A wind turbine test bench time delay identification method based on fractional time delay model

Abstract

The application discloses a wind turbine test bench time delay identification method based on a fractional time delay model, first, the integer order of the wind turbine test bench system time delay is acquired, then a fractional time delay model is constructed, and a transmission chain continuous model containing the time delay is obtained; an equivalent transmission chain discretization model of the wind turbine test bench containing the fractional time delay is constructed, and the stability of the equivalent transmission chain discretization model is analyzed, finally, the relationship between the time delay and the simulation multiple under the critical state is determined according to the analysis result, and the time delay identification is completed. Compared with the general time delay measurement or identification method, the scheme of the application can not rely on any measuring equipment, and does not intervene in the communication loop of the wind turbine test bench system, thereby ensuring the non-invasiveness of the identification method, further improving the accuracy of the time delay identification, and realizing the low-cost accurate identification of the time delay.

Description

Technical Field

[0001] This invention belongs to the field of time delay identification, specifically relating to a time delay identification method for wind turbine test benches based on a fractional time delay model. Background Technology

[0002] Wind turbine test benches are used to accurately simulate the electromechanical dynamics of actual wind turbines to conduct various studies and tests on behalf of actual wind turbines. Wind turbine test benches with low inertia require rotational inertia compensation strategies to bridge the inertia gap with actual wind turbines in order to achieve accurate simulation of slow dynamics. However, due to the influence of time delay, wind turbine test benches using rotational inertia compensation strategies suffer from instability and oscillation problems. Existing research has addressed this instability problem and improved system stability by adding a digital filter with parameters matched to the time delay in the compensation loop. The appropriate setting of the filter parameters is highly dependent on the accurate identification of the time delay in the inertia compensation loop.

[0003] Existing methods for obtaining time delay can be broadly categorized into two types: 1) Time delay identification methods based on known sample data, which focus on processing and analyzing known sample data to obtain time delay. These methods lack physical measurement methods for the data, have unclear application scenarios, and lack case studies for transformation and application. 2) Time delay measurement methods that focus on the acquisition path of sample data, which emphasize the use of advanced technologies or equipment to achieve direct measurement of time delay. These methods often require connection to the system's control loop and are difficult to apply to wind turbine test bench systems involving mechanical, electrical, and numerous subsystems.

[0004] Currently, the time delay identification system of wind turbine test bench is based on extracting the integer order of the time delay based on instability features, such as patent CN112906210A. However, this type of method is based on an integer model of time delay and can only identify the order of the time delay, that is, the integer part of the time delay relative to the control period. However, in the actual engineering environment, the size of the time delay is not exactly an integer multiple of the control period. Therefore, the time delay identification accuracy of this method needs to be further improved. Summary of the Invention

[0005] Based on the problems mentioned above, the purpose of this invention is to provide a time delay identification method for wind turbine test benches based on a fractional time delay model, which further improves the accuracy of time delay identification while ensuring the non-invasiveness of the identification method.

[0006] The technical solution to achieve the purpose of this invention is as follows:

[0007] A time delay identification method for wind turbine test benches based on a fractional time delay model includes the following steps:

[0008] Step 1: Obtain the integer order k0 of the time delay of the wind turbine test bench system;

[0009] Step 2: Construct a fractional time delay model D(s) and obtain a continuous transmission chain model W(s) with time delay;

[0010] Step 3: Construct the discretized model Φ0(z) of the equivalent transmission chain of the wind turbine test bench with fractional time delay;

[0011] Step 4: Stability analysis of the discretized model of the equivalent transmission chain;

[0012] Step 5: Based on the analysis results of Step 4, determine the time delay τ and the simulation multiple N under critical conditions. crit The relationship between them.

[0013] Compared with general time delay measurement or identification methods, the solution of this invention can be implemented without the aid of any measurement equipment or by intervening in the communication circuit of the wind turbine test bench system. While ensuring the non-invasiveness of the identification method, it further improves the accuracy of time delay identification and achieves accurate time delay identification at low cost.

[0014] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the steps of a wind turbine test bench delay identification method based on a fractional delay model, as described in an embodiment of the present invention.

