Frequency stability in an isolated grid can be easily impacted by sudden load or wind speed changes. Many frequency regulation techniques are utilized to solve this problem. However, there are only few studies designing torque compensation controllers based on power performances in different Speed Parts. It is a major challenge for a wind turbine generator (WTG) to achieve the satisfactory compensation performance in different Speed Parts. To tackle this challenge, this paper proposes a gain scheduled torque compensation strategy for permanent magnet synchronous generator (PMSG) based wind turbines. Our main idea is to improve the anti-disturbance ability for frequency regulation by compensating torque based on WTG speed Parts. To achieve higher power reserve in each Speed Part, an enhanced deloading method of WTG is proposed. We develop a new small-signal dynamic model through analyzing the steady-state performances of deloaded WTG in the whole range of wind speed. Subsequently, H_∞ theory is leveraged in designing the gain scheduled torque compensation controller to effectively suppress frequency fluctuation. Moreover, since torque compensation brings about untimely power adjustment in over-rated wind speed condition, the conventional speed reference of pitch control system is improved. Our simulation and experimental results demonstrate that the proposed strategy can significantly improve frequency stability and smoothen power fluctuation resulting from wind speed variations. The minimum of frequency deviation with the proposed strategy is improved by up to 0.16 Hz at over-rated wind speed. Our technique can also improve anti-disturbance ability in frequency domain and achieve power balance.

The virtual synchronous generator (VSG) technology of inverter is widely used to provide the inertia and damping support for power system. However, an additional measurement device PLL (phase-locked loop) is required in the virtual synchronous generator grid connection to track the voltage phase, amplitude, and frequency, which restricts the flexible output of the distributed power generation system. To tackle this challenge, a method for grid-connected control of virtual synchronous generator based on virtual impedance is proposed. It is assumed that there is a virtual power exchange between the synchronous machine and the power grid when the virtual synchronous generator is off-grid, the virtual impedance is developed to calculate the virtual current, and when the virtual current is zero, the output voltage of the VSG can be synchronized with the voltage of the power grid, thereby seamlessly switching between off-grid and grid-connected VSG. A semiphysical simulation platform is built based on RT-LAB; simulation and experimental results show that the proposed grid synchronization control strategy of the VSG can achieve seamless transform between different VSG modes, which is simpler than the conventional synchronization control, while having a good active and reactive power tracing performance.

To tackle frequent voltage violations in distribution networks with high photovoltaic penetration ratios, voltage regulation methods that dispatch the active and reactive powers of PV panels are widely studied. Because of the costly global communication networks, in current distributed voltage regulation methods based on multi-agent reinforcement learning, each agent determines control actions according to its individual observation, which implies a partially observable Markov decision process. However, the determined control actions are far from optimal, resulting in less active power generated by PV panels and voltage violations. Some recent voltage regulation methods, where each agent determines control actions according to its historical data of the most recent time steps, can improve the optimality, but the improvement is still not enough. This paper proposes a spatiotemporal multi-agent reinforcement learning voltage regulation (ST-MARL-VR) method where each agent determines control actions according to its historical observations and those of the neighboring agents. Historical observations of neighboring agents are transferred through a point-to-point communication network, which can largely reduce the costs for communication network establishment and improve the communication robustness compared to global communication networks. Furthermore, in practice, our proposed method can be deployed on the terminal units in distribution networks. Subsequently, we validate the proposed ST-MARL-VR method under two test scenarios, and the quantitative performance improvements are above 20% compared with the two comparative methods. Comparative analyses against state-of-the-art benchmarks demonstrate that the developed ST-MARL-VR algorithm substantially enhances the optimality of the determined control actions.

Provincial power grids with high wind and photovoltaic penetration face poor out-of-sample security under limited data, whereas robust scheduling can be overly conservative. This study develops an integrated scheduling framework for secure provincial-grid operation using established two-stage Wasserstein distributionally robust optimization. The first stage determines the day-ahead schedule and reserves flexibility. After renewable forecast errors are observed, the second stage adjusts generation, renewable curtailment, reserve deployment, and energy storage charging and discharging. A Wasserstein ambiguity set is constructed from synchronized error samples. The model jointly enforces plant output, ramping, line flow, incremental loss, reserve, and storage state constraints. Dual reformulation and linearization yield a mixed-integer linear program. For the four selected day-specific cases, the estimated reductions in power-shortage frequency relative to sample-based stochastic optimization are 61.24%, 67.39%, 39.01%, and 56.11% in spring, summer, autumn, and winter, respectively. The maximum winter reserve shortfall decreases from 32.69 MW to 5.51 MW, accompanied by lower real-time balancing requirements. Compared with the robust optimization benchmark, the reported schedules and ex post revenue profiles indicate greater revenue opportunities for some renewable generators. These results demonstrate the framework’s potential to balance operational security and renewable generation revenue opportunities in provincial-grid scheduling under limited data.

With the increasing capacity of wind power generators (WTGs), the volatility of wind power could significantly challenge the stability and economy of electric and heating networks. To tackle this challenge, this paper proposes an optimal dispatch framework based on controllable load (including controllable electric load and controllable thermostatically load) to reduce wind power curtailment. A forecasting model is developed for the controllable load, which comprehensively considers autocorrelation, weather factor, and consumers’ behavior characteristics. With adjusting controllable load, an optimal dispatch model of power system is then established and resolved by Sequential Least Squares Programming (SLSQP) method. Our method is verified through numerous simulations. The results show that, compared with the state-of-the-art techniques of support vector machine and recurrent neural networks, the root mean square error with the proposed long short-term memory can be reduced by 0.069 and 0.044, respectively. Compared with conventional method, the peak wind power curtailment with dispatching controllable load is reduced by nearly 10% and 5% in two cases, respectively.