Adaptive comprehensive control strategy for primary frequency regulation of coal-fired units assisted by flywheel energy storage system

To improve the flywheel energy storage system (FESS) assisting the primary frequency regulation (PFR) of coal-fired units, an adaptive comprehensive control strategy for PFR taking into account state of charge (SOC) self-recovery is proposed. The strategy introduces an adaptive frequency deviation coefficient so that it can adaptively adjust the inertia coefficient and droop coefficient according to the PFR depth and SOC. In the SOC self-recovery stage, the SOC self-recovery output strategy of the FESS is proposed, which improves the bidirectional adjustment performance of the FESS. Finally, the proposed control strategy is verified in Matlab/Simulink.


Introduction
Randomness and volatility of new energy sources seriously threaten the stability of the power grid frequency [1].Therefore, it is necessary to introduce new means, such as FESS, to improve the overall PFR capability of the grid [2].As a high-quality PFR resource, FESS is expected to become the mainstream auxiliary PFR method for new power systems [3].
FESS has a remarkable effect on PFR [4].In [5], a coordinated operation strategy between the FESS and the wind turbine was proposed to improve the effect of PFR.Hong et al. [6] proposed a quantitative prediction model for unit real-time output increment to realize the adaptive adjustment of FESS.Luo et al. [7] proposed a coordinated control strategy to reduce unit wear, suppress reverse adjustment, and improve the SOC retention rate.
This paper fully considers the characteristics of PFR of coal-fired units assisted by FESS.It proposes an adaptive comprehensive control strategy of PFR that takes SOC self-recovery into consideration.Firstly, the three control strategies of virtual inertia control and droop control are combined.An adaptive frequency deviation coefficient is introduced to propose a control that can adaptively adjust the inertia coefficient and droop coefficient according to the PFR depth and SOC method.Then, to improve the two-way adjustment ability of the FESS, the output strategy of the FESS in the SOC self-recovery stage is proposed.Finally, a FESS auxiliary coal-fired unit PFR simulation model is built to verify its effectiveness.

Model of PFR of coal-fired unit assisted by FESS
The model of PFR of a coal-fired unit assisted by FESS is shown in Figure 1.
sF T sT sT The model of PFR of coal-fired unit assisted by FESS.

Adaptive coordinated control strategy
To improve the PFR effect, the output of the FESS is: To improve the SOC retention rate of the FESS, this paper adopts the method of variable coefficient control.The expression of the droop coefficient is:   The expression of the virtual inertia coefficient is: The conventional drooping coefficient and the inertial coefficient are only related to the state of the lotus and its power.It cannot be adjusted according to the depth of the PFR.To this end, this article proposes a control method that can adjust the coefficient according to the PFR depth.Through the introduction of adaptive frequency deviation coefficients, its expression is: The droop coefficient after introducing the adaptive frequency deviation coefficient is: The inertia coefficient after introducing the adaptive frequency deviation coefficient is: From the above formula, the adaptive droop coefficient curve can be obtained, as shown in Figure 2, and the adaptive inertia coefficient curve can be shown in Figure 3. Figure 3. Adaptive inertia coefficient curve.

SOC self-recovery control strategy
To improve the two-way adjustment capability of the FESS, within the PFR dead zone, the SOC of the FESS can be self-recovery controlled so that the FESS SOC can be restored to the best PFR range, which is the best time for the next PFR.be prepared.Assuming that the optimal SOC interval of the FESS is, the output of the FESS under SOC self-restoring control is shown in Formula (9).
It can be seen from Formula (9) that when the SOC of the FESS is less than low SOC , it is charged at a higher power; when the FESS SOC is greater than high SOC , it will be discharged at a higher power; when the system is between low SOC and high SOC , the FESS is in a standby state.Taking n as 0.08, the SOC self-recovery output curve of the FESS can be obtained, as shown in Figure 4.

