A genetic algorithm is proposed to optimize the yaw control system used for the stable and efficient operation of turbines in wind power plants. In particular, the factors that produce yaw static deviation are analyzed. Then, the sought optimization method for the yaw static deviation of the wind turbine is implemented by using a lidar wind meter in the engine room in order to solve the low accuracy problem caused by yaw static deviation. It is shown that fuzzy control can overcome problematic factors such as the randomness of wind direction and track the change of wind direction accurately. Power control implementation is simple, as only the voltage and current of the generator need to be measured.

Wind energy is the fastest growing clean energy among renewable energy, and the development and utilization of wind energy has become a hot spot for new energy investment. As one of the core components of the wind turbine control system, the yaw control system plays a very important role in the safe, stable and efficient operation of the wind turbine. Therefore, it is necessary to study the yaw control system of wind turbines. The yaw system is a set of devices that make the wind turbine automatically orient to the wind direction, including drive motors, gear combinations, control system devices, and wind measurement devices [

At present, scholars have studied the optimization of wind turbine yaw control strategy from the aspects of dynamics, control technology, data analysis, etc., mainly to solve the influence of yaw control performance error on yaw error, but there are few studies on yaw static deviation optimization. This article analyzes the factors that produce yaw static deviation. Taking the power generation performance of wind turbines as the optimization goal, the nacelle lidar anemometer is used to propose a wind turbine yaw static deviation optimization method to solve the problem of low wind measurement accuracy of wind turbines. In order to achieve the efficient tracking of wind direction by the wind turbine, improve the power generation efficiency of the unit, and ensure the reliability of the unit, Lakshmanan et al. [

The power absorbed by the impeller of the wind generator from the air source can be expressed as follows: ^{3}cos

Fuzzy control does not require precise mathematical models. It is not affected by nonlinear factors, can integrate expert knowledge efficiently, and has good dynamic performance and robustness. The turbine system achieves ideal control effects in terms of maximum wind energy capture of the unit, generator speed tracking monitoring, maximum power capture of the unit and wind energy robustness.

(1) Determine the structure of the fuzzy controller

According to the actual situation of the controlled object, the fuzzy controller determines the form and quantity of input variables and output variables. In general, the input variable is the error between the output variable and the given variable and the error rate of change, which is usually expressed by E (or E) and

(2) Ambiguity of input and output variables

According to the design method of the fuzzy controller, the domain of the input language variables E and is quantified into 13 levels, namely According to the design method of the fuzzy controller, the domain of the input language variables E and is quantified into 13 levels, namely. Its fuzzy subset

(3) Define fuzzy control rules

For fuzzy input and output variables, the corresponding fuzzy control rules can be determined according to a certain expert experience or fuzzy model, expressed by If... Then conditional sentences. The author also referred to the control method summarized by the actual operator, and finally obtained a series of control rules that comply with the yaw control system composed of fuzzy conditional sentences. In summary, if the error is large, the control quantity must be as short as possible; if the error is relatively small, the error must be minimized. Of course, attention must be paid to ensure the stability of the entire system and minimize unnecessary over-harmonic impact. By summarizing the fuzzy control rules, the fuzzy control rules of the yaw system are obtained as shown in

Ec | PB | PM | PS | ZE | NS | NM | NB |

PB | NB | NB | NB | NM | ZE | ZE | ZE |

PM | NB | NB | NM | NM | ZE | ZE | ZE |

PS | NM | NM | NS | NS | PS | PS | PS |

ZE | NM | NS | NS | ZE | PS | PS | PM |

NS | NS | NS | ZE | PS | PM | PM | PM |

NM | ZE | ZE | PS | PM | PB | PB | PB |

NB | ZE | ZE | PS | PM | PB | PB | PB |

Fuzzy decision of output information

Any fuzzy control rule in the fuzzy control rule table also has its own definite fuzzy relationship, in which _{1}, _{2} ⋅ ⋅ ⋅ _{49}’s respective algorithms are:

After the “union” operation process of the above 49 fuzzy relations _{i} = (

The yaw control process based on power control is shown in _{f} is the obtained power situation at A specific moment; Δ_{1}, _{2} are the power change under the condition of change, _{2} = _{f} −

The specific implementation process of yaw control is as follows: When the wind generator is connected to the grid, the yaw control system is initialized and the wind direction is determined. (1) When the wind direction _{d} changes the amount >15°, that is, jumps to Part A for fuzzy control based on the wind vane sensor to carry out the wind control, until the change of wind direction _{d} is less than 15°, the yaw motor needs to continue to rotate 5° in the original direction and then yaw 3° before power control; Judge the change of Δ_{1}, if _{d} is less than 15°, the power change will be judged, if _{d} is less than 15°, it will be run to part B to judge the power change; (3) If Δ_{1} − Δ_{2} ≤ 0 is established after 5° counterclockwise rotation of the yaw motor, it means that the yaw control direction is correct and the power control method is still used in the original direction of yaw control, then judge the change of Δ_{1}. If Δ_{1} is greater than _{1} − Δ_{2} ≤ 0 is not valid, run part C, and the yaw motor rotates 5° clockwise, and then judge according to the power change value; (4) If the yaw motor rotates counterclockwise for 5°, Δ_{1} − Δ_{2} > 0 is established, indicating that the power change is caused by the wind speed change, and the yaw motor no longer rotates, return to the initial position directly from part C through part D without yaw control, otherwise, power control is carried out, according to Δ_{1}, if the change value of Δ_{1} is greater than

Establishment of wind farm multi-objective reactive power optimization model

Within the scope of meeting many constraints, minimize the deviation of wind farm and grid voltage as much as possible, and achieve the lowest possible reactive power compensation capacity, and maximize the quality of the wind farm voltage when the conditions are met, and promote the wind farm reactive power balance is a major goal. Based on the above analysis, this article made the following key considerations when establishing the reactive power optimization model:

① How to minimize the capacity of reactive power compensation;

② To improve the voltage quality;

③ The voltage deviation of wind farm parallel point is reduced to the maximum extent.

In order to minimize the compensation capacity, the following formula is the benchmark:

In _{1} is the compensation points of SVC; _{ci} is the compensation capacity of node I.

In order to achieve the highest voltage quality, the following formula is used as a reference:

In _{i} represents the actual value of the voltage at node I; _{iset} represents the voltage rating at node I.

In _{set} represents the rated voltage value that has been clearly defined; The voltage of the junction point expressed by _{CQ} is the voltage value after reactive power compensation.

Constraints to be met: Power constraints

In _{Gi} is the value of the active power output of the wind turbine numbered I, and _{Gi} is the value of the reactive power output of the wind turbine numbered I. _{Li} represents the value of active power at node I within the wind farm, and _{Li} represents the value of reactive power at node I within the wind farm. _{i} is the voltage value at node I; _{ij} is the phase difference of the conductance between node 1 and node j, _{i}_{j} is the phase difference of the suscance between node I and node j, _{ij} is the phase difference of the voltage between node I and node j. _{2} is the number of nodes located in the wind farm.

The constraint conditions of the control variables are as follows:

In _{CiMAX} is the highest value of reactive power compensation capacity at the compensation point. _{Ci} is the value of the required compensation capacity at the compensation node I; _{1} represents the number of compensation points required by SVC; _{iMIN} is the lowest value of the variation ratio, _{i}_{MAX} is the maximum value of the variation ratio; _{i} is the specific ratio of the transformer; _{3} denotes the number of required on-load voltage regulating transformers.

Constraint conditions of state variables that should be met are expressed as follows:

In _{CiMIN} is the minimum reactive power accepted by the wind turbine, and _{CiMAX} is the maximum reactive power accepted by the wind turbine. _{Gi} is the value of reactive power that can be accepted by the unit located at node I; _{4} is the total number of generators; _{iMIN} is the minimum permissible value of the voltage at node I, _{iMAX} is the maximum permissible value of the voltage at node I; _{i} is the voltage value at all nodes within the wind farm; _{3} represents the total number of all on-load regulating transformers.

From the six expressions listed above, the required mathematical model of reactive power optimization compensation can be completed. Only by solving the appeal mathematical model can the compensation points and specific compensation capabilities required by SUC be obtained [

Because it is a power control algorithm, the detection of the power value is indispensable, and the requirements are relatively high, but it is also relatively valued. This article uses the rotor side measurement, because the input and output power capacity of the rotor side is smaller than the stator side, so the measurement results will be more accurate. The following explains the superiority of the yaw control algorithm based on power control and fuzzy logic control. By setting the following parameters, the engine room and so on the rotational inertia value is small, using A 1.5 MW wind turbine. The selected radius of the wind wheel is r = 35 m, the area of the wind wheel rotation is A = 3848 m^{2}, the efficiency of the gear box is η = 0.97, the transmission ratio is 79.8, air density ρ = 1.225 kg/m^{3}.

The simulation diagram is shown in

Fuzzy control can overcome unfavorable factors such as randomness of wind direction and accurately track changes in wind direction. The power control is simple to implement. It only needs to measure the voltage and current of the generator, and then the rotation of the yaw motor can be controlled by the change of power. The combination of the two can better reflect its superiority. The application of this method has important practical significance and provides a good idea for the operation of wind farms.