WO2020252813A1

WO2020252813A1 - Double-layer adaptive inertia control method and device for inverter interfaced distributed generator

Info

Publication number: WO2020252813A1
Application number: PCT/CN2019/094149
Authority: WO
Inventors: 黄文焘; 李美依; 邰能灵; 余墨多
Original assignee: 上海交通大学
Priority date: 2019-06-20
Filing date: 2019-07-01
Publication date: 2020-12-24
Also published as: CN110266062A; CN110266062B

Abstract

A double-layer adaptive inertia control method and device for an inverter interfaced distributed generator. The method comprises: obtaining the relationship between a virtual inertia constant of VSG and an output angular frequency of IIDG on the basis of a control topology structure of the VSG-based IIDG, and then establishing a small signal model; using a small signal model to simulate the actual operating scene to select, using a double-layer adaptive control strategy, the control priority between power regulation and frequency regulation, so that the IIDG can adapt to different operating scenes. The method and device can give consideration to both output stability and dynamic response speed, effectively improve the system operating performance, and strengthen the control effect.

Description

Double-layer adaptive inertia control method and device for inverter type distributed power supply

Technical field

The invention relates to a technology in the field of smart grids, in particular to a double-layer adaptive inertia control method and device for an inverter-type distributed power supply.

Background technique

With the continuous development of renewable energy power generation technology, Distributed Generator (DG) has become the main way for the distribution network to utilize new energy. How to make full use of high-permeability DG and improve its stability is a key issue that needs to be resolved. Most of the DGs in the distribution network are inverter-type power sources, such as photovoltaics, energy storage, etc., and their power generation units are mostly direct current. They are connected to the grid with inverter-type interfaces. The power electronic devices have fast response speed, small output impedance, and low moment of inertia. Inverter-interfaced DG (IIDG) has insufficient stability. Virtual synchronous generator (VSG) technology simulates the inertia characteristics of synchronous generators and adds virtual inertia control to the IIDG control strategy, which can greatly improve the transient output characteristics of IIDG, making IIDG a friendly grid-connected power supply. Thereby improving the IIDG's ability to participate in the operation and regulation of the distribution network.

At present, the VSG strategy mainly focuses on the frequency stabilization mechanism and the power angle curve trajectory, but there is little research on the influence mechanism of the output active power under different virtual inertia. The existing literature shows that the use of large inertia can effectively reduce the output frequency fluctuation and enhance the dynamic stability of the system frequency; however, the use of small inertia can obtain fast and stable output power. The virtual inertia has different effects on the output frequency and power of the VSG-IIDG. Power adjustment requires small inertia. However, if the VSG adopts low inertia, the frequency fluctuates greatly, which is not conducive to the stable operation of the system. Therefore, if the two are not considered at the same time, it may result in poor dynamic response characteristics of the VSG-IIDG output power under different operating scenarios, and may even fail to meet the power supply requirements.

Summary of the invention

In view of the above-mentioned shortcomings in the prior art, the present invention proposes a double-layer adaptive inertia control method and device for an inverter-type distributed power supply. The virtual inertia adopted by the VSG controller is an adaptively variable control parameter, which can be adjusted according to the operating state of the system. And real-time dynamic frequency deviation adaptively adjusts the control priority to meet the system's requirements for inverter-type distributed power output under different operating conditions.

The present invention is realized through the following technical solutions:

The invention relates to a double-layer adaptive inertia control method for an inverter-type distributed power supply. A small signal model is established according to the control topology of the IIDG based on the VSG, and the relationship between the virtual inertia constant of the VSG and the IIDG output angular frequency and active power is analyzed. Furthermore, the small-signal model is used to simulate the actual operating scenario, and the control priority is selected with a double-layer adaptive control strategy between power adjustment and frequency adjustment, so that IIDG can adapt to different operating scenarios.

The power adjustment refers to: when the IIDG's own active power reference value has a sudden change, the smaller virtual inertia constant H is used to adjust the active power;

The frequency adjustment refers to: when the common bus frequency has a small-range sudden change, the virtual inertia constant H is adaptively adjusted according to the IIDG frequency output change, specifically: when the deviation is small, the control gives priority to the system adjustment speed, and the output is overshoot When the problem is later, the influence of external fluctuations is suppressed through rapid response; when the deviation is large, the control focuses on smoothing the overshoot, and the system response speed is appropriately slowed down.