Comprehensive control strategy
According to the previous analysis of the FESS participating in the PFR control strategy, the action timing of each control strategy can be determined.Then a PFR adaptive comprehensive control strategy is proposed.
(1) In the first PFR stage, when the frequency exceeds the PFR dead zone, the droop control is immediately activated, and it is used as the main PFR method throughout the entire PFR process.At the same time, the positive and negative virtual inertia control auxiliary droop control is enabled respectively according to Formula (1) to achieve a better PFR effect.
(2) In the PFR preparation stage, that is, the frequency is in the PFR dead zone, to ensure that the FESS is in the area with the strongest two-way adjustment ability, it is necessary to perform SOC selfcontrol for the FESS according to Formula (9).

PFR evaluation index
Common load disturbances in power systems are mainly divided into two types, namely step load disturbances and continuous load disturbances.The frequency characteristics of the two load disturbances are different.To accurately describe the PFR effect, different evaluation indicators are usually selected for the two load disturbances.
For step load disturbances, the evaluation indicators usually selected include the value of the maximum frequency deviation In order to more intuitively show the PFR effect under continuous disturbance, this paper proposes a comprehensive evaluation index Q, whose expression is: It can be seen that the smaller the comprehensive evaluation index is, the better the comprehensive effect of PFR and charge state retention rate is.

Simulation analysis
In this paper, a 1000 MW ultra-supercritical coal-fired unit assisted by a 6MW FESS is used as an example.According to Figure 1, a model of the PFR of a coal-fired unit assisted by FESS is built in Matlab/Simulink.To facilitate the calculation and construction of the simulation model, taking the rated capacity of the coal-fired unit of 1000 MW and the rated frequency of 50 Hz as the reference value, the simulation parameters are converted into per-unit values, and the simulation parameters are shown in Table 1.The simulation is realized under 0.015 p.u. step load disturbance and 30 min continuous load disturbance.It is compared with the constant K method and variable K method.

Step load disturbance
The frequency curve is shown in Figure 5, and the output increment curve of the coal-fired unit is shown in Figure 6.The output curve of the FESS is shown in Figure 7, the SOC curve of the FESS is shown in Figure 8, and the PFR evaluation indicators are shown in Table 2.
Figure 5. Frequency deviation curve.Figure 6.Coal-fired unit output curve.It can be seen that under the step load disturbance, the PFR effect under this article strategy is the best, which can better reduce the PFR pressure of the coal-fired unit and improve the service life of the unit.At the same time, this paper's strategy can improve the PFR effect while taking into account the SOC maintenance effect, avoiding overcharging or over-discharging of the FESS, and improving the service life of the FESS.

Continuous load disturbance
30 min continuous load disturbance is added in Figure 9.The frequency curve is shown in Figure 10, and the output increment curve of the coal-fired unit is shown in Figure 11.The output curve of the FESS is shown in Figure 12, the SOC curve of the FESS is shown in Figure 13, and the PFR evaluation indicators are shown in Table 3.  Figure 11.Coal-fired unit output curve.
Figure 12.FESS output curve.Figure 13.SOC change curve.It can be seen that under continuous disturbance, the comprehensive PFR effect under the control strategy in this paper is the best.It can take into account the SOC maintenance and PFR effect and reduces the PFR pressure of the coal-fired unit and prolongs the service life of the unit.At the same time, the SOC of the FESS under the strategy of this paper can be kept in the optimal PFR range for a long time, which is conducive to improving the PFR performance.

Conclusion
In order to alleviate the PFR pressure of coal-fired units under the high proportion of new energy penetration, a PFR adaptive comprehensive control strategy that takes into account SOC self-recovery is proposed.The simulation analysis is carried out in Matlab/Simulink, and simulation results show that the comprehensive PFR performance of this paper strategy is the best.It takes into account the SOC maintenance effect and the PFR effect and can effectively reduce the PFR pressure of the coalfired unit.
m f Δ , the frequency deterioration rate m v , and the frequency recovery time r t .The smaller m f Δ , mv , and r t are, the better the PFR effect will be.For continuous load disturbance, the root mean square value of frequency and SOC are as follows: