The invention discloses a variable parameter nonlinear feedback control method achieving wind turbine MPPT control. The method is achieved through a nonlinear feedback control system composed of a controller, a power generator and a wind turbine. The actual rotating speed of a wind wheel, the estimated pneumatic torque of the wind turbine and the optimal rotating speed of the wind wheel serve as the input of the controller; an electromagnetic torque reference value of the power generator is obtained by adjusting the parameters of the controller to serve as the output of the controller, wherein the output of the controller is the input of the power generator; and the electromagnetic torque of the power generator serves as the output of the power generator to control the actual rotating speed of the wind wheel.

Variable parameter nonlinear feedback control method achieving wind turbine MPPT control

The invention discloses a fault-tolerant control method of the stroke of a generator hydraulic servomotor. When a unit control system operates normally, the stroke of the hydraulic servomotor is controlled through a PI containing nonlinear integral. When failures of a hydraulic servomotor sensor are detected, a controller starts a software mathematic model, obtains the feedback value of the software mathematic model, the feedback value is output to a mechanical hydraulic servo system after going through PWM, power amplification and filtration, and a generator unit continues to conduct closed-loop real-time control in an automatic mode. Consequently, the unit is made to continue closed-loop real-time control in the automatic mode without need of switching to a manual operation state and halting. Moreover, at load shedding time of the unit, the unit can automatically switch to an empty-load state, so that over speed of the unit is prevented, control precision and control performance are better than those of a manual operation state, and an important role is played on safety operation of the generator unit.

Fault-tolerant control method of stroke of generator hydraulic servomotor

The invention discloses a maximum power point tracking control method for improving performance of multi-type wind turbines. The method is characterized in that a decreased torque gain control methodis applied to realize the maximum power point tracking control; and an electromagnetic torque gain coefficient is set according to a rotational inertia in wind turbine structural parameters, wherein the adopted formula is Kd=1-alpha*J, and an adjustment coefficient alpha of the gain coefficient Kd is determined according to an upper limit value and a lower limit value of the gain coefficient Kd and an upper limit value and a lower limit value of the rotational inertia J. According to the improvement method, the gain coefficient is dynamically set according to the rotational inertia of the windturbine, the wind energy capture efficiency of the multi-type wind turbines can be improved, and good adaptability is realized; meanwhile, the gain coefficient is limited to be close to the optimal gain coefficient, so that the wind turbine obtains higher wind energy capture efficiency; and the control method only depends on the wind turbine structural parameter, namely the wind turbine rotational inertia, complicated iterative calculation is not needed, and the method is simple and convenient to implement.

Maximum power point tracking control method for improving performance of multi-type wind turbines

The invention discloses a BP neural network-based low-speed rotor position observer of a permanent-magnet synchronous motor. The BP neural network-based low-speed rotor position observer comprises a BP neural network and a phase locking ring. Inductance and a magnetic flux parameter of the motor are not needed, and the BP neural network-based low-speed rotor position observer does not need to be similar to a traditional sliding-mode observer to extract counter electromotive force. By neural network self-learning, the observer can be approximate to an actual motor model. In the BP neural network-based low-speed rotor position observer, an input of the observer is voltage and current data under a stator coordinate system, trigonometric function operation relevant to coordinate conversion isnot needed, the operation quantity is reduced, and the running rate is improved. In the observer, the input of the observer comprises historical voltage and current data of the motor, the dimensionality of an effective input variable of the observer can be improved by introducing historical data, and the accuracy of the observer is further improved.

BP neural network-based low-speed rotor position observer of permanent-magnet synchronous motor

The invention relates to a tracking method and device of a maximum power point of a photovoltaic power generation system. The tracking method comprises the steps that an initial virtual error variablee0 is defined, an initial subsystem is constructed; a lyapunov function of the subsystem is constructed, on the basis that the stability condition of the lyapunov function of the subsystem is met, aninitial virtual control function is constructed, wherein the lyapunov function is shown in the description; the two steps are executed circularly until the new constructed virtual control function isthe function of a control target; when a new virtual error variable e(n+1) is defined, a variable in the last virtual control function is taken as the variable, and a lyapunov function of a new subsystem is constructed, wherein the new lyapunov function is shown in the description, and n is an integer larger than or equal to zero; according to the function of the control target, controlling is conducted to track the maximum power point so that the photovoltaic power generation system can be stable. By using an inverse method, the maximum power point of the photovoltaic power generation systemcan be rapidly tracked, and the photovoltaic voltage and photovoltaic current which are output by a photovoltaic battery are more stable.

Tracking method and device of maximum power point of photovoltaic power generation system

PURPOSE: A method for controlling a sensorless online neural network for a wind power generation control system is provided to apply an online neural network adaptive control algorithm for the maximum power point tracking (MPPT) of a wind power generation system CONSTITUTION: A method for controlling a sensorless online neural network for a wind power generation control system includes: a step of modeling mechanical power generated by a wind turbine, computing a tip-speed ratio (TSR), calculating torque due to a wind force, computing a wind energy conversion system (WECS) first model equation, and then modeling a nonlinear control equation corresponding to the uncertainty and nonlinearity of a WECS system; and a step of determining an ideal control input through an ideal nonlinear input/output linearization controller, measuring a differential value for an actual wind velocity using a differential estimator, identifying torque due to an unknown wind force through radial-basis function network (RBFN), designing a controller based on the output value of the differential estimator and the ideal control input, selecting Lyapunov function, and then implementing a control algorithm analyzing stability

Park Jang Hyun

Papers

Method of sensorless mppt neural control for wind energy conversion systems