DC bus voltage, the fourth control loop is used. Hence,

- 9 -

嘉兴学院本科生毕业设计

the voltage regulation of the DC bus voltage can be obtained by using the power converter to generate a fundamental voltage in phase or out of phase with the fundamental component of the active power filter current. The fourth control signal v4(t) can be represented as:

v4(t)=k3ia1(t) (9)

Whrer ial(t) is the fundamental component of the active power filter current.Then,the power converter acts as a positive/negative fundamental resistor absorb/regenerate the real power from/to the utility,to maintain the DC bus voltage at a constant value.

To adjust the reactive power supplied from the active power filter, the power converter must generate a fundamental voltage that is in phase with the utility fundamental voltage. Equation (2) shows that controlling the amplitude of this fundamental voltage can control the compensating reactive power.The fifth control signal v5(t) can be represented as: v5(t)=k4vs1(t)

(10)

where vs1(t) is the fundamental component of the utility voltage. The major function of the proposed active power filter is the harmonic suppression, and the reactive power compensation is the minor function. Hence, the priority of the harmonic suppression is higher than the compensating reactive power, and the compensating reactive power is proportional to the amplitude of load harmonic current. The maximum compensating reactive power is the basic reactive power for the power capacitor of the series-connected inductor and capacitor set under the utility fundamental voltage,and the power converter generates no fundamental voltage that is in phase with the utility voltage under full-load condition.

Finally, the reference signal for suppressing harmonic current and adjusting the reactive power can be obtained and represented as:

Vref(t)=v1(t)+v2(t)+v3(t)+v4(t)+v5(t) (11)

4 Control Block Diagram

Figure 6 shows the control block diagram of the proposed active power filter. It consists of five control loops. From Fig. 6, it can be found that five feedback signals, namely the load current, the utility current, the output current of power converter, utility voltage and the DC bus voltage, are used in the control circuit of the proposed active power filter to calculate the reference voltage of the power converter. The first control loop is used to implement the product of harmonic components of the load current and the impedance of the series-connected inductor and capacitor set shown in (5). The load current is detected

- 10 -

嘉兴学院本科生毕业设计

and sent to the band-rejection filter to filter out its fundamental component.

Fig.6 Control block of proposed active power filter

From(5), it can be found that the product of harmonic components of the load current and the impedance of the series-connected inductor and capacitor set can be obtained by feeding the harmonic components of load current to a proportional integral differential (PID)controller. The proportional, integral and differential coefficients are the resistance R, capacitance C and inductance L, respectively,as shown in (5). Then, the output of the first control loop is obtained. To avoid the effect of noise, a low-pass filter is used in the front of the differential controller and a high-pass filter is inserted at the end of integral controller to reject the DC component due to the initial condition. Since, the series-connected inductor and capacitor set is located at the output of the power converter, the capacitor also can block the DC component owing to the initial condition. Hence, the effect of the initial condition caused by the switch-on of the proposed active power filter can be suppressed.

To improve the compensating performance, the second control loop is used to modify the error of the compensating results of the first control loop. In the second control loop, the detected utility current is sent to the band-rejection filter to filter out the fundamental component.Then,the uncompensated harmonic components of the utility current are obtained. The output of the band-rejection filter is fed to the amplifier to obtain the output of the second control loop.

The third control loop is used to generate a virtual harmonic resistor to be connected in series with the series-connected inductor and capacitor set to act as a damper. The output current of the power

- 11 -

嘉兴学院本科生毕业设计

converter is sent to a bandpass filter to obtain the fundamental component, and then the detected output current of the power converter and its fundamental component are fed to a subtractor to obtain the harmonic components. The harmonic components are fed to an amplifier to obtain the output of the third control loop.

The fourth control loop is used to regulate the DC bus voltage. The fourth control loop is composed of a lowpass filter to filter out the DC bus voltage ripple and a subtractor to subtract a setting value from the output of low-pass filter. After this,the subtractor result is sent to a PI controller. The output of the band-pass filter is the fundamental component of the active power filter current, and the output of the fourth control loop is the product of the output of the PI controller and the output of the band-pass filter. The fifth control loop is used to adjust the compensating reactive power.First, the reactive power of load is calculated. The utility voltage is sent to a phase shift circuit to shift its phase by 90°and then multiplied to the load current; the product is sent to a low-pass filter to obtain the reactive power of the load. Since,the DC bus voltage is constant and the priority of the harmonic suppression is higher than the compensating reactive power in the proposed active power filter, the compensating reactive power must be limited. Hence,a limit circuit is used to restrict the compensated reactive power. The limit value of the limit circuit is varied and depends on the amount of compensated harmonic current. Under the no-load or light-load condition, the amount of compensated harmonic current is small, and the value of the limit circuit is large. This means that the power converter can supply a large reactive power in this condition. However, the value of the limit circuit is nearly zero under the heavy load condition. This means that the maximum reactive power is the reactive power supplied from the series-connected inductor and capacitor set in this condition. The utility voltage and the output of limit circuit are sent to a multiplier to obtain the output of the fifth control loop.

