Inverter parallel current sharing control method based on virtual reactive voltage self-adjustment

Abstract

The invention discloses an inverter parallel current sharing control method based on virtual reactive voltage self-adjustment, and belongs to the technical field of coordination control of parallel inverters. Compared with traditional droop control, the reactive power equalization degree can be well improved, and the reactive power equalization effect and the response speed when the load power changes are enhanced. According to the invention, the relation between the output power and the output impedance of the inverter is deeply analyzed, the characteristics of loop chain control and traditional droop control are combined, and a reactive loop chain and self-adjusting virtual reactive voltage control are introduced, so that the virtual impedance can realize self-adaptive adjustment according to circular chain type tracking of reactive power output by a parallel module; mismatch between output impedances of the parallel modules is compensated, and output reactive power balance of the parallel modules is achieved. In addition, the voltage of a common coupling point does not need to be measured, the requirement for a communication network is low, the system reliability and safety are higher, and good reference value is provided for engineering application.

Description

Technical Field

[0001] The present invention belongs to the technical field of coordinated control of parallel inverters, and in particular relates to a method for controlling parallel current sharing of inverters based on virtual reactive voltage self-regulation. Background Art

[0002] As people pay more and more attention to the environment, renewable energy is gradually developed and utilized, and microgrids and distributed power generation have also emerged. As a key technology of microgrids and distributed power generation, inverter parallel technology is also gaining attention. Inverter parallel technology has many advantages, such as modular design, high redundancy and economy, and can enhance the function and application scope of inverters and improve the reliability of power supply systems.

[0003] One of the key issues of inverter parallel connection is the balanced distribution of load current and the suppression of circulating current between parallel systems. Only when the output voltage amplitude, frequency and phase of each parallel module are the same, the load current will be evenly distributed according to the capacity of the parallel modules. However, when the output impedance and circuit parameters of the parallel modules are inconsistent, there will be circulating current between the parallel systems, resulting in unbalanced output power of the parallel modules, causing overload or damage to the parallel modules. At present, a variety of inverter parallel current sharing control strategies have been proposed. Since power droop control does not require interconnection lines, the spatial dispersion and scalability of parallel modules are strong, and they have high redundancy and reliability, so power droop control is the mainstream control strategy for inverter parallel current sharing control. However, when the output impedance of the parallel modules is unequal, the output reactive power of the parallel modules is unbalanced under power droop control, and the current sharing effect of the parallel system is poor.

[0004] In order to solve the above problems, how to balance the output reactive power when the output impedance of the parallel modules is mismatched under power droop control has become a research hotspot. Some literatures have proposed a robust droop control, adding an integral link in the reactive-amplitude droop control loop, which can eliminate the influence of output impedance on reactive power in steady state and realize reactive power balancing. However, this method is highly dependent on the voltage of the common coupling point (PCC) to eliminate the influence of line impedance. Recently, some literatures have adopted the harmonic signal injection method, which mainly injects a harmonic with a very small amplitude of about 1% to 2% of the fundamental voltage into the reference voltage, and controls the reactive power according to the harmonic power to achieve reactive power balancing. However, this method will increase the harmonic content of the system output voltage due to the injection of harmonic voltage. There are also literatures that use the virtual impedance method, which can enhance system stability and realize reactive power balancing by subtracting the current feedback term from the reference voltage to compensate for the mismatch of the output impedance. However, since the line parameters of the parallel system are difficult to measure and the output impedance of the parallel modules cannot be accurately known, the traditional virtual impedance method cannot completely compensate for the mismatch of the output impedance, and the power sharing effect is still not ideal. Summary of the invention

[0005] In order to overcome the defects of the above-mentioned prior art, improve the reactive power balancing degree based on power droop control, and enhance the reactive power balancing effect and response speed when the load power changes, the present invention proposes an inverter parallel current sharing control method based on virtual reactive voltage self-adjustment, which can realize accurate reactive power balancing.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The present invention provides a method for controlling parallel current sharing of inverters based on virtual reactive voltage self-adjustment, comprising the following steps:

[0008] 1) Several inverters, all of which are connected in parallel, with the DC sides of all inverters connected to a DC source, and the AC sides of all inverters connected to an AC bus, respectively. Each inverter is equally controlled as a voltage source using the same control strategy.

[0009] 2) Each inverter samples voltage and current information for power calculation, and then uses the calculated active power and reactive power for droop control, and obtains the voltage reference value for droop control. The voltage phase information is obtained through a SOGI-based phase-locked loop. The phase-locked loop cooperates with the voltage and current dual closed-loop control to ensure that the frequency reference value is near the base frequency (50Hz) and keep each inverter synchronized, achieving stable parallel operation.

[0010] 3) According to the reactive power calculated in step 2), reactive chain and self-adjusting virtual reactive voltage control are performed to obtain the voltage adjustment value of the self-adjusting virtual reactive voltage, and then the voltage reference value obtained by the droop control in step 2) is adjusted to obtain the final voltage reference value of the inverter.

[0011] 4) According to the final voltage reference value of the inverter obtained in step 3), the voltage and current dual closed-loop regulator is input to generate a PWM modulation signal.

[0012] Preferably, the specific method for implementing droop control in step 2) is as follows:

[0013]

[0014] Among them, ω and E are the reference amplitude and frequency of the inverter voltage obtained by droop control, ω o and E o is the no-load output voltage amplitude and frequency of the inverter, m and n are the active-frequency droop adjustment coefficient and reactive-amplitude droop adjustment coefficient respectively, P and Q are the active power and reactive power of power calculation, and i is the serial number of the inverter.

[0015] In step 3), the reactive power chain and self-adjusting virtual reactive voltage control are specifically performed as follows:

[0016] The reactive power value of the previous inverter is obtained through the reactive chain, and input into the self-adjusting virtual reactive voltage control as the reactive power reference value of this inverter. The integral control strategy is used to generate the self-adjusting virtual impedance to eliminate the steady-state reactive power error between the inverters. The specific expression is as follows:

[0017]

[0018]

[0019] Among them, M v is the self-adjusting virtual impedance, k v is the self-adjustment coefficient, N is the number of parallel inverters, s is the Laplace operator, u vQ is the voltage regulation value of the self-adjusting virtual reactive voltage, i o is the output current.

[0020] Preferably, for the self-adjustment coefficient k v The selection includes the following steps:

[0021] 1) The calculation formula for the inverter output active power and reactive power is as follows:

[0022]

[0023] Among them, R t is the total output resistance, is the inherent output resistance R o , Feeder resistance R f and the sum of the self-adjusting virtual resistance; X t is the total output reactance, is the inherent output reactance X o , Feeder reactance X f In order to simplify the analysis, the self-adjusting virtual resistance and self-adjusting virtual reactance are approximately equal to the self-adjusting virtual impedance M. v .U o is the PCC voltage, δ is the inverter no-load output voltage E o and PCC voltage U o The angle between.

[0024] 2) Perform small signal modeling analysis, and the linear expressions of the inverter output active power and reactive power are as follows:

[0025] ΔP i =k peΔ E i +k pδ Δδ i +k pm ΔM v

[0026] ΔQi =k qe ΔE i +k qδ Δδ i +k qm ΔM v

[0027] where k pe , k pδ , k pm and k qe , K qδ , k qm are the corresponding partial differentials of active power and reactive power at the steady-state point, and their expressions are as follows:

[0028]

[0029] 3) The linear expressions of droop control and self-adjusting virtual reactive voltage control are as follows:

[0030] Δω=-mΔP i

[0031] ΔE=-nΔQ i

[0032]

[0033] 4) Step 3) adds a low-pass filter and the expression after combining with step 2) is as follows:

[0034]

[0035]

[0036]

[0037] Among them, ω c are the low-pass filter parameters.

[0038] 5) Combining step 4) and Δω=s.Δδ, the characteristic equation of the parallel system is expressed as follows:

[0039] s 4 +b1s 3 +b2s 2 +b3s+b4=0

[0040] Among them, the expressions of coefficients b1, b2, b3, and b4 are as follows:

[0041]

[0042] Preferably, the self-adjustment coefficient k can be plotted according to the characteristic equation vChanging root locus diagram. During design, in order to coordinate the stability and dynamic response of the system, the self-adjustment coefficient should not be too large.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] The present invention proposes a method for controlling the parallel current sharing of inverters based on virtual reactive voltage self-adjustment. Compared with the traditional droop control, the method can well improve the reactive power sharing degree, and enhance the reactive power sharing effect and response speed when the load power changes. The present invention introduces a reactive ring chain self-adjusting virtual impedance by deeply analyzing the relationship between the inverter output power and the output impedance, combining the characteristics of the ring chain control and the traditional droop control, and the virtual impedance can realize adaptive adjustment according to the cyclic chain tracking of the reactive power output of the parallel module, thereby compensating for the mismatch between the output impedances of the parallel modules, and realizing the reactive power balance of the parallel modules. In addition, the present invention can still work normally during the load jump process, greatly saving the adaptive adjustment time and accelerating the dynamic response speed of the system. Moreover, the present invention can also realize faster parallel and offline steady-state operation when the inverter is offline, and has good hot-swap function. Finally, the present invention does not need to measure the PCC point voltage, has low requirements on the communication network, and has higher system reliability and safety, which provides a good reference value for engineering applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is a control block diagram of the present invention;

[0046] Figure 2 This is a structural diagram of the self-adjusting virtual reactive voltage in the present invention;

[0047] Figure 3 is the self-adjusting coefficient k in the present invention v Variation root locus diagram;

[0048] Figure 4 The output power waveforms of three inverters under three control methods are shown;

[0049] Figure 5 The self-adjusting virtual impedance waveforms of the three inverters under different control methods;

[0050] Figure 6 The voltage waveforms of the common coupling point under three control methods are shown;

[0051] Figure 7 The following are simulation waveforms of reactive power, output current and common coupling point voltage of three inverters when the load power switches from full load to light load and then to heavy load under the control method proposed in the present invention. DETAILED DESCRIPTION

[0052] In order to further understand the content of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments, which are intended to explain the present invention rather than limit it.

[0053] See also Figure 1 and Figure 2 The present invention provides a method for controlling parallel current sharing of inverters based on virtual reactive voltage self-adjustment, comprising the following steps:

[0054] 1) Several inverters, all of which are connected in parallel, with the DC sides of all inverters connected to a DC source, and the AC sides of all inverters connected to an AC bus, respectively. Each inverter is equally controlled as a voltage source using the same control strategy.

[0055] 2) Each inverter samples voltage and current information for power calculation, and then uses the calculated active power and reactive power for droop control, and obtains the voltage reference value for droop control. The voltage phase information is obtained through a SOGI-based phase-locked loop. The phase-locked loop cooperates with the voltage and current dual closed-loop control to ensure that the frequency reference value is near the base frequency (50Hz) and keep each inverter synchronized, achieving stable parallel operation.

[0056] 3) According to the reactive power calculated in step 2), reactive chain and self-adjusting virtual reactive voltage control are performed to obtain the voltage adjustment value of the self-adjusting virtual reactive voltage, and then the voltage reference value obtained by the droop control in step 2) is adjusted to obtain the final voltage reference value of the inverter.

[0057] 4) According to the final voltage reference value of the inverter obtained in step 3), the voltage and current dual closed-loop regulator is input to generate a PMW modulation signal.

[0058] 5) The specific method for implementing droop control in step 2) is as follows:

[0059]

[0060] Among them, ω i and E i is the inverter voltage reference amplitude and frequency obtained by droop control, ω o and E o is the inverter no-load output voltage amplitude and frequency, m and n are the active-frequency droop adjustment coefficient and reactive-amplitude droop adjustment coefficient, respectively. i and Q i is the active power and reactive power of power calculation, and i is the serial number of the inverter.

[0061] 6) The specific method for controlling the reactive power chain and self-adjusting virtual reactive voltage in step 3) is as follows:

[0062] The reactive power value of the previous inverter is obtained through the reactive chain, and input into the self-adjusting virtual reactive voltage control as the reactive power reference value of this inverter. The integral control strategy is used to generate the self-adjusting virtual impedance to eliminate the steady-state reactive power error between the inverters. The specific expression is as follows:

[0063]

[0064]

[0065] Among them, M v is the self-adjusting virtual impedance, k v is the self-adjustment coefficient, N is the number of parallel inverters, s is the Laplace operator, u vQ is the voltage regulation value of the self-adjusting virtual reactive voltage, i o is the output current.

[0066] 7) For the self-adjustment coefficient k in step 6) v The calculation formulas for the inverter output active power and reactive power are as follows:

[0067]

[0068] Among them, R t is the total output resistance, is the inherent output resistance R o , Feeder resistance R f and the sum of the self-adjusting virtual resistance; X t is the total output reactance, is the inherent output reactance X o , Feeder reactance X f In order to simplify the analysis, the self-adjusting virtual resistance and self-adjusting virtual reactance are approximately equal to the self-adjusting virtual impedance M. v .U o is the PCC voltage, δ is the inverter no-load output voltage E o and PCC voltage U o The angle between.

[0069] 8) After small signal modeling analysis, the linear expressions for the inverter output active power and reactive power in step 7) are as follows:

[0070] ΔP i =k pe ΔE i +k pδ Δδ i +k pm ΔM v

[0071] ΔQ i =k qe ΔE i+k qδ Δδ i +k qm ΔM v

[0072] where k pe , k pδ , k pm and k pe , K qδ , k qm are the corresponding partial differentials of active power and reactive power at the steady-state point, and their expressions are as follows:

[0073]

[0074] 9) The linearized expressions of step 2) and step 3) are as follows:

[0075] Δω=-mΔP i

[0076] ΔE=-nΔQ i

[0077]

[0078] 10) Step 9) adds a low-pass filter and the expression after combining with step 8) is as follows:

[0079]

[0080] Among them, ω c are the low-pass filter parameters.

