Active power smoothing control strategy of offshore wind power system based on reduced matrix converter

On account of the fluctuation and randomness of wind speed, the output power of wind turbines fluctuates greatly, which may cause the frequency change of the power grid and increase the difficulty of power system dispatching. To solve this problem, an active power smoothing control (APSC) strategy based on RMC for offshore wind power systems is proposed in this paper. Compared with existing wind power converter systems, RMC adopted in this paper has the advantages of compact structure, small volume and high conversion efficiency. Based on the analysis of the exponential moving average (EMA) algorithm, a method for determining active power based on the fuzzy EMA algorithm is proposed. The method adjusts the cycle number N in real-time according to the wind speed, taking into account both the wind energy utilization rate and the smoothing performance of active power. The control principle of offshore wind power generation system (WPGS) based on RMC is analyzed in detail and the coordinated control strategy of RMC and inverter is proposed. Simulation results demonstrate the validity of the proposed method and control strategy.


Introduction
With the depletion of fossil energy, new energy generation has become a feasible way to solve the energy crisis problem.The WPGS is a renewable energy technology that uses wind power to generate electricity, which takes advantage of the strong wind power at sea and converts the wind energy into electricity through wind turbines.
However, because of the random fluctuation of wind speed in nature, the change of wind speed will lead to a constant change of generator speed so that the active power generated by the generator will fluctuate with the change of wind speed.When this part of the energy is fed into the grid through the transmission line, it may affect the power quality of the grid, such as voltage deviation, frequency fluctuation, voltage flicker and so on, which increases the difficulty of the power system scheduling [1].Therefore, it is of great significance to effectively control the active power output of the WPGS to make the WPGS output smooth active power.
To solve sharp fluctuations problem in the output power of wind turbines, in recent years, research has been done on the smooth control strategy of the active power of wind power systems and put forward some solutions to the problem.The existing solutions can be divided into the following categories: (1) installation of energy storage devices; (2) adopting variable pitch control; (3) adopting generator speed control; (4) adopting coordinated control of variable pitch and speed change.In [2], by installing battery storage devices, the output power of WPGS is clipped and valley-filled to reduce the amplitude of active power fluctuation.Although this method can smooth the active power of the WPGS to a certain extent, it needs to install an auxiliary energy storage device, which increases the investment cost of the WPGS.In [3], the active power output of the WPGS is controlled by changing the pitch angle of wind turbines.Although this method can alleviate the fluctuation of active power, frequent changes of pitch angle will add mechanical stress to the pitch angle actuator and increase the mechanical strength requirement of the WPGS.The combined variable pitch and variable speed control are used to adjust the output active power of the generator.On the premise of not adding energy storage devices, the power adjustment capability of the WPGS itself is used to smooth active power output [4].
In recent years, grid-connected permanent magnet synchronous WPGS has been widely used and researched because of its unique advantages.As a new topological power converter, RMC has the following characteristics: fewer conversion stages, high power density, compact structure, high reliability and high efficiency.Based on the above advantages, RMC is very suitable for offshore wind power, which has a strict limit on installation space and has a broad application prospect.
This paper presents an APSC strategy for offshore wind power systems based on RMC.According to the analysis of the mathematical model of the WPGS, aiming at the contradiction between wind energy utilization and active power smoothing performance in wind power generation, an active power smoothing control method based on the fuzzy EMA algorithm was proposed to optimize the parameters of RMC active power reference value.The mathematical model of RMC and grid-side inverter is established.Then, the coordinated control strategy of RMC and inverter is respectively proposed.

Mathematical model of wind turbine
In the light of Betz's theory [5], the wind energy captured by the WPGS is 23 p 0.5 ( , ) where ρ is the air density, R is the radius of the blade, ν is the wind speed, λ is the blade tip velocity ratio and ω is the angular velocity of the wind turbine, which satisfies the relationship: λ=Rω/ν.β is the pitch angle.Cp (λ, β) is the wind energy utilization coefficient, which is a function of λ and β which can be expressed as: When β is constant, the relationship between Cp and λ is shown in Figure .1. From Figure .1, under a certain wind speed, if the system operates at the peak of the curve, the maximum wind energy can be obtained.At this time, the WPGS operates with the optimal blade tip velocity ratio λopt, Cp = Cpmax.Therefore, the output power of the generator is:

Reference active power setting method based on EMA algorithm
When the MPPT method is adopted, the output power of the generator fluctuates sharply with wind speed variation.Therefore, this paper proposes an APSC method based on the exponential moving average (EMA) algorithm.
The mathematical expression of the EMA algorithm is as follows: EMA(C) A (1 ) B  = + − (4) where A is the value of the current moment and B is the mean value of the previous time.α is the weight factor, α = 2/(1+N), where N is the number of periods.
The formula for obtaining the smooth active power instruction value by using the EMA algorithm is: where Popt (k) is the maximum power value when the MPPT method is adopted at the current moment.
* (k 1) P − is the smooth active power reference value calculated at the last sampling time.
It is not difficult to find from Equations ( 4) and ( 5) that the utilization rate of wind energy reaches the maximum value.The calculated reference active power is equal to the active power calculated by the MPPT method when N = 1.By taking different values of N, it can obtain the following law.i.e. the larger the value of N is, the smoother the curve of the active power reference value is, and the lower the utilization rate of wind energy is; The smaller the value of N is, the greater the curve fluctuation of the active power reference value is, the greater the utilization rate of wind energy is.