Finally, the modulated signal can be obtained by summing the outputs of the first, second, third, fourth and fifth control loops. Then, the modulated signal is sent to a pulse-width modulator in order to drive the power electronics devices of the power converter.

5 Experimental results

To demonstrate the performance of the proposed active power filter,a three-phase 20KVA prototype was devel-oped. The major parameters of the prototype are shown in Table 1. The utility power is supplied by a three-phase three-wire utility system operating at 380V and 60Hz. A comparison of the proposed active power filter and the conventional parallel active power filter is shown in Table 2. Because the

- 12 -

嘉兴学院本科生毕业设计

inductance of the series-connected inductor and capacitor set is only 200H, a ferrite core can be used to reduce the power loss, weight and volume. Hence, the volume and weight of the proposed active power filter is clearly smaller than that of the conventional parallel active power filter. In addition, the hardware cost is also reduced significantly owing to the low-voltage rating of DC capacitor and power electronic devices, small inductance of the filter inductor and small dimension of the overall system. The tested load is a six-pulse rectifier charger. Table 1: Major parameters of prototype

Table 2: comparison results

Fig. 7 Test results for proposed active power filter at steady state a Utility voltage b Utility current c Load current

d Active power filter current

- 13 -

嘉兴学院本科生毕业设计

Fig. 8 Test result of proposed active power filter under transient of applying nonlinear load a Utility voltage b Utility current c Load current

d Active power filter current

Fig. 9 Test results under the condition with large system impedance before applying proposed active power filter a Utility voltage b Utility current.

Figure 7 shows the experimental results of the proposed active power filter in the steady state. The load current shown in Fig. 7c is rich in harmonics and its total harmonic distortion(THD)is 26%. The waveform of the utility current shown in Fig.7b is nearly sinusoidal, and its THD is only 3% after compensation by the proposed active power filter. The test results show that the harmonic suppression

- 14 -

嘉兴学院本科生毕业设计

performance of the proposed active power filter is excellent.

Figure 8 shows the experimental results of the proposed active power filter under the transient of applying nonlinear load. As seen in Fig. 8b, the proposed active power filter has an excellent transient response.

In an industrial distribution power system, a turbine generator is often used for backup power. However, the power source of the turbine generator can be regarded as a weak power source because the capacity of the turbine generator is not large enough. The system impedance of a weak power source is large, and the utility voltage will be clearly degraded under the condition of nonlinear load. The experimental results shown in Figs.9 and 10 are obtained under the condition of large system impedance. An inductor with 0.6mH is inserted into the utility power feeder to simulate the large system impedance condition. As seen in Fig.9, the voltage waveform of the utility is seriously distorted owing to the nonlinear load. The THD of the utility voltage and the utility current are 9% and 39%, respectively.The distorted utility voltage may disturb the normal operation of the power equipment itself or the neighbouring load in the same power feeder. Fig.10 shows that both the voltage and current waveforms of the utility are nearly sinusoidal after applying the proposed active power filter. The THD of the utility voltage and current are 1.4% and 4.9%, respectively. Hence, this verifies that the proposed active power filter can not only suppress the input current harmonics but also avoid voltage waveform distortion under nonlinear loads.

Fig. 10 Test results under the condition with large system impedance after applying proposed active power filter a Utility voltage b Utility current

- 15 -

嘉兴学院本科生毕业设计

Fig. 11 Test results of proposed active power filter under condition of no load a Utility voltage b Utility current

Figure 11 shows the experimental result of the proposed active power filter under the condition of no load. The maximum fundamental reactive current can be obtained by calculating the utility voltage and the impedance of the series-connected inductor and capacitor set. The utility voltage is 380V; the RMS value of the fundamental reactive current of the active power filter is changed from 15A to 7.2A after compensation. This verifies that the proposed active power filter can supply adjustable reactive power.

6 Conclusions

The input features of power electronics related facilities are high input current harmonics and poor input power factor. In this paper, a novel active power filter and control method is proposed. The proposed active power filter can effectively suppress the harmonic current and supply an adjustable reactive power.It also has the advantages of lower voltage rating for DC capacitor and power electronic devices, smaller filter inductor, smaller dimensions, light weight, better filter performance and low electromagnetic interference(EMI).A three-phase 20KVA active power filter has been developed to demonstrate the performance of the proposed method. The experimental results show that the proposed active power filter has excellent performance in suppressing harmonic current. Furthermore, the hardware cost of the proposed active power filter is very competitive for harmonic loads whose input is a diode-rectifier or phase-controlled rectifier with a low-level voltage below 480 V.

- 16 -

嘉兴学院本科生毕业设计

7 Acknowledgement

The authors would like to acknowledge the UIS Abler Electronics Co.,Ltd.,the financial support and engineers helped with the hardware implementation.

