Nonlinear model predictive control method and system for wind generating set

Abstract

The invention discloses a nonlinear model predictive control method and system for a wind generating set. The method comprises the following steps: constructing a nonlinear fan model; wherein the nonlinear fan model comprises a nonlinear aerodynamic model, a fan transmission chain model, a fan blade-tower drum coupling model and an actuator response model; based on a nonlinear fan model, taking wind speed disturbance, rotating speed, pitch angle, tower displacement estimation and tower speed estimation as input variables, and predicting the state of the wind turbine generator; wherein the tower drum displacement estimation and the tower drum speed estimation are obtained through estimation of a nonlinear observer; the input variables of the nonlinear observer are the measured tower acceleration, wind speed disturbance, rotating speed and pitch angle. The blade root and tower load of the wind turbine generator can be effectively reduced, weight reduction of all large components is facilitated, the cost of the generator is reduced, meanwhile, the running stability of the generator can be improved, and the running life of the generator can be prolonged.

Description

Technical Field

[0001] The present invention mainly relates to the field of wind power technology, and in particular to a nonlinear model predictive control method and system for a wind turbine generator set. Background Art

[0002] During the operation of wind turbines, wind speed disturbances can cause fluctuations in the dynamic load of the turbine, thus affecting the stability of the turbine operation. In addition, with the continuous increase in wind turbine capacity and blade flexibility, it is increasingly difficult to accurately describe the wind turbine system with a mathematical model. The traditional industrial standard PID control method relies more on the model parameters and theoretical assumptions of the control object, resulting in poor control performance, long adjustment time, and large overshoot when the input disturbance changes rapidly and significantly. Therefore, how to add the calculation of measurable disturbances to the turbine control system and increase the feedforward compensation for wind speed disturbances will be a huge bottleneck in improving the control stability of wind turbines.

[0003] The wind turbine control system plays an important role in the stability of the whole machine structure dynamics, power optimization, fatigue load reduction, etc. At present, there are mainly the following control technologies:

[0004] 1) Traditional PID control method, which linearizes the unit model and designs control parameters near the working point to achieve torque and pitch control;

[0005] 2) Intelligent control methods, including fuzzy control methods and neural network control methods. Fuzzy control uses language rules to represent experience and knowledge for control, which can overcome the influence of nonlinear factors to a certain extent; neural networks use wind speed, rotation speed, power, etc. as inputs and torque, pitch angle, etc. as outputs to construct a BP neural network to achieve control goals.

[0006] The above method can achieve the control of torque and pitch to a certain extent, but the traditional PID control method has poor anti-interference ability to input and may cause oscillation. Intelligent control methods require a large amount of data as training samples, the calculation process is relatively complex, and the interpretability is poor. Summary of the invention

[0007] In view of the technical problems existing in the prior art, the present invention provides a nonlinear model predictive control method and system for wind turbines, which can effectively reduce the blade root and tower loads of wind turbines and reduce the cost of the units, while also improving the stability of unit operation and extending the operating life of the units.

[0008] In order to solve the above technical problems, the technical solution proposed by the present invention is:

[0009] A nonlinear model predictive control method for a wind turbine generator set comprises the following steps:

[0010] Construct a nonlinear wind turbine model; the nonlinear wind turbine model includes a nonlinear aerodynamic model, a wind turbine transmission chain model, a wind turbine blade-tower coupling model and an actuator response model;

[0011] Based on a nonlinear wind turbine model, the state of the wind turbine is predicted with wind speed disturbance, rotation speed, pitch angle, tower displacement estimation and tower speed estimation as input variables; the tower displacement estimation and tower speed estimation are estimated by a nonlinear observer; the input variables of the nonlinear observer are the measured tower acceleration, wind speed disturbance, rotation speed and pitch angle.

[0012] Preferably, the fan transmission chain model is:

[0013]

[0014] Where θ is the torsion angle of the transmission shaft, ω r is the rotor speed, ω g is the generator speed, N g is the gearbox transmission ratio, J r is the equivalent moment of inertia of the rotor side (including the rotor, hub, and main shaft), J g is the equivalent moment of inertia of the generator side (including gearbox, high-speed shaft and generator), K θ is the equivalent torsional stiffness coefficient of the transmission chain system, B θ is the equivalent damping coefficient of the transmission chain system, T r is the aerodynamic torque, T g is the generator torque.