Reference active power setting method based on fuzzy EMA algorithm
According to the analysis in Section 2.2, the specific cycle number N in Equation ( 4) is a core parameter of the EMA algorithm.However, in the traditional EMA algorithm, the value of N is a constant and cannot be adjusted automatically with the fluctuation of wind speed, so it cannot achieve a high utilization rate and smooth active power at the same time.Therefore, a reference active power calculation based on a fuzzy EMA algorithm is proposed.This method can dynamically adjust the value of N according to the change of ν.When the ν fluctuates greatly, N takes a larger value to ensure smooth active power.When the fluctuation of ν is small, N takes a smaller value to ensure a higher utilization rate of wind energy.The reference active power control block diagram based on a fuzzy EMA algorithm is shown in Figure .2. Fuzzy Control Under normal circumstances, the more dimension the fuzzy controller has, the higher its precision and the better its effect will be.However, since the formulation of fuzzy rules requires experience and many attempts to complete, too many dimensions will make the setting of fuzzy rules too complicated, thus increasing the difficulty of operation.This paper uses a two-dimensional fuzzy controller, which has two inputs: one input is instantaneous wind speed ν(k) at the moment and the other input Δν(k) is the difference between the wind speed at the moment and the previous sampling moment.The output of the fuzzy controller is the number of cycles N. The relationship of the fuzzy controller is shown in Figure .3.
For each input-output variable, the range of values taken over the theoretical domain of things is called the domain of discourse of that variable.In this fuzzy controller, the domain of wind velocity ν(k) is [5.5, 13] .6, the variables are discretized into 7 files and a total of 49 fuzzy control rules are set.The design basis of fuzzy control is as follows: (1) when the fluctuation of v is small, a smaller value of N is adopted to ensure that the wind power system can obtain a higher utilization rate of wind energy; (2) when the v fluctuates greatly, N takes a larger value to make the wind turbine generate a relatively smooth active power.According to the above design rules, a fuzzy control rule table can be obtained, as shown in Table 1.

Structure and modulation strategy of offshore WPGS based on RMC
The main circuit structure of the offshore WPGS based on RMC is shown in Figure .7. It is mainly composed of the wind turbine, permanent magnet synchronous generator (PMSG), RMC and grid-side inverter.Unlike conventional SVPMW modulation, RMC outputs positive and negative alternating highfrequency voltages.RMC has six basic vectors: Iab, Iac, Icb, Iba, Ica and Iba, where S is the sector number as shown in Figure.9. Any current vector is synthesized in some sector in which the reference input phase current is situated by both adjacent basic vectors Ix1 and Iy1 (used to output positive current Ir), two basic vectors Ix2 and Iy2 (used to output negative current -Ir) and zero vector, as shown in Figure .10. Taking sector 1 as an example, the basic vectors Iab, Iac and zero vector synthesize the vector Ir, and the basic vectors Iba, Ica and zero synthesize the vector -Ir.Because the RMC can output polarity is positive and negative, the modulation strategy is called bipolar current space vector modulation strategy (B-C-SVM).
The B-C-SVM is a novel modulation strategy suitable for the current type of RMC converter.The strategy has the characteristics of sinusoidal input current, adjustable unit power factor, low output voltage and current ripple and low switching loss.

RMC converter control strategy
For the surface-mounted PMSG, the stator alternating axis and direct axis meet the following relations: Ld = Lq.Then, the generator electromagnetic torque and the stator voltage equation are: where Te is the electromagnetic torque of the generator, f is the flux linkage of the rotor, p is the number of poles of PMSG.isd, isq, usd and usq are the d and q axis components of stator current and voltage respectively.Ld and Lq are the d axis and q axis components of stator inductance respectively.s is the electromechanical angular velocity.
The circuit equation of the RMC input filter capacitor is as follows: By combining Equations ( 6) and ( 7) and simplifying them, we can get: (1 ) According to Equation (8), isd and isq can be controlled by controlling ipd and ipq to control electromagnetic torque and generator speed.
The RMC converter on the generator side controls the generator to output smooth active power according to the reference values in section 2.2.Considering the loss of the switching, the RMC inputoutput power relationship is as follows: * s 0 Cu Fe

P P P P P P P
where ∆P is the sum of all losses on the generator side, P0 is the mechanical loss of the WPGS, PCu is the copper consumption of the WPGS and PFe is the iron consumption of the WPGS.As shown in Figure .11, the RMC converter adopts a double closed-loop control strategy.The outer loop is a power loop and the actual value of the power outer loop can be calculated by measuring the voltage and current signals on the DC side of RMC.After the power deviation passes through the PI regulator, the active current component * sq i of the q axis is output and the reactive current component * sd i of the d axis is generally set to 0. The inner loop is the current loop, and the output of the PI regulator of the d and q axes is combined with decoupling components to obtain the current control components ipa and ipq of d and q axes respectively.Then, the three-phase input reference current signal of RMC is obtained after dq/abc transformation.Finally, the B-C-SVM is used to generate control signals for each switch.