8 References

1 Henderson, R.D., and Rose, P.J.:‘Harmonics: the effects on power quality and transformers', IEEE Trans.Ind.Appl., 1994,41, pp. 528–532

2 Wu, C.J., Chiang, J.C., Yen, S.S., Liao, C.J., Yang, J.S., and Guo,T.Y.:‘Investigation and

mitigation of harmonic amplification problems caused by singleuned filters’,IEEE Trans. Power Deliv.,1998,13, pp.800–806

3 Chiang, S.J., Ai, W.J., and Lin, F.J.:‘Parallel operation of capacity-limited three-phase four-wire active power filters’,IEE Proc., Electr. Power Appl., 2002, 149, (5), pp. 329–336

4 Wu, J.C., and Jou, H.L.:‘A simpli?ed control method for single-phase active power filter’,IEE Proc.Electr.Power Appl., 1996, 143,pp. 219–224

5 Akagi, H.:‘Trends in active power line conditioners’,IEEE Trans.Power Electron., 1994, 9, (3), pp. 263–268

6 Abellan, A., Benavent, J.M., and Garcera, G.:‘A new control method for obtaining reference currents of shunt active power filters in unbalanced and non-sinusoidal conditions’.Proc. IEEE ISIE Conf.,1999, pp. 831–836

7 Singh, B., Al-Haddad, K., and Chandra, A.:‘A new control approach to three-phase active power filter for harmonics and reactive power compensation’, IEEE Trans. Power System, 1998, 13, (1), pp. 133–138

8 Wu, J.C., and Huang, S.J.: ‘Design and operations of cascaded activepower filters for the reduction of harmonic distortion in a power system’, IEE Proc. Gener. Transm. Distrib., 1999, 146, (2), pp. 193–199

9 Kawabata, T., and Komatsu, Y.: ‘Characteristics of three phase active power filter using extension pq theory’. Proc. IEEE ISIE, Conf., 1997,pp. 302–307

10 Yang, J., Wang, Y., and Wang, Z.: ‘A DSP controlled hybrid power filter used to compensate the harmonics and reactive power caused by electrical traction loads’.Proc.IEEE PESC Conf.,2003,pp. 1615–1620

- 17 -

嘉兴学院本科生毕业设计

11 Fujita, H., and Akagi, H.:‘A practical approach to harmonic compensation in power system, series connection of passive and active filters’, IEEE Trans. Ind. Appl., 1991, 38, pp. 1020–1025

12 Dixon, J., Espinoza, J., Moran, L., and Rivas, D.: ‘A simple control scheme for hybrid active power filter’. Proc. IEEE PESC Conf., 2000,pp. 991–996

13 Rahmani, S., Al-Haddad, K.A., and Fnaiech, F.:‘A new control technique based on the instantaneous active current applied to shunt hybrid power filters’. Proc. IEEE PESC Conf., 2003, pp. 808–813

14 EI-Habrouk, M., Darwish, M.K., and Mehta, P.: ‘Active power filters: a review, IEE Proc., Electr. Power Appl., 2000,147,(5),pp. 403–412

15 Mohan, Undeland,and Robbins,:‘Power electronics converters,applications and design’ (Media Enhanced, John Wiley & Sons, Inc.,2003, 3rd edn.)

四、译文正文

1 引言

电力电子设备由于非线性输入特性，如负载，可产生显著的谐波电流。这种谐波电流会在电力系统中产生如变压器过热，旋转机械振动，较差的电压特性，电力元件损坏，医疗设施故障等问题。为了有效地减少谐波污染，很多限制谐波的标准已经制定出来，如IEEE519-1992，IEC1000-3-2，IEC1000-3-4等。因此，解决谐波问题是当今电力系统的一个重要议题。传统上用无源电力滤波器来解决这个问题，但是有一些缺点，如共振和拙劣的滤波性能。

近年来，在电力电子技术基础上已经开发出很多谐波抑制设备。这些设备中的有些能同时抑制非线性负载中不同的谐波成分，用同样的谐波抑制设备称为有源电力滤波器

[3-9][2]

[1]

。图1显示了一

个常规有源电力滤波器的系统单线图。这个系统包括一个滤波电感器，电源转换器和一个直流电容器。电力转换器用来将产生的补偿电流经滤波电感器注入电力馈线。滤波电感器是用来抑制电力电子器件在电源转换器中进行转换时所产生的高频率纹波电流。滤波电感器的电感取决于开关频率，直流电压和纹波电流值。直流电容器位于电源转换器的直流母线上，相当于一个能源缓冲器。虽然常规的有源电力滤波器能抑制谐波，但它有以下几个缺点：

- 18 -

嘉兴学院本科生毕业设计

Fig. 1 常规有源电力滤波器的系统单线图

1 抑制电源转换器、滤波电感器中电力电子设备的倒闸操作所产生的纹波电流需要一个很大的电感。

2 电源转换器需要一个较高直流母线电压，引起较高的开关功率损耗和需要高电压等级的直流电容器和电力电子设备。

3 使用较大的滤波电感器会产生一个较大的功率损耗，能源效率低，散热多，体积尺寸和重量大。 4 使用较大的滤波电感器，也造成高频率响应。

因此，上述这些不利因素限制了常规有源电力滤波器的应用。

另一种解决谐波问题的办法，是采取一系列电源转换器与无源电力滤波器构成的一种混合型电力滤波器

[10-15]

。图2显示了一个无源电力滤波器串联电源转换器组成的常规混合电力滤波器。一

个或多个调谐滤波器组成的无源电力滤波器，它是用来过滤主要的谐波成分。因此，电力转换器的容量可减少。不过，它也有一些缺点：

Fig. 2 混合电力滤波器的系统单线图

1 在制造过程中，无源电力滤波器的无源元件的参数可能有错误。因此，如果无源电力滤波器的无源元件的参数不准确，不能降低有源电力滤波器的容量。

2 一般来说，无源滤波器的电感是铁合金做的，如硅钢片，大尺寸和重量导致资源浪费。此外，当电力转换器产生的高频率的纹波电流流经由硅钢片组成的电感，产生明显的功率损耗，使得电感中的热量增加。

- 19 -

嘉兴学院本科生毕业设计

3 减小无源电力滤波器的尺寸，必须减少无源电力滤波器的电感和使用较大电容的电容器。但是，当操作这种负荷轻、容量大的无源电力滤波器，可引起很大的超前电流，降低了功率因数，增加了母线电压。

本文提出了一种新的有源滤波器和控制方法。这种有源电力滤波器具有谐波抑制和可调无功补偿。电源电流经过新型有源电力滤波器补偿后，可近似为正弦波形，在轻负荷条件下其输入功率因数与常规混合电力滤波器比较具有改善。为了证明该新型有源电力滤波器的性能，一个20千伏安的原型机已开发和测试。

2 系统配置及工作原理

有源电力滤波器的系统配置如图3所示，它包括了串联的电感器和电容器组，直流电容器，电源转换器，以及高频率的纹波滤波器。电压模式控制是用来控制电源转换器，电源转换器生成一个补偿电压，再转换成补偿电流流经该串联的电感器和电容器组，补偿电流流入电力馈线是为了过滤谐波电流和补偿非线性负载产生的无功功率。有源电力滤波器的配置与如图2所示的混合电力滤波器类似。不过功能及尺寸与无源元件（电感电容型）不相同。在有源电力滤波器中，串联的电感器和电容器组的电感非常少，而且它是用来过滤电源转换器中的开关纹波的。串联的电感器和电容器组中的电容器是用来提供基波的无功功率。不过，无源元件（电感电容型）在混合电力滤波器是用来调占主导地位的负载电流的谐波分量。混合电力滤波器中的电感比有源电力滤波器中的电感大，所以混合电力滤波器中的电感的尺寸和重量也大于有源电力滤波器中的电感。高频率纹波滤波器由一组电容器和电阻器组成，它是用来过滤掉更多的开关纹波，形成了功率转换器。

Fig. 3 新型有源电力滤波器的系统配置

图4所示的是有源电力滤波器的等效电路。它包括两个电压源：一种是有功电压源及另一种

是功率转换器产生的补偿电压源。电力转换器产生补偿电压，电力转换器是一个独立的电压源，其电压取决于负载电流的谐波成分和基波的无功成分。等效电路如图4所示，图4可进一步划分为如图5所示基波频率的等效电路和谐波频率等效电路。图5a所示的是基波频率下串联的电感器和电容器组的阻抗在基波频率是呈电容性的等效电路。调整产生的基波无功电流，功率转换器必须产

- 20 -

嘉兴学院本科生毕业设计

生一个与有功电压相位相同的基本电压。基波电压Vc1(RMS)通过电容器的公式可表示为：

Vc1=Vs1-Va1 (1)

Fig. 4 新式有源电力滤波器的等效电路

Fig. 5 新型有源电力滤波器的等效电路 a 基本频率时的等效电路 b 谐波频率时的等效电路 这里的Vs1和Va1分别是有功电压及电源转换器的输出电压相位有效值。因此，无功功率Qa提供给有源电力滤波器，可以推导出

Qa=Qc(1-Va1/Va1) (2) Qc是串联的电感器和电容器组中的电力电容器在电源基波电压下产生的基波无功功率。由式(2)可以得出,有源电力滤波器提供可调的无功功率，无功功率可以通过调整电源转换器输出电压的基波组成部分。

图5b所示的是谐波频率下的等效电路。如果频率低于共振频率，则串联的电感器和电容器组是呈容性的。反之如果频率高于共振频率，则串联的电感器和电容器组是呈感性的。电源转换器的开关频率明显高于串联的电感器和电容器组的共振频率。因此，串联的电感器和电容器组作为一个电感器，用来过滤电源转换器的开关频率。抑制负载谐波电流，理想的补偿电压可表示为： Vah=ZLCILH (3) Ilh是负载电流的谐波分量，Zlc是串联的电感器和电容器组的阻抗。如果电力转换器可以产生式(3)表示的电压，此电压转换成负载谐波电流相反的补偿电流。因此，可以抑制负载谐波电流。式(3)表明，理想的补偿电压依赖于负载谐波电流和串联的电感器和电容器组的阻抗。

- 21 -

嘉兴学院本科生毕业设计

从桥式电力转换器的操作理论分析，电源转换器的直流母线电压必须高于常规有源电力滤波器的电源电压的峰值。然而，串联的电感器和电容器组是用来串联电力转换器的，它可以阻止大多数的电源电压的基波组成部分。因此，有源电力滤波器的直流母线电压可小于有用电压的峰值，它仅仅依赖于补偿电压的振幅，补偿电压小于电源电压的峰值。这就是说，该新型有源电力滤波器的直流母线电压比传统的有源电力滤波器的直流母线电压明显降低很多。因此，直流电容器和电力电子设备的开关功率损耗和电压等级的也可降低。此外，电力转换器的纹波电流依赖于直流母线电压和滤波电感。这意味着，直流母线电压越低，需要指定纹波电流限制的滤波电感越小。因此，串联的电感器和电容器组中的滤波电感较小，是由于直流母线电压较低。

相比传统的有源电力滤波器，可以看出该新型有源电力滤波器用到了3个额外的交流电容器，以减少三相滤波电感器的电感，直流电容器的额定电压，电力电子设备和散热片的尺寸的电压和电力损失大。相反，感应器的核心是由氧化铁材料做成的小电感，具有体积小，重等级。在实际中，电感器的核心是由铁合金做成的大电感，使得尺寸大，质量重量轻，涡流损耗低

[15]

。电源转

换器的开关产生的电磁干扰(EMI)，也依赖于直流母线电压。因此，该新型有源电力滤波器的直流电容器和电力电子设备的电压等级低，滤波电感小，尺寸小，重量轻，具有良好的滤波性能和很低的电磁干扰等明显的优点。此外，较小的滤波电感可以改善有源电力滤波器的高频率响应。由于直流母线电压的容量取决于补偿电流，而不是实用的电压，有源电力滤波器的应用可以扩大到更大的电压范围。限制可变电压的应用，如在220-480V范围内，只能改变串联的电感器和电容器组的额定电压。不过，传统的有源电力滤波器中的有源和无源部分的额定电压必须改变。此外，该新型有源电力滤波器可用于50/60赫兹的电力系统中，只需调整控制电路的参数。不过，改变混合电力滤波器中无源电力滤波器中的电感电容参数，是由于在50/60赫兹的电力系统中主要的谐波频率是不同的。

上述表明，该有源电力滤波器具有抑制谐波电流，提供可调的无功电流的性能。当有源电力滤波器在轻负载下操作，谐波负载电流很小。电力转换器主要是用来生成一个基本电压，以减少有源电力滤波器的无功电流供应。当有源电力滤波器负荷重条件下操作，电力转换器是不能用来产生基波电压而是一个谐波电压。因此，电源电压的整个组成部分就会下降导致串联的电感器和电容器组产生最大的无功电流。为了提高有功电流的输入功率因数，有源电力滤波器在轻的或重的负载条件下能够产生一个可调无功电流。因此，有源电力滤波器可以改善传统的混合电力滤波器产生的无功功率是常数的劣势。相比传统的有源电力滤波器，该新型有源电力滤波器具有成本优势，由于直流电容、电力电子设备额定电压较低和散热面积较小。因此，有源电力滤波器的硬件成本对于非线性负载是极具竞争力的，其输入是二极管整流或相位控制整流器和一个低于480V的低级别的电压 。

- 22 -

嘉兴学院本科生毕业设计

3 控制方法

按照常理，有源电力滤波器已经可以用电流模式控制。然而，在低滤波电感下难以实施，因为它具有极高的开关纹波，而且在脉宽调制传送的时候可能产生复杂的通道。这种复杂通道在传送时期会引起一个以上的倒闸操作。要解决这个问题，有源电力滤波器需要用电压模式控制，用电压模式控制来控制三相电源转换器，作为一个电压放大器的增益可表示为： Kcon=Vdc/2Vtri (4)

Vdc是直流母线电压，Vtri是脉宽调制器载波信号的振幅。因此，电压模式控制的电力转换器的控制电路是用来确定一个参考电压除以理想的补偿电压得到公式(4) 。

从以上可以发现由式(3)可以推导出电力转换器产生理想的补偿电压是为了过滤负载谐波电流 。因此,首先控制信号V1(t)可以进一步推导出式(3)，可表示为：

(5)