The control priority refers to the use of sensitivity coefficients to adaptively select the control priority to make IIDG adapt to different operating scenarios.

Among them: k _g is a preset coefficient, k _d reflects the frequency offset weight, the larger k _d is, the actual control curve is close to Γ ₁ ; when k _a decreases, the actual control curve is close to Γ ₄ , then the whole during the adjustment the virtual volume using the minimum inertia, Γ ₁ Γ ₄ corresponding to decreasing sensitivity coefficient, Γ ₄ the corresponding sensitivity coefficients are zero.

Preferably, the upper and lower limits of the selection range of the virtual inertia constant are obtained through the second-order system damping coefficient, droop coefficient, damping coefficient and control margin of the VSG.

The present invention relates to a device for realizing the above method, including: a priority adaptive adjuster and an offset adaptive adjuster, wherein: the priority adaptive adjuster adaptively selects control between power adjustment and frequency adjustment according to actual operating scenarios The priority is to improve the dynamic characteristics of IIDG power and frequency output as a whole, and the offset adaptive regulator adjusts the virtual inertia adaptively according to the frequency output of IIDG to achieve a control strategy that takes into account both small overshoot and fast response speed.

Technical effect

Compared with the prior art, the present invention realizes the adaptive control of the grid-connected inverter type distributed power source. This strategy takes into account the angular frequency and the stable control ability of power output, and has the characteristics of small overshoot and rapid response, making IIDG effective The output can be fast and stable when the system is disturbed, and can be used to deal with different types of disturbances in the system. Compared with the fixed inertia VSG control and Bang-bang control, the proposed control strategy takes into account the angular frequency and power output characteristics, has the characteristics of small overshoot and rapid response, and the control effect is better.

Description of the drawings

Figure 1 is a schematic diagram of the control topology of IIDG based on VSG;

Figure 2 is a schematic diagram of active power-frequency control;

Figure 3 is a schematic diagram of the flow of the VSG control algorithm;

Figure 4 is a schematic diagram of the trajectory of the characteristic root of the system when H changes from small to large;

Figure 5 shows the response curve of IIDG output active power offset ΔP;

Figure 6 is the response curve of IIDG output frequency offset Δω;

Figure 7 is the relationship curve between adaptive virtual inertia H and ω;

Figure 8 is a schematic diagram of dual-layer adaptive control;

Figure 9 is a schematic diagram of the topology of the simulation system;

Figure 10 is a schematic diagram of simulation results of grid-connected operation;

In the figure: a is the output of IIDG, b is the virtual inertia;

Figure 11 is a schematic diagram of simulation results of island operation;

In the figure: a is the output of IIDG, b is the virtual inertia.

Detailed ways

As shown in Figure 1, it is a control topology of IIDG based on VSG, in which the PWM signal controls the on and off of the switch in the inverter bridge under the drive of the drive circuit, and the bridge arm output voltage simulates the internal potential of the synchronous generator. L _f and C _ac are the filter inductance and capacitance respectively. After LC filtering, the inverter output voltage simulates the terminal voltage of the synchronous generator. Through the disconnection of the common coupling point, IIDG can switch between grid-connected and off-grid operation modes.

The frequency control of IIDG refers to:

Among them: P is the active power output by the inverter port under VSG control, k is the damping coefficient, ω is the angular frequency of the IIDG output, ω _grid is the angular frequency at the common coupling point, D is the active droop coefficient, and H is the virtual inertia of the VSG The constant, 2H, represents the time required for the VSG to accelerate from the stop state where the angular velocity is zero to the rated angular velocity under the rated power command.

As shown in Figure 2, it is a schematic diagram of active power-frequency control. When IIDG works in grid-connected mode, frequency control mainly relies on the damping term k(ω-ω _grid ) to track and synchronize with the frequency of the external power grid; when running off-grid, Frequency control uses active-frequency droop control to simulate the primary frequency modulation function of the power system to provide frequency support for the IIDG system.