[0081] 11) Combining step 10) and Δω=s·Δδ, the characteristic equation of the parallel system is expressed as follows:

[0082] s 4 +b1s 3 +b2s 2 +b3s+b4=0

[0083] Among them, the expressions of coefficients b1, b2, b3, and b4 are as follows:

[0084]

[0085] 12) According to the characteristic equation of step 11), Figure 3 The self-adjustment coefficient k is given v From the root locus diagram, we can draw the following conclusions:

[0086] The parallel system after the introduction of self-adjusting virtual reactive voltage control has four characteristic roots, one pair of conjugate complex roots and one pair of real roots, all located in the left half plane, and the system is in a stable state. The conjugate complex roots change in an arc shape as the self-adjusting coefficient increases, that is, they first approach the real axis, then move away from the real axis, and finally approach the real axis. According to control theory, the increase in the imaginary part of the characteristic root will lead to a decrease in system damping, an increase in overshoot, and a longer oscillation adjustment time. Therefore, the self-adjusting coefficient should be selected when the conjugate complex root changes for the first time and approaches the real axis. In this stage, the system has a small overshoot, fast adjustment, and good dynamic performance. The other pair of real roots gradually move away from the imaginary axis as the self-adjusting coefficient increases, and has little effect on the stability of the parallel system. Therefore, the self-adjusting coefficient should not be too large, otherwise the stability will deteriorate.

[0087] Specifically, the simulation model of the present invention is composed of three inverters sharing the load and setting different feeder impedances. The traditional droop control method, the virtual impedance control method, and the control method proposed by the present invention are simulated respectively. The simulation parameters are shown in Table 1.

[0088] Table 1 Simulation parameters

[0089] parameter Numeric DC input voltage <![CDATA[V dc =400V]]> Rated capacity <![CDATA[S 1,2,3 =2.2kVA]]> Rated output voltage <![CDATA[U o =220V]]> Rated output frequency <![CDATA[f o =50Hz]]> Switching frequency <![CDATA[f s =20kHz]]> LC filter (parasitic resistance r) L=2mH,C=3.17μF(r=5mΩ) Voltage controller PI parameters <![CDATA[k pu =0.0085,k iu =524.7]]> Current controller PI parameters <![CDATA[k pi =0.203,k ii =6132]]> Droop coefficient m=0.0003,n=0.0008 Self-adjusting coefficient <![CDATA[k v =0.01]]> Feeder impedance 1 <![CDATA[Z f1 =0.05+j0.251Ω]]> Feeder impedance 2 <![CDATA[Z f2 =0.1+j0.314Ω]]> Feeder impedance 3 <![CDATA[Z f3 =0.15+j0.377Ω]]> Load power <![CDATA[S L =(1.76+j1.32)kVA]]>

[0090] Figure 4 The output power waveforms of three inverters under three control methods are given. The control methods at different stages are from left to right: traditional droop control method, virtual impedance control method, and the control method proposed by the present invention. Figure 4 In the conventional droop control method (first stage), the active power can achieve a good balancing effect, while the reactive power is in an unequal distribution state, and its size is inversely proportional to the total output impedance of the inverter module. Under the virtual impedance control method (second stage), the active power is still in an unequal distribution state, but the output active power is reduced, while the reactive power balancing degree increases, but it is still in an unequal distribution state. Under the control method (third stage) proposed in the present invention, both active power and reactive power can achieve power balancing and evenly distribute the load power.

[0091] Figure 5 The self-adjusting virtual impedance waveforms of the three inverters under different control methods are given. In steady state, the virtual impedance of the module with smaller total output impedance is positive, and the virtual impedance of the module with larger total output impedance is negative. Under its regulation, the total output impedance of the parallel modules tends to be consistent.

[0092] Figure 6The common coupling point voltage waveforms under the three control methods are given. It can be seen that the common coupling point voltage is reduced under the virtual impedance control method, while the common coupling point voltage has no obvious change under the traditional droop control and the control method proposed in the present invention. Therefore, the control method proposed in the present invention will not sacrifice the output voltage accuracy, realize power sharing, and has a certain voltage regulation effect.