其中L和C是串联的电感器和电容器组的电感和电容，R是有源电力滤波器其他部分的损耗。 如果电力转换器能产生谐波电压等于第一控制信号V1(t)和由串联的电感器和电容器组转换成的补偿电流，那么从理论上负载电流的谐波分量就可以得到补偿。然而，滤波性能的退化是由于串联的电感器和电容器组的参数，参数的变化是由于在实际使用中寿命、变频、生产和温度。为了提高补偿的性能，第二个控制回路必须用来修改补偿效果的误差。第二个控制回路的概念是在传统混合电力滤波器的理论基础上建立的

[10-12]

。第二个控制信号V2由检定的谐波成分的有功电

流乘以放大的增益（K1）获得，可表示为：

v2(t)=k1ish(t) (6) ish是有功电流的谐波成分，如果电力转换器产生的电压等于第二个控制信号V2(t)，有功谐波电流可由图4获得，表示为：

(7)

Ilh是负载电流的谐波分量，由式(7)可以发现K1项是加在分母中，当电力转换器产成一个电压作为第二个控制信号V2(t)。因此，第二控制回路是用来控制电力转换器以形成一个虚拟的谐波电阻K1。虚拟谐波电阻K1是在一系列有用阻止负载电流未补偿谐波成分再回到有用源电力滤波器中，第一个控制回路作为粗调，第二个控制回路作为微调。

由于在提供的新型有源电力滤波器使用的电压模式控制，串联的电感器和电容器组可能在电源及有源电力滤波器引起高频率振荡。因此，第三个控制回路的应用是避免高频率的振荡。第三

[12]

。在提供的有

- 23 -

嘉兴学院本科生毕业设计

个控制回路是用来生成一个虚拟谐波电阻成为连接一系列串联的电感器和电容器组。虚拟谐波电阻作为一个谐波阻尼器，实现电力转换器产生谐波电压，谐波电压与有源电力滤波器的电流谐波成分成正比。因此，第三个控制信号V3(t)可表示为:

v3(t)=k2iah(t) (8)

iah(t)是有源电力滤波器的谐波电流分量。因此，电力转换器可以作为一个虚拟谐波电阻(k2) 。 直流电容器位于电压源电力转换器直流母线上，用来提供直流电压给电力转换器，以及作为一个能源缓冲。直流母线电压被认为是恒电压。然而，虚拟谐波电阻在第二个和第三个控制回路和功率变换器的开关损耗将消耗真正的功率，直流母线上电压的变化不能避免的。保持直流母线电压相对恒定，使用第四控制回路。因此，直流母线电压电压调节的，利用电力转换器产生与有源电力滤波器电流基本成分同相或异相的基波电压。第四控制信号V4(t)可表示为：

v4(t)=k3ia1(t) (9)

ial(t)是有源电力滤波器电流的基本组成部分。因此电力转换器作为一个正或负基本电阻从有用的东西那里吸收或释放功率，以维持直流母线电压为恒定的值。

调整无功功率给有源电力滤波器，电力转换器产生一个与电源基波电压同相的基波电压。公式(2)表明，控制基波电压的振幅可控制无功功率补偿。第五个控制信号的V5(t)可表示为：

v5(t)=k4vs1(t) (10)

Vs1(t)是电源电压的基本组成部分。有源电力滤波器的主要功能是谐波抑制，次要功能是无功补偿。因此，谐波抑制的优先级高于无功功率补偿，补偿的无功功率与负荷谐波电流的振幅成正比。最大补偿无功功率是串联的电感器和电容器组在有用基本电压下电力电容器的基本无功功率，功率转换器产生的不是与满负载条件下电源电压同相的基波电压。

最后，可以得到抑制谐波电流和调整无功功率的参考信号，可表示为：

Vref(t)=v1(t)+v2(t)+v3(t)+v4(t)+v5(t) (11)

4 控制框图

图6所示的是有源电力滤波器的控制框图，它由五个控制回路组成。从图6，可以发现5个反馈信号，即负载电流，公用电流，电力转换器输出电流，公用电压和直流母线电压，是用在控制有源电力滤波器的电路来计算电力转换器的参考电压。第一个控制回路是用来执行该产品负载电流的谐波成分和图5所示的串联的电感器和电容器组的阻抗 。负载电流被发现并送往频带限制滤波器过滤出基本组成部分。

- 24 -

嘉兴学院本科生毕业设计

Fig.6 新型有源电力滤波器的控制块

由式(5)可以发现该产品负载电流的谐波成分和串联的电感器和电容器组的阻抗可通过把负载电流的谐波成分输送到比例积分微分(PID控制)控制器来获得。比例，积分和微分系数分别用电阻R ，电容C和电感L表示，如式(5)所示。因此，得到了第一个控制回路的输出。为了避免噪声影响，低通滤波器是用在微分控制器和一个高通滤波器的前面，在积分控制器的后面由于初始条件来抑制直流分量。因此，串联的电感器和电容器组位于电力转换器的输出端，电容器由于初始条件可以限制直流分量。因此，初始条件所造成的影响是由有源电力滤波器开关引起的，可以抑制。 为了提高补偿的能力，第二个控制回路是用来修改第一个控制回路补偿结果的误差。在第二个控制回路中，检测到的电源电流被送到带宽限制滤波器过滤掉基波组成成分。因此，获得未补偿的电源电流谐波成分。带宽滤波器的输出反馈给放大器获得第二个控制回路的输出。 第三个控制回路是用来生成一个虚拟谐波电阻器连接一系列串连的电感器和电容器组，以形成一个阻尼器。电力转换器的输出电流是被送到一个带通滤波器来获得基本的组成部分，然后将检测到的电力转换器输出电流和它的基本成分反馈给减法器获得谐波成分。谐波成分输入反馈给放大器获得第三个控制回路的输出。

第四控制回路是用来调节直流母线电压。第四控制回路由一个过滤直流母线电压纹波的低通滤波器和一个低通滤波器的输出减去一个设定值的减法器组成。在此之后，减法器的结果被送到一个PI控制器。带通滤波器的输出是有源电力滤波器电流的基本组成部分，第四个控制回路的输出是PI控制器输出和带通滤波器输出的乘积。

第五控制回路是用来调节无功功率补偿。第一，无功负荷的计算方法。电源电压被送到一个

- 25 -

嘉兴学院本科生毕业设计

相位变换电路把相位变换90°，然后再乘以负载电流;乘积被送到一个低通滤波器获得负载的无功功率。由于直流母线电压是不变的和谐波抑制的优先级高于有源电力滤波器补偿无功功率，补偿无功功率必须被限制。因此，一个限制电路是用来限制补偿无功功率。限制电路的极限值是变化的，取决于一些补偿的谐波电流。在没有负荷或轻负荷条件下，一些补偿的谐波电流很小，限制电路的值很大。这意味着电力转换器在这个条件可提供大量无功功率。尽管限制电路的值在重负载条件下几乎接近于零。这意味着最大无功功率是这个条件下提供给串联的电感器和电容器组的无功功率。电源电压和限制电路的输出被送到一个乘法器来获得第五个控制回路的输出。 最后，调制信号可通过第一次，第二次，第三次，第四次和第五次控制回路的输出相加得到。然后，调制信号被送到一个脉宽调制器，来驱动电力转换器的电力电子设备。

5 实验结果

为了证明该新型有源电力滤波器的能力，开发出一个三相20KVA原型。原型的主要参数如表1所示。有功功率是由一个三相三线系统运行在380V，60Hz条件下提供的。该新型有源电力滤波器和常规并联型有源电力滤波器的比较如表2所示。因为串联的电感器和电容器组的电感只有200H，铁芯可以用来降低功率损耗，重量和体积。因此，该新型有源电力滤波器的体积和重量显然比常规并联型有源电力滤波器要小。此外，由于直流电容器和电力电子器件的电压等级低，滤波器电感值小，整体系统尺寸小，硬件成本也大大减少。测试的负荷是一个六脉冲整流充电器。 表1 ：原型的主要参数

表2 ：比较结果

- 26 -

嘉兴学院本科生毕业设计

Fig. 7 新型有源电力滤波器在稳态下的测试结果

a 电源电压 b 电源电流 c 负荷电流

d 有源电力滤波器电流 图7显示了该新型有源电力滤波器在稳定状态下实验结果。曲线7C所示的负载电流谐波较多

及总谐波失真(THD)是26％。电源电流的波形如图7b所示，接近于正弦波，其总谐波失真在提供的有源电力滤波器补偿后只有3％。测试结果表明，该新型有源电力滤波器的谐波抑制能力是非常好的。

图8显示了该新型有源电力滤波器在非线性负载瞬态运行下的试验结果。由曲线8b可以发现该新型有源电力滤波器具有良好的瞬态响应。

在一个工业配电系统中，涡轮发电机是常常被用于后备电源。然而，气轮发电机组的电源可以被视为一个弱电源，因为气轮发电机组的容量不够大。该弱电源的系统阻抗非常大，电源电压在非线性负载条件下将明显退化。实验结果显示在大的系统阻抗条件下可得到图9和图10。一个0.6mH的电感器插入到应用电力馈线来模拟大的系统阻抗条件。图9可以发现，由于非线性负载，电压波形严重扭曲。电源电压和电流的总谐波失真分别为9％和39％ 。扭曲的电源电压可能扰乱同一电力馈线中电力设备本身和相邻负载的正常运作。图10说明了电压和电流波形接近于正弦在运行的有源电力滤波器中。电压和电流的总谐波失真分别为1.4％和4.9％。因此，这个验证说明该新型有源电力滤波器不仅能抑制输入电流谐波，而且在非线性负载下能避免电压波形畸变。