[0015] Preferably, the wind turbine blade-tower coupling model is:

[0016]

[0017] In the formula, m bld is the total blade mass, m twr Tower equivalent mass, d twr is the tower equivalent damping coefficient, c twr is the tower equivalent stiffness coefficient, R bs is the distance from the center of gravity of the blade to the center of gravity of the tower top, R bt is the distance from the thrust center to the center of gravity of the tower top, φ b is the blade flapping angle, d b is the blade damping coefficient, c b is the blade stiffness coefficient.

[0018] where d twr and c twr The calculation formula is:

[0019] m Te=0.25m T +m N +m H +3m B

[0020] d twr =4πm Te d s f0

[0021] c twr =m Te (2πf0) 2

[0022] Where m T is the tower mass, m N is the cabin mass, m H is the wheel mass, m B is the mass of a single blade, d s is the structural damping ratio, and f0 is the frequency of the tower moving forward and backward.

[0023] Preferably, the nonlinear aerodynamic model comprises:

[0024] Nonlinear aerodynamic torque T r expression:

[0025]

[0026] Nonlinear thrust F t expression:

[0027]

[0028] The wind energy utilization coefficient C p (β,λ) and air thrust coefficient C t (β,λ) is obtained by looking up the table, β is the pitch angle, λ is the tip speed ratio expression:

[0029]

[0030] v rel is the relative wind speed, and the expression is:

[0031]

[0032] where v w is the wind speed.

[0033] Preferably, the actuator response model is:

[0034]

[0035] in is the natural frequency, is the damping coefficient, and u is the pitch angle change rate control quantity.

[0036] Preferably, the nonlinear fan model is:

[0037]

[0038] Preferably, in the model predictive control, the optimization control problem of the wind turbine is described as:

[0039]

[0040] in:

[0041]

[0042] Constraints:

[0043]

[0044] x(t0)=x0

[0045]

[0046] The objective function is a quadratic expression whose weights are independent of the system state x and input u, but are allowed to depend on the external disturbance d, where the objective function is set as:

[0047]

[0048] Where W ω is the speed deviation weight, W T is the weight of the tower's forward and backward swing speed, W P is the rated power deviation weight, is the pitch rate weight, W M is the torque change weight, W θ is the pitch angle weight.

[0049] Preferably, in model predictive control, the constraint set H(x(t),u(t),d(t)) is set as follows:

[0050] The speed ω(t) limit is within the rated speed ω rated Within 114% of:

[0051] ω(t)≤1.14ω rated

[0052] The pitch angle limit is within the feasible range:

[0053] θ min ≤θ yip (r)≤θ max

[0054] The rate of change of the pitch angle and generator torque is limited to within the feasible range:

[0055]

[0056] Adding a tip speed ratio limit during control can improve the power coefficient without changing the speed limit:

[0057] λ min (v0(t))≤λ(t)≤λ max (v0(t)).

[0058] Preferably, the ordinary differential equation of the nonlinear observer is:

[0059]

[0060] Where A, B, C, D are state matrices:

[0061]

[0062] The present invention further discloses a nonlinear model predictive control system for a wind turbine generator set, comprising a memory and a processor connected to each other, wherein a computer program is stored in the memory, and when the computer program is run by the processor, the steps of the method described above are executed.

[0063] Compared with the prior art, the advantages of the present invention are:

[0064] The nonlinear model predictive control method for a wind turbine generator set of the present invention mainly includes mechanism modeling and nonlinear model predictive control method, which can effectively reduce the blade root and tower load of the wind turbine generator set, which is not only beneficial to the weight reduction of various large components and the reduction of unit cost, but also can improve the stability of unit operation and extend the unit operation life.

[0065] The present invention realizes the torque and pitch control of the unit by writing algorithm logic into the main control program, and observes the algorithm state quantity based on existing sensors and other data acquisition equipment without adding additional hardware costs; based on nonlinear model predictive control, the unit can predict the future state according to the model and current measurement values, and increase feedforward compensation control, which can reduce the extreme load caused by sudden changes in wind speed and ensure the safe operation of the unit; by reducing the pitch frequency and the vibration of the tower, the stability of the generator output power is improved, which is beneficial to improving the power generation quality of the unit. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 It is a structural schematic diagram of the transmission chain model in the present invention.

[0067] Figure 2 This is a structural block diagram of the model predictive control of the present invention.

