The maximum power point tracking (MPPT) control of variable-speed wind turbine generators (WTGs), such as doubly-fed induction generators (DFIGs) and fully-rated converter-based WTGs, can extract the maximum energy from the wind, thereby providing more economic benefit than fixed-speed WTGs. However, because the MPPT control function causes the WTGs to respond to the wind speed instead of the system frequency variation, the frequency fluctuation caused by continuously varying wind speeds are severe unless mitigated. Furthermore, for electric power systems that have a high penetration level of wind power, the stochastic wind speed variation and resultant output power fluctuation of a large-scale WTG significantly deteriorates the system frequency fluctuation because the frequency response of synchronous generators is slower than the variability in the output power of a WTG caused by varying wind speeds. This paper proposes a frequency-regulating (FR) scheme of a DFIG that can decrease the maximum frequency deviation while mitigating the frequency fluctuation, particularly for an electric power system that has a high penetration level of wind power. To achieve these objectives, the proposed scheme employs an additional control loop relying on the frequency deviation operating in conjunction with the MPPT control loop. Two types of control gain of the frequency deviation loop are proposed; one control gain is modified with the rotor speed only, while the other is modified with the rotor speed and frequency deviation. In the former case, which varies with the rotor speed, the control gain is separately defined for the under-frequency section (UFS) and over-frequency section (OFS) in order to improve the FR capability of a DFIG in the OFS. To improve the FR capability in the UFS, the control gain is set to be proportional to the rotor speed. To prevent over-deceleration of the rotor speed, the control gain is set to zero at the minimum rotor speed. In contrast, in the OFS the control gain is set to be inversely proportional to the rotor speed to improve the FR capability. The control gain in the low-rotor-speed region, where a DFIG in the OFS has a high potential for absorbing the surplus energy from the power system into its rotating masses, is significantly larger than that in the UFS. Therefore, in the proposed scheme the control gain of the OFS at any rotor speed is greater than or equal to that of the UFS. In this way, the proposed scheme can increase the FR capability in the OFS. In the latter case, where the gain varies with both rotor speed and frequency deviation, a term that depends on frequency deviation is multiplied with the abovementioned gain, which varies only with the rotor speed, in order to suppress the maximum frequency deviation while mitigating the frequency fluctuation. The control gain thus varies with both rotor speed and frequency deviation. This control gain becomes larger as the frequency deviation increases. Therefore, the proposed scheme can suppress the maximum frequency deviation while mitigating the frequency fluctuation. The performance of the proposed FR scheme is investigated under various wind conditions for the IEEE 14-bus system. In the test system, to simulate a power system that has a low ramping capability, the synchronous generators are all assumed to be steam turbine generators. The effect of the frequency deviation term on the performance of the proposed scheme is also investigated. Two kinds of continuously varying input wind speeds are used in this paper as case studies for different wind power penetration levels (15% and 30%). Furthermore, to investigate the performance under larger disturbance, a load variation (medium disturbance) and a synchronous generator trip (large disturbance) are simulated in the case studies. The simulation results clearly demonstrate that the proposed FR scheme significantly lessens the output power fluctuation of a DFIG by modifying the control gain with the rotor speed and frequency deviation, thereby mitigating frequency fluctuation while suppressing the maximum frequency deviation. In addition, the FR capability of a DFIG is further improved for a higher penetration of wind power. Further, under a large disturbance the proposed scheme can help arrest the frequency decline if there exists a sufficient amount of stored kinetic energy in a WTG. The advantages of the proposed FR scheme are that it can significantly mitigate the frequency fluctuation while suppressing the maximum frequency deviation under varying wind speeds, especially in an electric power system that has high penetration level of wind. In addition, the operating speed range of a DFIG can be extended beyond that allowed by conventional schemes, helping reduce the energy storage system (ESS) capacity required to regulate the frequency fluctuation caused by varying wind speeds. However, because the proposed scheme relies on the stored kinetic energy in the rotating masses of a WTG in the UFS, it provides less contribution to FR than ESSs in low-wind conditions. In this situation, a combination of ESSs and the proposed scheme would be a promising solution to mitigate the frequency fluctuation.
List of Figures ivList of Tables viiList of Symbols viiiAbstract xiii1. Introduction 11.1 Background 11.2 Conventional FR schemes 51.3 Proposed FR scheme 82. DFIG model 102.1 Mechanical power model 112.2 Shaft model 132.3 Induction generator model 142.4 Controllers of a DFIG 163. Conventional FR schemes of a DFIG 193.1 Scheme #1 [28] 203.2 Scheme #2 [32] 234. Proposed FR scheme based on a variable gain depending on the rotor speed frequency deviation 264.1 Proposed FR scheme based on a variable gain depending on the rotor speed 274.2 Proposed FR scheme based on a variable gain depending on the rotor speed and frequency deviation 304.3 Variable gain with consideration of the power limit of a DFIG 335. Model system 405.1 IEEE 14-bus system 405.2 Synchronous generator model 426. Case studies 436.1 Varying wind speeds at wind power penetration level of 15% 456.1.1 Case 1 (Wind speeds: low/low/low) 456.1.2 Case 2 (Wind speeds: low/medium/high) 536.1.3 Case 3 (Wind speeds: medium/medium/medium) 616.1.4 Case 4 (Wind speeds: high/high/high) 656.2 Varying wind speeds at wind power penetration level of 30% 696.2.1 Case 5 (Wind speeds: low/low/low) 696.2.2 Case 6 (Wind speeds: low/medium/high) 776.2.3 Case 7 (Wind speeds: medium/medium/medium) 856.2.4 Case 8 (Wind speeds: high/high/high) 906.3 Effects of the Δf term 956.4 Effect of different varying wind speeds 1056.4.1 Case 9 (Wind speeds: low/low/low) 1066.4.2 Case 10 (Wind speeds: low/medium/high) 1146.4.3 Case 11 (Wind speeds: medium/medium/medium) 1226.4.4 Case 12 (Wind speeds: high/high/high) 1276.4.5 Effects of the Δf term 1326.5 Summary of comparison 1416.5.1 For Case 1 to Case 8 1416.5.2 For Case 9 to Case 12 1466.6 Medium/large disturbance with varying wind speeds with wind power penetration level of 30% 1496.6.1 Case 13 (Load connection and disconnection) 1496.6.2 Case 14 (Synchronous generator trip) 1547. Conclusions 159References 161요약(국문초록) 170