For synchronous generators, the moment of inertia signifies the rotation characteristics, which are related to physical quantities such as the size and mass of the generator. In the VSG control system, since the inertia constant is a virtual control quantity, its value can be flexibly selected, and the constant can be used according to actual needs. The value or function value has a wider value range than the synchronous generator, and it has a better control effect.

According to the frequency control of IIDG, the small signal model after the linearization of the VSG near the given value can be obtained, that is, a system composed of two input quantities and two output quantities, as shown in Figure 3. When the common bus frequency is constant as the power frequency, the transfer function between the input and output of the VSG active power is established as:

When the VSG active power reference value does not change, a transfer function with the common bus angular frequency fluctuation as input and VSG angular frequency as output is established:

among them:

δ _s and E _s are the output voltage parameters of IIDG when the power is _Pref and Q _ref .

It can be seen from the transfer function expression that the VSG controller is a second-order system, and the characteristic root of the system is:

Figure 4 plots the trajectory of the characteristic roots of the system when the system is under-damped and the damping coefficient k is used. It can be seen from the figure that when the virtual inertia constant H continues to increase, the absolute value of the characteristic root of the system decreases, the characteristic root is closer to the virtual axis, and the system stability margin gradually decreases.

The two transfer functions of the VSG control structure can be used to analyze the response of the system when the IIDG's own active power is stepped and analyze the response of the system when the IIDG is subjected to frequency disturbances from the external system. The following is an analysis of the IIDG output characteristics for power and frequency changes.

1) When IIDG's own active power reference value has a sudden change, ΔP _ref =α*u(t), where: α represents the sudden change amplitude, u(t) is the unit step function, and the IIDG output power deviation is:

among them:

The first peak time t _p and the maximum overshoot ΔP(t _p ) of the offset are:

As shown in Figure 5, it is the response curve of IIDG output active power offset ΔP. It can be seen from the figure that as the virtual inertia constant H increases, the overshoot of the IIDG output active power offset ΔP continues to increase, and the fluctuation becomes more severe. In other words, within this value range, a smaller virtual inertia constant H is more beneficial to active power regulation.

2) When the common bus frequency has a small-range sudden change, Δω _grid = α(u(t)-u(t-τ ₀ )), where: α represents the sudden change amplitude, u(t) is the unit step function, τ ₀ It is a sudden change period, at this time, the IIDG output frequency deviation is:

The first peak time of the offset

Maximum overshoot

As shown in Figure 6, it is the response curve of IIDG output frequency offset Δω when the frequency changes suddenly; as can be seen from the figure, as the virtual inertia constant H increases, the IIDG output frequency offset Δω overshoot continues to decrease, and the overall fluctuation More gentle. In other words, within this value range, a larger virtual inertia constant H is more beneficial to frequency adjustment.

There are contradictions and differences between the magnitude of the virtual inertia H and the output active power and frequency of the IIDG system under different operating scenarios. On the one hand, when the IIDG is subject to sudden changes in the frequency of disturbances from the external system, the increase in the virtual inertia constant can reduce the IIDG output frequency offset. Smaller, the frequency fluctuation is smoother; on the other hand, when the IIDG’s own active power output changes, reducing the virtual inertia constant can reduce the IIDG’s output active power offset and make the active power adjustment more stable.

Therefore, the two-layer adaptive inertia control method involved in this embodiment can not only adjust the offset adaptively, but also select the control priority adaptively. The relationship between the adaptive virtual inertia H and ω is shown in Figure 7, and the adaptive control virtual Inertia

Among them: k _a is the adaptive control sensitivity factor, H ₀ is the virtual inertia constant used by the control algorithm when the IIDG works at power frequency, and H _h is the virtual inertia constant corresponding to the infinite frequency offset; when the frequency offset reaches 1/k _{At a} , the virtual inertia will be (H ₀ +H _h )/2, which is the median value of the adaptive adjustment zone. Frequency offset area is less than 1 / k _a sensitive region in response to a smaller volume used in this area. The area where the frequency deviation is greater than 1/k _a is the over-leveling suppression area, and the inertia of this area is relatively large.

Therefore, in Fig. 7 ω= _ωref ±1/k _a becomes the boundary between the over-leveling suppression area and the response-sensitive area, and k _a can be used to adjust the relative size of the response-sensitive area and the over-leveling suppression area. This parameter characterizes the sensitivity of adaptive control: as k _a increases, the response sensitive area becomes smaller, and the adjustment scale continues to decrease; but as k _a increases, the slope of the curve at the same frequency offset increases, which means that the control system becomes more sensitive. To be sensitive, a small state change can cause parameter adjustment. The four curves in FIG. 7 Γ ₁ Γ ₄ corresponding to decreasing sensitivity factor, Γ ₄ wherein the sensitivity factor of the corresponding zero.

The adaptive adjustment offset refers to: taking Γ ₁ as an example, when the IIDG system suffers from frequency disturbances, the frequency operating state deviates from the stable operating point, the control system will enter the response sensitive area, the system responds quickly, and the influence of external fluctuations is suppressed . When the frequency deviation is more serious, the control will enter the over-leveling zone. The inertia in this area is relatively large, which greatly reduces the influence of external frequency fluctuations on the frequency output of IIDG itself, and the output frequency of IIDG will remain flat without major fluctuations. In the extreme case, when the frequency offset is infinite, the virtual inertia will be H _h , so H _h is the upper limit of the entire adaptive virtual inertia adjustment. When the IIDG output frequency has no deviation, the virtual inertia constant used by the control algorithm is H ₀ , which is the lower limit of adaptive virtual inertia constant adjustment. During the adjustment process of the virtual inertia H with the change of the IIDG output frequency ω, its value is always greater than zero, and the control system runs above the asymptote. This makes the control system always have positive damping, and the characteristic root is always located on the left side of the virtual axis, ensuring that the system stability is not damaged during the adjustment process.

The adaptive selection control priority refers to: in order to adapt the IIDG to different operating scenarios, the sensitivity coefficient is used

Among them: k _g is a preset coefficient, k _d reflects the frequency offset weight, the larger the k _d , the more serious the output frequency offset, so k _{a is} used to adaptively select the control priority; when the output frequency offset When it is more serious, k _a increases, and the actual control curve is close to Γ ₁ ; when the output power deviation is serious, k _a decreases, and the actual control curve is close to Γ ₄ , and in extreme cases, it reaches Γ _4. During the entire adjustment process The virtual inertia is at the minimum value, which can ensure fast response of active power output, small overshoot and good dynamic response characteristics.

In summary, in actual operation, the control system can not only adjust the output horizontally along the control curve (the x-axis direction of the graph), but also adjust the control priority order longitudinally (the y-axis direction of the graph). The above-mentioned double-layer adaptive control block diagram is shown in Figure 8.

The virtual inertia constant H is the core parameter in the VSG algorithm. Its selection range directly affects the adjustment time scale of the IIDG system, making the output characteristics of the power grid more diversified. The control system parameters of the VSG are selected in the following ways, according to this series Design reference can get the upper and lower limits of the virtual inertia constant selection range H _h , H ₀ .

1) Second-order system damping coefficient: Since the VSG control system is a second-order model, its damping coefficient

The damping coefficient ζ affects the nature of the system response. In order to make the system transient response reach a stable value faster, limit the system overshoot and make the adjustment time smaller, ζ should be 0.4-0.8, that is, the design parameters should meet:

At this time, the system is in an underdamped state, and the time response shows attenuated oscillation.

2) Droop coefficient: The droop coefficient D is determined by the grid standard. This design parameter indicates the degree of change of the inverter's output active power and reactive power for every 1Hz change in frequency and 1kV change in output voltage amplitude. The specific design standards should refer to the provisions in the relevant standards.