[0068] Figure 3 Schematic diagram of the nonlinear observer estimating tower displacement and velocity in the present invention. DETAILED DESCRIPTION

[0069] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

[0070] The nonlinear model predictive control method for a wind turbine generator set provided by an embodiment of the present invention specifically comprises the following steps:

[0071] Construct a nonlinear wind turbine model; wherein the nonlinear wind turbine model consists of a nonlinear aerodynamic model, a transmission chain model, a blade-tower model and an actuator response model;

[0072] like Figure 1 As shown in Figure 2, the fan transmission chain model is:

[0073]

[0074] Where θ is the torsion angle of the transmission shaft, ω r is the rotor speed, ω g is the generator speed, N g is the gearbox transmission ratio, J r is the equivalent moment of inertia of the rotor side (including the rotor, hub, and main shaft), J g is the equivalent moment of inertia of the generator side (including gearbox, high-speed shaft and generator), K θ is the equivalent torsional stiffness coefficient of the transmission chain system, B θ is the equivalent damping coefficient of the transmission chain system, T r is the aerodynamic torque, T g is the generator torque, T ls =K θ θ is the low speed shaft torsion, is the high-speed shaft torque.

[0075] Model (1) can be transformed into:

[0076]

[0077] Discretizing model (1) yields:

[0078]

[0079] The wind turbine blade-tower coupling model is:

[0080]

[0081] In the formula, m bld is the total blade mass, m twr Tower equivalent mass, dtwr is the tower equivalent damping coefficient, c twr is the tower equivalent stiffness coefficient, R bs is the distance from the center of gravity of the blade to the center of gravity of the tower top, R bt is the distance from the thrust center to the center of gravity of the tower top, φ b is the blade flapping angle, d b is the blade damping coefficient, c b is the blade stiffness coefficient.

[0082] where d twr and c twr The calculation formula is:

[0083]

[0084] Where m t is the tower mass, m N is the cabin mass, m H is the wheel mass, m B is the mass of a single blade, d s is the structural damping ratio, and f0 is the frequency of the tower moving forward and backward.

[0085] The aerodynamic model includes:

[0086] Nonlinear aerodynamic torque T r expression:

[0087]

[0088] Nonlinear thrust F t expression:

[0089]

[0090] The wind energy utilization coefficient C p (β,λ) and air thrust coefficient C t (β,λ) is obtained by looking up the table, β is the pitch angle, λ is the tip speed ratio expression:

[0091]

[0092] v rel is the relative wind speed, and the expression is:

[0093]

[0094] where v w is the wind speed.

[0095] The pitch angle actuator response model is:

[0096]

[0097] in is the natural frequency, is the damping coefficient, and u is the pitch angle change rate control quantity.

[0098] The nonlinear wind turbine model consisting of the above transmission chain model, blade-tower model, nonlinear aerodynamic model and actuator response model is:

[0099]

[0100] The state variables in model (11) are defined as follows:

[0101]

[0102] Transforming (11) into state space form is:

[0103]

[0104] In particular, the generator torque loss needs to be considered when modeling, considering the generator torque T loss The loss includes mechanical transmission loss T loss,mec and torque loss T caused by power loss loss,el , calculated by looking up the table.

[0105] Mechanical transmission loss T loss,mec The calculation is as follows (unit: kNm):

[0106] Shaft input torque 0 200 2000 4000 60000 Shaft loss torque 86 86 92 107 112

[0107] The shaft loss torque divided by the transmission ratio is the mechanical transmission loss T. loss,mec .

[0108] Torque loss T caused by power loss loss,el The calculation is as follows (unit: MW):

[0109] Generator input power 0 1 2 3 4 5 6 7 Generator power loss 0.05 0.09 0.13 0.17 0.21 0.25 0.29 0.33

[0110] The torque generated by the power loss of the generator divided by the generator speed is the torque T loss,el , considering the torque loss of the generator, in model (1) T loss Added to the generator torque T g .

[0111] Based on the nonlinear wind turbine model constructed in the above steps, the wind speed disturbance, rotation speed, pitch angle, tower displacement estimation and tower speed estimation are used as input variables to predict the state of the wind turbine; the corresponding model predictive control is as follows Figure 2 As shown, the specific implementation steps are as follows:

[0112] The optimal control problem of wind turbines can be described as follows:

[0113]

[0114] in:

[0115]

[0116] Constraints:

[0117]

[0118] x(t0)=x0

[0119]

[0120] The objective function is a quadratic expression whose weights are independent of the system state x and input u, but are allowed to depend on the external disturbance d, where the objective function is set as:

[0121]

[0122] Where W ω is the speed deviation weight, W T is the weight of the tower's forward and backward swing speed, W P is the rated power deviation weight, is the pitch rate weight, W M is the torque change weight, W θ is the pitch angle weight. In particular, the pitch angle weight W θ Valid below rated wind speed.

[0123] The constraint set H(x(t),u(t),d(t)) is set as follows:

[0124] 1. The speed ω(t) limit is within the rated speed ω rated Within 114% of:

[0125] ω(t)≤1.14ω rated

[0126] 2. The pitch angle limit is within the feasible range:

[0127] θ min ≤θ tip (t)≤θ max

[0128] 3. The rate of change of pitch angle and generator torque is limited within the feasible range:

[0129]

[0130] 4. If the tip speed ratio is simply tracked during control, the load on some axes may increase significantly. Therefore, the tip speed ratio limit is added to improve the power coefficient without changing the speed limit:

[0131] λ min (v0(t))≤λ(t)≤λ max (v0(t))

[0132] Prediction time domain T f The time step is set to 10s, and the differential equation is solved by the fourth-order explicit Runge-Kutta method. The time step is set to 0.2s, which is consistent with the update frequency of the anemometer, so there will be 50 segments. After each optimization, one segment is selected for system feedforward control.

[0133] The model predictive controller requires a full state vector at the beginning of the optimization range. In practical applications, the displacement and speed of the tower's forward and backward swing cannot be directly measured, so it is necessary to construct a Figure 3 The observer shown is used to reconstruct x T and

[0134] The nonlinear observer consists of a static nonlinear observer for aerodynamic thrust and a linear Romberg observer, which is used to calculate the dynamic state quantity x of the forward and backward motion of the tower. T and Make an estimate.

[0135] The filtered measured output pitch angle and speed are set to 0 through a nonlinear equation to ignore the influence of tower movement and estimate the aerodynamic thrust. Then pass and the filtered tower acceleration To estimate

[0136] The ordinary differential equation of the observer is:

[0137]

[0138] Where A, B, C, D are state matrices:

[0139]

[0140] where u L is the aerodynamic thrust, L yL is the state machine gain matrix, k T is the aerodynamic constant, c Te is the aerodynamic damping constant, m is the tower mass, T e is the kinetic time constant.

[0141] The nonlinear model predictive control method for a wind turbine generator set of the present invention mainly includes mechanism modeling and nonlinear model predictive control method, which can effectively reduce the blade root and tower load of the wind turbine generator set, which is not only beneficial to the weight reduction of various large components and the reduction of unit cost, but also can improve the stability of unit operation and extend the unit operation life.

[0142] The present invention realizes the torque and pitch control of the unit by writing algorithm logic into the main control program, and observes the algorithm state quantity based on existing sensors and other data acquisition equipment without adding additional hardware costs; based on nonlinear model predictive control, the unit can predict the future state according to the model and current measurement values, and increase feedforward compensation control, which can reduce the extreme load caused by sudden changes in wind speed and ensure the safe operation of the unit; by reducing the pitch frequency and the vibration of the tower, the stability of the generator output power is improved, which is beneficial to improving the power generation quality of the unit.

[0143] The present invention can effectively reduce the extreme tower and blade root loads during sudden wind speed changes; it can effectively reduce power fluctuations and improve power generation quality; the algorithm has strong versatility and can be quickly expanded and transplanted according to units of different power levels and specifications.

[0144] The present invention also discloses a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above method are executed. The present invention further discloses a nonlinear model predictive control system for a wind turbine generator set, including a memory and a processor connected to each other, on which a computer program is stored, and when the computer program is executed by the processor, the steps of the above method are executed. The medium and system of the present invention correspond to the above method and also have the advantages described in the above method.

[0145] The present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by hardware related to computer program instructions. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned method embodiment when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. Computer-readable storage media include: any entity or device that can carry computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. The memory is used to store computer programs and / or modules, and the processor implements various functions by running or executing computer programs and / or modules stored in the memory, and calling data stored in the memory. The memory may include high-speed random access memory and may also include non-volatile memory, such as a hard disk, an internal memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0146] The above are only preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should be regarded as the protection scope of the present invention.

Claims

1. A nonlinear model predictive control method for a wind turbine generator set, characterized in that: Includes steps: Construct a nonlinear wind turbine model; the nonlinear wind turbine model includes a nonlinear aerodynamic model, a wind turbine transmission chain model, a wind turbine blade-tower coupling model and an actuator response model; Based on the nonlinear wind turbine model, the state of the wind turbine is predicted with wind speed disturbance, rotation speed, pitch angle, tower displacement estimation and tower speed estimation as input variables; The tower displacement estimation and tower velocity estimation are obtained by estimating through a nonlinear observer; the input variables of the nonlinear observer are the measured tower acceleration, wind speed disturbance, rotation speed and pitch angle.

2. The nonlinear model predictive control method for wind turbine generator set according to claim 1, characterized in that: The fan transmission chain model is: Where θ is the torsion angle of the transmission shaft, ω r is the rotor speed, ω g is the generator speed, N g is the gearbox transmission ratio, J r is the equivalent moment of inertia of the rotor side, J g is the equivalent moment of inertia of the generator side, K θ is the equivalent torsional stiffness coefficient of the transmission chain system, B θ is the equivalent damping coefficient of the transmission chain system, T r is the aerodynamic torque, T g is the generator torque.

3. The nonlinear model predictive control method for wind turbine generator set according to claim 2, characterized in that: The fan blade-tower coupling model is: In the formula, m bld is the total blade mass, m twr Tower equivalent mass, d twr is the tower equivalent damping coefficient, c twr is the tower equivalent stiffness coefficient, R bs is the distance from the center of gravity of the blade to the center of gravity of the tower top, R bt is the distance from the thrust center to the center of gravity of the tower top, φ b is the blade flapping angle, d b is the blade damping coefficient, c b is the blade stiffness coefficient; where d twr and c twr The calculation formula is: m Te =0.25m T +m N +m H +3m B d twr =4πm Te d s f0 c twr =m Te (2πf0) 2 Where m T is the tower mass, m N is the cabin mass, m H is the wheel mass, m B is the mass of a single blade, d s is the structural damping ratio, and f0 is the frequency of the tower moving forward and backward.

4. The nonlinear model predictive control method for wind turbine generator set according to claim 3, characterized in that: The nonlinear aerodynamic model includes: Nonlinear aerodynamic torque T r expression: Nonlinear thrust F t expression: The wind energy utilization coefficient C p (β,λ) and air thrust coefficient C t (β,λ) is obtained by looking up the table, β is the pitch angle, λ is the tip speed ratio expression: v rel is the relative wind speed, and the expression is: where v w is the wind speed.

5. The nonlinear model predictive control method for wind turbine generator set according to claim 4, characterized in that: The actuator response model is: in is the natural frequency, is the damping coefficient, and u is the pitch angle change rate control quantity.

6. The nonlinear model predictive control method for wind turbine generator set according to claim 5, characterized in that: The nonlinear fan model is:

7. The nonlinear model predictive control method for a wind turbine generator set according to any one of claims 1 to 6, characterized in that: In model predictive control, the optimization control problem of wind turbines is described as: in: Constraints: x(t0)=x0 The objective function is a quadratic expression whose weights are independent of the system state x and input u, but are allowed to depend on the external disturbance d, where the objective function is set as: Where W ω is the speed deviation weight, W T is the weight of the tower's forward and backward swing speed, W P is the rated power deviation weight, is the pitch rate weight, W M is the torque change weight, W θ is the pitch angle weight.

8. The nonlinear model predictive control method for wind turbine generator set according to claim 7, characterized in that: In model predictive control, the constraint set H(x(t),u(t),d(t)) is set as follows: The speed ω(t) limit is within the rated speed ω rated Within 114% of: ω(t)≤1.14ω rated The pitch angle limit is within the feasible range: i min ≤θ tip (t)≤θ max The rate of change of the pitch angle and generator torque is limited to within the feasible range: Adding a tip speed ratio limit during control can improve the power coefficient without changing the speed limit: λ min (v0(t))≤λ(t)≤λ max (v0(t)).

9. The nonlinear model predictive control method for a wind turbine generator set according to any one of claims 1 to 6, characterized in that: The ordinary differential equation of the nonlinear observer is: Where A, B, C, D are state matrices:

10. A nonlinear model predictive control system for a wind turbine generator, comprising a memory and a processor connected to each other, wherein a computer program is stored in the memory, characterized in that: When the computer program is executed by a processor, the computer program performs the steps of the method according to any one of claims 1 to 9.

Images