Grid-side inverter control strategy
The control block diagram of the grid-side inverter is shown in Figure .12. uga, ugb and ugc are threephase AC power grid voltages.iga, igb and igc are three-phase AC power grid currents.ugd and ugq are the components of three-phase AC power grid voltage on d and q axes respectively.igd and igq are components of the three-phase AC grid current on d and q axes respectively.ωg is the angular frequency of the grid.Lg is the inductance connecting the inverter to the power grid.The phase-locked loop (PLL) is used to detect the phase of the grid voltage.
The inverter adopts the vector control technology of power grid voltage orientation.Assuming that the grid voltage synthesis vector is directed to the d axis of the d-q coordinate system.The grid voltage is projected to 0 on the q axis.In the d-q coordinate system, the active and reactive power feed into the three-phase AC grid by the inverter is: It can be seen from Equation (10) that the active and reactive power connected to the grid can be controlled by controlling the current components igd and igq on the d and q axes respectively.In Figure .12, the inverter uses double closed-loop control.The outer loop is a voltage loop and the output * gd i of its PI regulator is taken as the reference value of active current.The reactive current component * gq i is set to 0. The inner loop is a current loop and the output of the PI regulator is coupled with the decoupling term to obtain the reference input voltage ucd and ucq of the inverter under the dq axis.Then, after the dq/abc transformation, the three-phase reference input voltage uca, ucb and ucc of the inverter under the static coordinate system is obtained.Finally, the control signal of each switch of the inverter is obtained by means of SVPWM modulation.

Simulation analysis
To verify the feasibility of the proposed method, a simulation platform was built on Matlab software.Simulation parameters are as follows: 1) Wind turbine: blade radius R = 4 m, air density ρ = 1.225 kg/m 3 , rated wind speed is 12 m/s, maximum wind energy utilization coefficient Cpmax = 0.48.
3) Three-phase power grid: rated voltage is 200 V, rated frequency is 50 Hz.4) DC transmission line: dc rated voltage Vdc = 600 V, line length l =100 km. Figure .13 shows the typical curve of wind speed over a period of time.From Figure .13, it can be seen that wind speed has large fluctuations and random characteristics.Figure .14 is the active power curve of wind turbines respectively working in MPPT and power smoothing modes.It can be observed in Figure .14 that the power fluctuation range of the APSC proposed in this paper is significantly smaller than the conventional MPPT control strategy.Figure .15 shows the output voltage waveform of the RMC.The voltage is a high-frequency positive and negative alternating pulse signal.Figure .16 shows the output voltage and current waveform in phase-a of the wind generator, the phase of the voltage and current is consistent, indicating that all the active power is emitted by the generator.Figure.17 shows the curve of cycle number N. It can be seen that after adjustment by the fuzzy controller, the value of N can be adjusted in the light of variation in wind speed.

Conclusion a)
To solve the problem that the active power in offshore WPGS fluctuates sharply with the change of wind speed, this paper proposes an offshore wind power system based on RMC and its active power smooth control strategy without adding auxiliary energy storage devices.b) In this paper, a fuzzy EMA algorithm is adopted to optimize the parameters of the active power instruction value.The value of dynamic adjustment period N is used to adjust active power in real-time.
c) The B-C-SVM of RMC is analyzed.The active power smooth coordination control strategy of RMC and grid-side inverter is proposed.The simulation results validate that the method adopted in the paper can ensure a high utilization rate of wind energy and smooth output of active power.

Figure 1 .
Figure 1.Curve of Cp and λ of wind turbine.

Figure 2 .
Figure 2. Reference active power control block diagram based on fuzzy EMA algorithm.
. The domain of wind velocity change rate Δν(k) is [-5, 5].The domain of period number N is [2, 12].The fuzzy controller adopts Mamdani control rules.Membership functions for the variables are shown in Figures.4-6.

Figure 7 .Figure 8 .Figure 9
Figure 7. Structure of offshore wind power generation system based on RMC.

Figure 10
Figure 9 Input current vector distribution.Figure10Input current reference vector synthesis.

Figure 14 .
Figure 14.Active power output curve in two modes.

Figure 16 .Figure 17 .
Figure 16.Output voltage and current waveform in phase-a by generator.