3) Damping coefficient: When IIDG is connected to the grid and operates stably, ω _ref is equal to ω _grid , the primary frequency modulation term (ω _ref -ω _grid )/D is zero, and the damping controller k(ω-ω _grid ) determines the active power control signal; When the IIDG island is running stably, ω is equal to ω _grid , the damping term k(ω-ω _grid ) is zero, and the primary frequency modulation controller (ω _ref -ω _grid )/D determines the active power control signal; while in actual operation, IIDG When the system is subjected to disturbances, the inverter output deviates from the set value. At this time, the damping term and the primary frequency modulation term are not zero, and they exist in the active-frequency control equation at the same time, in order to make it work under various operating conditions. The controller's regulating effect, 1/D and k should be similar in magnitude. Only in this way can the corresponding frequency offset be reasonably converted into an active power control signal.

4) Control margin: The characteristic root distribution of the transfer function shows that this VSG control system is a minimum phase system. According to the system design principle: the phase angle margin is at least 30°, and the general design is 40°～60°, namely:

among them:

Is the phase of the transfer function at ω _g ; the amplitude margin should be at least 6dB, and the general design is 10～20dB, namely:

Among them: A(ω _t ) is the amplitude of the transfer function at ω _t .

According to the above design principles, the upper and lower limits of the virtual inertia constant selection range H _h , H ₀ can be obtained.

This embodiment specifically builds a system in PSCAD/EMTDC for simulation verification. According to the system topology shown in Figure 9 and the VSG-IIDG adopts a dual-layer adaptive control strategy, the simulation parameters are shown in Table 1. The upper limit of adaptive control inertia is 1, the lower limit is 0.1, and k _g is 10.

Table 1 Main parameters of VSG simulation

In order to observe the control effect during grid-connected operation, when the operation reaches 4.2s, the fluctuation of the upper-level distribution network system causes the common bus frequency oscillation, and the fluctuation is eliminated after two power frequency cycles. At 6.2s, the power reference value increased from 0.3MW to 0.4MW. Figure 10(a) shows the change of the IIDG output frequency under the fixed virtual inertia constant and adaptive inertia control, and Figure 10(b) shows the virtual inertia value corresponding to the adaptive change.

It can be seen from the figure that after the oscillation occurs, the output frequency of the IIDG is affected and shifted. Under the action of inertia, it will eventually stabilize and return to the original operating state. It can be seen that, compared with the fixed virtual inertia constant, on the one hand, the frequency under adaptive control, the active power overshoot is smaller, and the system output is more stable; at the same time, the disturbance process under the adaptive control proceeds extremely fast, the entire oscillation is compressed, and the system can Quick recovery.

In order to observe the effect of the control strategy in island operation (breaker 2 is disconnected), when the system is connected to the grid for 4s, the power reference value drops from 0.4MW to 0.3MW. In addition, the frequency of the common bus changes at 7.2s, and the change lasts for 0.2s. Figure 11 shows the simulation results.

In the absence of frequency support from the distribution network, the IIDG output will shift after the disturbance, and then finally return to a stable operating state under the adjustment of the control system. Compared with the fixed virtual inertia constant, on the one hand, the frequency and active power overshoot under adaptive control are reduced, and the system output is more stable; at the same time, compared with large inertia control, the response speed under adaptive control is faster, and the system can recover quickly .

In summary, the double-layer adaptive inertial control strategy can take into account both output stability and dynamic response speed, effectively improve system operating performance and strengthen control effects.

The above specific implementations can be locally adjusted by those skilled in the art in different ways without departing from the principle and purpose of the present invention. The protection scope of the present invention is subject to the claims and is not limited by the above specific implementations. All implementation schemes within the scope are bound by the present invention.

Claims

A dual-layer adaptive inertia control method for inverter-type distributed power sources is characterized in that the relationship between the virtual inertia constant of the VSG and the IIDG output angular frequency is obtained according to the IIDG control topology based on the VSG, and a small signal model is established. Through the small signal model to simulate the actual operating scenario, the two-layer adaptive control strategy of adaptive adjustment of offset and adaptive control priority between power adjustment and frequency adjustment makes IIDG adapt to different operating scenarios;

The relationship between the virtual inertia constant and the IIDG output angular frequency refers to:

Among them: P is the active power output by the inverter port under VSG control, k is the damping coefficient, ω is the angular frequency of the IIDG output, ω grid is the angular frequency at the common coupling point, D is the active droop coefficient, and H is the virtual inertia of the VSG Constant, 2H represents the time required for the VSG to accelerate from the stopped state with the angular velocity of zero to the rated angular velocity under the rated power command;