[0093] Figure 7 The reactive power, output current, and common coupling point voltage simulation waveforms of three inverters are given when the load power switches from full load to light load and then to heavy load under the control method proposed in the present invention. It can be seen that under the control method proposed in the present invention, after the load is switched, the reactive power and output current are quickly balanced, with a faster dynamic response speed and better reactive power distribution. In addition, under the control method proposed in the present invention, the common coupling point voltage fluctuates less during the load change process and has a higher voltage stabilization accuracy.

[0094] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, it should be understood by those skilled in the art that the specific implementation modes of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be included in the scope of protection of the claims of the present invention.

Claims

1. A method for controlling parallel current sharing of inverters based on virtual reactive voltage self-adjustment, characterized in that: The following steps are involved: 1) Several inverters, all of which are connected in parallel, with the DC sides of all inverters connected to a DC source, and the AC sides of all inverters connected to an AC bus, respectively. Each inverter is equally controlled as a voltage source using the same control strategy; 2) Each inverter samples voltage and current information for power calculation, and then uses the calculated active power and reactive power for droop control, and obtains the voltage reference value for droop control. The voltage phase information is obtained through a SOGI-based phase-locked loop. The phase-locked loop cooperates with the voltage and current dual closed-loop control to ensure that the frequency reference value is near the base frequency (50Hz) and keep each inverter synchronized, thus achieving stable parallel operation; 3) According to the reactive power calculated in step 2), reactive chain and self-adjusting virtual reactive voltage control are performed to obtain a voltage adjustment value of the self-adjusting virtual reactive voltage, and then the voltage reference value obtained by the droop control in step 2) is adjusted to obtain the final voltage reference value of the inverter; 4) According to the final voltage reference value of the inverter obtained in step 3), the voltage and current dual closed-loop regulator is input to generate a PWM modulation signal.

2. The method for controlling parallel current sharing of inverters based on virtual reactive voltage self-adjustment according to claim 1, characterized in that: The specific method for implementing droop control in step 2) is as follows: Among them, ω and E are the reference amplitude and frequency of the inverter voltage obtained by droop control, ω o and E o is the no-load output voltage amplitude and frequency of the inverter, m and n are the active-frequency droop adjustment coefficient and reactive-amplitude droop adjustment coefficient respectively, P and Q are the active power and reactive power of power calculation, and i is the serial number of the inverter; In step 3), the reactive power chain and self-adjusting virtual reactive voltage control are specifically performed as follows: Among them, M v is the self-adjusting virtual impedance, k v is the self-adjustment coefficient, N is the number of parallel inverters, s is the Laplace operator, u vQ is the voltage regulation value of the self-adjusting virtual reactive voltage, i o is the output current.

3. The method for controlling parallel current sharing of inverters based on virtual reactive voltage self-adjustment according to claim 1, characterized in that: For the self-adjustment coefficient k v The selection includes the following steps: 1) The calculation formula for the inverter output active power and reactive power is as follows: Among them, R t is the total output resistance, is the inherent output resistance R o , Feeder resistance R f The sum of the self-adjusting virtual resistance; X t is the total output reactance, is the inherent output reactance X o , Feeder reactance X f and the sum of the self-adjusting virtual reactance. To simplify the analysis, the self-adjusting virtual resistance and the self-adjusting virtual reactance are approximately equal to the self-adjusting virtual impedance M. v , U o is the voltage at the point of common coupling (PCC), δ is the inverter no-load output voltage E o and PCC voltage U o The angle between 2) Perform small signal modeling analysis, and the linear expressions of the inverter output active power and reactive power are as follows: ΔP i =k pe DE i +k pδ Dd i +k pm DM v ΔQ i =k qe DE i +k qδ Dd i +k qm DM v where k pe , k pδ , k pm and k qe , k qδ , k qm are the corresponding partial differentials of active power and reactive power at the steady-state point, and their expressions are as follows: 3) The linear expressions of droop control and self-adjusting virtual reactive voltage control are as follows: Δω=-mΔP i ΔE=-nΔQ i 4) Step 3) adds a low-pass filter and the expression after combining with step 2) is as follows: Among them, ω c is the low-pass filter parameter; 5) Combining step 4) and Δω=s·Δδ, the characteristic equation of the parallel system is expressed as follows: s 4 +b1s 3 +b2s 2 +b3s+b4=0 Among them, the expressions of each phase coefficient b1, b2, b3, and b4 are as follows:

4. The self-adjustment coefficient selection method according to claim 3, characterized in that: According to the characteristic equation, the self-adjustment coefficient k can be plotted v Changing root locus diagram. During design, in order to coordinate the stability and dynamic response of the system, the self-adjustment coefficient should not be too large.

Images