Home>>News

A new algorithm for harmonic and reactive current

2022-09-21

Detail

A new algorithm for harmonic and reactive current detection

0 introduction

with the development of power electronic technology, power electronic devices are more and more widely used, but the harmonic generated by them also brings harm to electricity pollution and electromagnetic interference. On the other hand, modern electrical equipment is more sensitive to power quality and puts forward higher requirements for power supply quality. Active power filter can eliminate harmonics and improve the stability of power system operation. Its research and application have attracted more and more attention

there are two basic principles for active filter to eliminate harmonics: one is to inject a current equal to the reactive power and harmonic current of the load and in the opposite direction to compensate reactive power and suppress harmonics, which is called parallel active filter; The other is to inject the fundamental compensation current into the secondary side of the series transformer, so that the series transformer has a low impedance to the fundamental current and a high impedance to the harmonic current [1], so as to suppress harmonics. This method is called series active filter. In addition, there are series parallel type, hybrid type, etc. However, no matter which one is adopted, the value of harmonic and reactive current must be detected first. At present, the more mature current detection methods mainly include p q detection method [3] and IP IQ detection method [4] based on instantaneous reactive power theory [2]. However, these two methods need two coordinate transformations, and the amount of calculation is large. Among them, IP IQ detection method needs to use phase-locked loop, which has the problems of complex implementation and low detection accuracy

this paper studies a new algorithm for harmonic and reactive current detection, and gives the simulation results and experimental results

1 principle of harmonic and reactive current detection method

Figure 1 is the system block diagram of the parallel active filter. Its basic principle is: calculate the harmonic and reactive current of the load through the detection link, then control the output of the inverter circuit, inject the compensation current with the same size and opposite direction as the reactive and harmonic current of the load into the electricity, so that the electric current only contains the fundamental active component. In this way, the device can not only filter the harmonic, but also provide the reactive current required by the power system, which can greatly improve the utilization rate of electric energy and improve economic benefits

Figure 1 system block diagram of shunt active filter

this paper proposes a new harmonic and reactive current detection algorithm. Figure 2 is the detection schematic diagram of load harmonic and reactive current, and the dotted line box in the figure is the DC side voltage control part. As shown in Figure 2, first detect the actual load current and electric voltage, and calculate these six quantities to obtain the required three-phase load harmonic and reactive current

Figure 2 Schematic diagram of harmonic and reactive current detection algorithm

for simplicity, assuming that the electric voltage is three-phase symmetrical and distortion free, then

ua=umsin t

ub=umsin (T-2/3) (1)

uc=umsin (T + 2/3)

load current IA, IB, IC can be expressed as the sum of fundamental wave and harmonic, that is,

ic=ic1 + ick

considering the asymmetric load, the current is divided into positive sequence, negative sequence and zero sequence, so the fundamental current is

IA1 = I1 + sin (t －) + I1 sin (T + 1 －) + I10

IB1 = I1 + sin (t － 2/3) +

I1 sin (it is understood that t + 1 － + 2/3) + I10

IC1 = I1 + sin (t － + 2/3) +

I1 － sin (T + 1 － 2/3) + I10 (3)

where: i1+, i1-, I10 is the amplitude of positive sequence, negative sequence and zero sequence components of fundamental wave

is the power factor angle

1 - is the initial phase of the negative sequence of the fundamental wave

harmonic current is also divided into positive sequence, negative sequence and zero sequence. The k-th harmonic current can be expressed as

iak=ik + sin (k t + K +) + IK sin (k t + k -) + ik0

ibk=ik + sin (K T + K + 2/3) + IK sin (low production concentration and small range K T + K + 2/3) + ik0

ick=ik + sin (k t + K + 2/2/3) + IK sin (k t + K － 2/3) + ik0 (4)

where: IK +, IK － and ik0 are k-th harmonic positive sequence, negative sequence Amplitude of zero sequence component

k + and K - are the initial phases of harmonic positive sequence and negative sequence

the instantaneous value p of three-phase active power can be obtained from equation (5)

p=uaia ＋ ubib ＋ ucic=ua (IA1 ＋ Iak) + UB (IB1 ＋ IBK) + UC (IC1 ＋ ick) = (uaia1 ＋ ubib1 ＋ ucic1) + (uaiak ＋ ubibk ＋ ucick) = {3umi1 ＋ cos/2 － 3umi1 － cos (2 T ＋ 1 －)/2} + {3umik ＋ cos [(k － 1) t ＋]/2} - {3umik cos [(K + 1) t + k -]/2} (5)

equation (5) contains DC and a series of harmonic components. Harmonic in basketball courts, volleyball courts and other open-air sports venues and cultural welfare facilities, the frequency of the ground wave can be as low as 100Hz. After low-pass filtering, the harmonic component in the power can be filtered out, leaving only the steady-state value p (3umi1 + cos/2), where I1 + COS is the amplitude of the active component of the fundamental positive sequence current. For phase a, the active component IA1 of the fundamental positive sequence current has =i1 + cos sin t. From equation (6), we can get that

ia1 has =i1 + cos sin t== (6)

similarly, we can get that the active component of other two-phase fundamental positive sequence current IB1 has =i1 + cos sin (T-2/3), and IC1 has =i1 + cos sin (T + 2/3)

subtract the above active components of the fundamental positive sequence current IA1, IB1 and IC1 from the actual load current IA, IB and IC to obtain the load harmonic and reactive current, which can be used as the compensation current command output by the three-phase inverter, that is,

IAC * =ia-ia1 has

IBC * =ib-ib1 has (7)

ICC * =ic-ic1 has

in addition, the voltage UD on the DC side of the inverter should be maintained constant during the operation of the active filter. The dotted box in Figure 2 shows the DC side voltage control part. As shown in Figure 2, input the difference between the given value UD * and the actual detected value UD into the PI regulator, multiply the output by the actual DC measured voltage UD, and the result is taken as the increment P of active power. Superimpose P on the output of the low-pass filter in Figure 2, so that there is a certain fundamental active current in IC *, so that the capacitor on the DC side of the inverter can obtain energy from the AC side, and compensate the operating power consumption of the active filter, so as to stabilize UD at the given value UD *

2 simulation and test results

this detection algorithm is simulated by using the Simulink module in MATLAB. The simulation results are shown in Figure 3. It can be seen from the simulation waveform that the fundamental active current calculated by the detection algorithm is completely in phase with the electric voltage, and it is a standard sine, which shows that the detected harmonic and reactive current are completely accurate

Figure 3 simulation waveform of load harmonic and reactive current detection

in the figure: 1 electric voltage 2 load current 3 detected fundamental active current 4 detected harmonic and reactive current waveform

the capacity of the experimental prototype is designed to be 6kW, the voltage is three-phase 380V, and the load is motor and uncontrolled rectifier bridge. The control part takes the DSP chip tms320s2407 of TI company as the core, and the harmonic and reactive current detection and PWM pulse signal generation are realized by the corresponding software

the main functional modules involved in the software are: event manager, a/d conversion module, interrupt service program. Use T1 timer to start a/d conversion at a fixed time, and sample the electric voltage, load current, electric current and DC side voltage in turn. Set the sampling frequency as 10kHz. After the completion of a/d conversion, ADC interrupt is generated, and the algorithm is implemented in the interrupt service subroutine to calculate the harmonic and reactive current, that is, the compensation current command. Among them, the low-pass filter adopts a second-order Butterworth filter with a cut-off frequency of 20Hz. The current control method adopts the triangular carrier modulation method. The compensation current command is compared with the actual compensation current, and the difference is sent to the digital PI regulator. The output of the PI regulator is modulated with the high-frequency triangular carrier. The PWM module generates six PWM control signals, in which the triangular carrier is realized by a timer with a frequency of 10Hz

send six PWM control signals to the drive circuit, and finally generate the corresponding compensation current through IGBT to inject electricity. The simulation results and experimental results of the whole system are shown in Figure 4 and figure 5

Figure 4 system simulation waveform

(a) electric voltage, load current

(b) electric voltage, electric current

Figure 5 system experimental waveform

the experiment and simulation have similar results. According to the system experimental waveform in Figure 5, the actual load current contains a large number of harmonics and reactive components, and the electric voltage is partially distorted due to the influence of the load. After compensation, the electric current is basically sinusoidal and in phase with the voltage

3 conclusion

this new detection algorithm of harmonic and reactive current in power system proposed in this paper can detect all harmful currents including fundamental reactive current, zero sequence current, negative sequence current and harmonic current. Simulation and experimental results verify the correctness and feasibility of this detection algorithm. This algorithm does not need phase-locked loop and matrix transformation, and has the characteristics of accurate calculation and simple implementation