The said small signal model establishes the transfer function between the input and output of VSG active power when the common bus frequency is constant as the power frequency remains unchanged:
among them:
δ s and E s are the output voltage parameters of IIDG when the power is Pref and Q ref ; when the reference value of VSG active power remains unchanged, a transfer function with common bus angular frequency fluctuation as input and VSG angular frequency as output is established:
The method according to claim 1, wherein the power adjustment refers to: when the IIDG's own active power reference value changes suddenly, a smaller virtual inertia constant H is used to adjust the active power.
The method according to claim 2, wherein when the IIDG's own active power reference value has a sudden change, ΔP ref =α*u(t), where: α represents the magnitude of the sudden change, and u(t) is the unit step function , The output power deviation of IIDG is:
among them:
The first peak time t p and the maximum overshoot ΔP(t p ) of the offset are:
The method according to claim 1, wherein the frequency adjustment refers to: when the common bus frequency has a small-range sudden change, the virtual inertia constant H is adaptively adjusted according to the IIDG frequency output change.
The method according to claim 4, wherein when the common bus frequency has a small-range sudden change, Δω grid = α(u(t)-u(t-τ 0 )), where: α represents the sudden change amplitude, u (t) is the unit step function, and τ 0 is the sudden change period. At this time, the IIDG output frequency deviation is:
The first peak time of the offset
Maximum overshoot
The method according to claim 4, characterized in that when the deviation is small, the control gives priority to the system adjustment speed, and the output overshoot problem is later, and the influence of external fluctuations is suppressed through quick response; when the deviation is large, the control The focus is on suppressing overshoot, and the system response speed is appropriately slowed down.
The method according to claim 1, wherein the adaptive adjustment offset refers to: taking Γ 1 as an example, when the IIDG system suffers frequency disturbance, the frequency operation state deviates from the stable operation point, and the control system will enter Respond to the sensitive area, the system responds quickly, and suppresses the influence of external fluctuations. When the frequency deviation is serious, the control will enter the over-leveling and flattening zone. The inertia of this area is large, which makes the influence of external frequency fluctuations on the frequency output of IIDG greatly reduced. The output frequency will remain flat without large fluctuations. In the extreme case, when the frequency offset is infinite, the virtual inertia will be H h , so H h is the upper limit of the entire adaptive virtual inertia adjustment, and when the IIDG output frequency When there is no deviation, the virtual inertia constant used by the control algorithm is H 0 , which is the lower limit of the adaptive virtual inertia constant adjustment. The virtual inertia H is always greater than zero during the adjustment process with the IIDG output frequency ω, and the control system runs Above the asymptote, this makes the control system always have positive damping, and the characteristic root is always on the left side of the virtual axis, ensuring that the system stability is not damaged during the adjustment process.
The method according to claim 1, wherein the adaptive control priority refers to adaptively selecting the control priority using a sensitivity coefficient to adapt the IIDG to different operating scenarios,
Among them: k g is a preset coefficient, k d reflects the frequency offset weight, the larger k d is, the actual control curve is close to Γ 1 ; when k a decreases, the actual control curve is close to Γ 4 , then the whole during the adjustment the virtual volume using the minimum inertia, Γ 1 Γ 4 corresponding to decreasing sensitivity coefficient, Γ 4 the corresponding sensitivity coefficients are zero.
The method according to claim 1, characterized in that the upper and lower limits of the virtual inertia constant selection range are obtained through the second-order system damping coefficient, droop coefficient, damping coefficient and control margin of the VSG.
A device for implementing the method according to any one of the preceding claims, comprising: a priority adaptive adjuster and an offset adaptive adjuster, wherein: the priority adaptive adjuster automatically adjusts power and frequency according to actual operating scenarios. The control priority is adaptively selected to improve the overall dynamic characteristics of IIDG power and frequency output. The offset adaptive regulator adjusts the virtual inertia adaptively according to the frequency output of IIDG to achieve a control strategy that takes into account both small overshoot and fast response speed.