- 27 -

嘉兴学院本科生毕业设计

Fig. 8 新型有源电力滤波器在非线性负载瞬态运行下的试验结果 a 电源电压 b 电源电流 c 负荷电流

d 有源电力滤波器电流

Fig. 9 在有大系统阻抗条件下没有运行新型有源电力滤波器的测试结果 a 电源电压 b 电源电流

- 28 -

嘉兴学院本科生毕业设计

Fig. 10 在有大系统阻抗条件下运行新型有源电力滤波器的测试结果

a 电源电压 b 电源电流

Fig. 11 新型有源电力滤波器在无负载条件下的测试结果

a 电源电压 b 电源电流

图11显示了该新型有源电力滤波器在无负载条件下运行的实验结果。通过计算电源电压和串联的电感器和电容器组的阻抗可获得最大基波电流。电源电压是380V，有源电力滤波器的基波电流的均方根值在补偿后从15A变成7.2A。这验证了该新型有源电力滤波器可提供可调的无功功率。

6 结论

电力电子设施的输入特征是具有高的谐波电流和低的功率因数。本文提出了一种新型的有源电力滤波器和控制方法。该新型有源电力滤波器可以有效地抑制谐波电流并提供可调的无功功率。它还具有低电压等级的直流电容器和电力电子器件，较小的滤波电感器，尺寸小，重量轻，更好

- 29 -

嘉兴学院本科生毕业设计

的滤波性能和低电磁干扰等优势。一个三相20KVA的有源电力滤波器已经开发出来，以显示本文所提出的控制方法所具有的性能。实验结果表明，该新型有源电力滤波器在抑制谐波电流上具有良好的性能。此外，对于输入是二极管整流器或480v以下低压相控整流器的谐波负载来说，该新型有源电力滤波器的硬件花费是非常有竞争力的。

7 感谢

作者感谢埃布勒电子有限公司在资金上的支持和工程师在硬件实现上的帮助。

8 参考文献

[1] Henderson,R.D.,and Rose,P.J.:谐波：对电能质量和变压器的影响,IEEE

Trans.Ind.Appl.,1994,41,pp.528–532

[2] Wu,C.J.,Chiang,J.C.,Yen,S.S.,Liao,C.J.,Yang,J.S.,and Guo,T.Y.:对单一滤波器引起的谐波放大问题进行的研究和减轻,IEEE Trans.Power Deliv.,1998,13, pp.800–806

[3] Chiang,S.J.,Ai,W.J.,and Lin,F.J.:容量限制的三相四线制有源电力滤波器的并联运行,IEE Proc.,Electr.Power Appl.,2002,149,(5),pp.329–336

[4] Wu,J.C.,and Jou,H.L.:一种单相有源电力滤波器简单控制方法,IEEE Proc.Electr.Power Appl.,1996,143,pp.219–224

[5] Akagi,H.:有源电力线调节装置的趋势,IEEE Trans.Power Electron.,1994,9,(3),pp.263–268 [6] Abellan,A.,Benavent,J.M.,and Garcera,G.:一种新的并联有源电力滤波器在不平衡和非正弦条件下获得参考电流.的方法Proc.IEEE ISIE Conf.,1999,pp.831–836

[7] Singh,B.,Al-Haddad,K.,and Chandra,A.:一种新的控制方法：三相有源电力滤波器谐波及无功功率补偿,IEEE Trans.Power System,1998,13,(1), pp. 133–138

[8] Wu,J.C.,and Huang,S.J.:级联型有源电力滤波器的设计和操作：减少电力系统中谐波失真,IEEE Proc.Gener.Transm.Distrib.,1999,146,(2),pp193–199

[9] Kawabata,T.,and Komatsu,Y.:三相有源电力滤波器使用扩展PQ理论的特点.Proc.IEEE ISIE,Conf.,1997,pp.302–307

[10] Yang,J.,Wang,Y.,and Wang,Z.:一个DSP控制的用于补偿谐波和电力牵引负荷引起的无功功率的混合型电力滤波器.Proc.IEEE PESC Conf.,2003, pp.1615–1620

[11] Fujita,H.,and Akagi,H.:电力系统、串联的无源和有源滤波器中谐波补偿的切实可行的方法，,IEEE Trans.Ind.Appl.,1991,38,pp.1020–1025

[12] Dixon,J.,Espinoza,J.,Moran,L.,and Rivas,D.:混合型有源电力滤波器的简单控制方案.Proc.IEEE PESC Conf.,2000,pp.991–996

- 30 -

嘉兴学院本科生毕业设计

[13] Rahmani,S.,Al-Haddad,K.A.,and Fnaiech,F.:一种新的基于瞬时有功电流提供给并联混合电力滤波器的控制技术.Proc.IEEE PESC Conf.,2003, pp.808–813

[14] EI-Habrouk,M.,Darwish,M.K.,and Mehta,P.:有源电力滤波器：评论,IEEE Proc.,Electr.Power Appl.,2000,147,(5),pp.403–412

[15] Mohan,Undeland,and Robbins,:电力电子转换器的设计和应用(Media Enhanced,John Wiley & Sons,Inc.,2003,3rd edn.)

- 31 -

嘉兴学院本科生毕业设计

签字： - 32 -

年 月五、指导教师评语

日

本文来源：https://www.bwwdw.com/article/xed6.html