In wind turbine control, maximum power point tracking (MPPT) control is the main control mode for partial-load regimes. Efficiency potentiation of energy conversion and power smoothing are both two important control objectives in partial-load regime. However, on the one hand, low power fluctuation signifies inefficiency of energy conversion. On the other hand, enhancing efficiency may increase output power fluctuation as well. Thus the two objectives are contradictory and difficult to balance. This paper proposes a flexible MPPT control framework to improve the performance of both conversion efficiency and power smoothing, by adaptively compensating the torque reference value. The compensation was determined by a proposed model predictive control (MPC) method with dynamic weights in the cost function, which improved control performance. The computational burden of the MPC solver was reduced by transforming the cost function representation. Theoretical analysis proved the good stability and robustness. Simulation results showed that the proposed method not only kept efficiency at a high level, but also reduced power fluctuations as much as possible. Therefore, the proposed method could improve wind farm profits and power grid reliability.
1 January 2015 in International Journal of Wireless and Mobile Computing
The smart grid has caught great attention in recent years, which is poised to transform a centralised producer-controlled power network to a decentralised consumer-interactive power network that is supported by the information technology. However, it also leads to the concern that security threats in cyber network (such as wireless sensor networks) could expand to the power grid. False data injection is one of the attack threats, which is a concern for it is difficult to detect. In this paper, a distributed attack detection scheme for false data injection is proposed. Firstly, a Jacobian matrix partition algorithm is designed to divide the power system to be checked into some subsystems. Then, the attack detection method based on null space mapping is executed in each subsystem to find the injection points. The analysis shows that the proposed algorithm can detect false data injection attacks in the smart grid.
For selective catalytic reduction (SCR) in coal-fired power plants, controlling the hourly mean outlet NOx requires balancing emissions against ammonia use. An hourly mean above the target increases NOx emissions, whereas an unnecessarily low mean requires more ammonia and raises operating costs. This balance is difficult to maintain under rapidly varying inlet disturbances and long process delays. We propose a hierarchical controller, DDPG-RDC-MPC, that couples deep deterministic policy gradient (DDPG) setpoint scheduling with retrospective dynamic correction model predictive control (RDC-MPC). The DDPG agent updates the outlet NOx setpoint according to the evolving hourly mean, while RDC-MPC regulates ammonia injection and compensates for inlet disturbances without explicitly forecasting future inlet NOx. We evaluated the controller in simulation using a mechanistic SCR model identified from operating data of a 660 MW coal-fired unit. In a one-hour test under identical inlet disturbances, the framework reduced peak outlet NOx from 81.95 to 69.13 mg/Nm³ (15.6%) and moved the hourly mean from 58.75 to 49.30 mg/Nm³. This reduced deviation from the 50 mg/Nm³ target by 92.0%, while the fluctuation range narrowed from 48.20 to 39.85 mg/Nm³ (17.3%). In a separate 10-minute comparison with conventional MPC, RDC-MPC reduced peak NOx by 7.0% but increased cumulative ammonia injection by 7.2%. These simulation results indicate more accurate hourly-target tracking and better peak suppression, while the ammonia-use trade-off and performance across wider plant conditions require further evaluation.