[0016] Figure 2 This is a flowchart illustrating the steps for obtaining the integer order of the delay of the wind turbine test bench system in an embodiment of the present invention.

[0017] Figure 3 This is a block diagram of an integer time-delay transmission chain control system with compensated torque as output and unbalanced torque as input, as shown in an embodiment of the present invention.

[0018] Figure 4 This is a comparative schematic diagram of the fractional time delay transmission chain control block diagram structure for three sets of verification experiments in an embodiment of the present invention. Detailed Implementation

[0019] Combination Figure 1 A time delay identification method for wind turbine test benches based on a fractional time delay model includes the following steps:

[0020] Step 1, Combining Figure 2 To obtain the integer order k0 of the time delay of the wind turbine test bench system, specifically:

[0021] Step 1-1: Establish a transmission chain model for the wind turbine test bench based on a rotational inertia compensation strategy, and obtain the compensation torque model for the test bench during dynamic simulation of the wind turbine. Specifically:

[0022] Step 1-1-1: All parameters of the simulated wind turbine model are converted to the high-speed side. The transmission chain of the wind turbine is then simplified to a single mass block model, and its equation of motion is:

[0023]

[0024] In the formula, T a For aerodynamic torque, n g For the gearbox ratio, T g For the electromagnetic torque of the generator, To calculate the overall rotational inertia of the wind turbine after conversion to the high-speed side, J r J is the actual moment of inertia of the wind turbine. g The moment of inertia of the generator rotor. It is the rotational acceleration;

[0025] The transmission chain model of the test bench is as follows:

[0026]

[0027] In the formula, J s Let T be the moment of inertia of the test bench. s The driving torque of the test bench It is the rotational acceleration;

[0028] Step 1-1-2, assuming the wind turbine and the test bench rotate at the same speed, subtract the two equations from Step 1-1-1 and transform them to obtain the calculation equation for the driving torque of the wind turbine test bench:

[0029]

[0030] Therefore, the transmission chain model equation of the wind turbine test bench is obtained as follows:

[0031]

[0032] Let the unbalanced torque of the transmission chain be ΔT = T a / n g -T g The compensation torque model for dynamic simulation of wind turbines on the test bench.

[0033] Step 1-2: Discretize the transmission chain model of the wind turbine test bench to obtain the z-domain transfer function H(z) with the compensation torque as the output and the unbalanced torque as the input, specifically:

[0034] Based on the transmission chain model of the wind turbine test bench, considering system delay To compensate for torque T comp As the output, the unbalanced torque ΔT of the transmission chain is the input, and a discrete z-transform is performed, such as... Figure 3 As shown, the transfer function H(z) is obtained:

[0035]

[0036] In the formula, β=(J t -J s ) / J s k0 is the multiple of the delay relative to the control period T, and is defined as the delay order.

[0037] Steps 1-3: Based on the discrete model of the transmission chain of the wind turbine test bench, the unbalanced torque is taken as a step input, the system is set to zero initial state, and the time-domain response T of the compensation torque is obtained through partial power series method and inverse z-transform. comp (k), specifically:

[0038] Let the unbalanced torque ΔT of the wind turbine test bench transmission chain be a step input, and its time-domain expression is:

[0039] ΔT(k)=m·u(0)

[0040] In the formula, ΔT(k) represents the time-domain response of the unbalanced torque, k is the time series, m is the step amplitude, and u(0) is the unit step response;

[0041] If the system is initialized to zero, the compensated torque output is T. comp (z):

[0042]

[0043] T is obtained through partial power series method and inverse z-transform. comp Time-domain expression:

[0044]

[0045] In the formula, T comp (k) Time-domain response of the compensation torque It is the remainder of time series k with respect to the delay order k0+1, where the delay order k0+1 includes the k0-order communication delay and the 1-order acceleration observation delay.

[0046] Steps 1-4: Analyze the time-domain response T of the compensation torque. comp (k) Convergence-divergence analysis is performed to extract the deterministic relationship τ(T) between the oscillation period and time delay of the compensation torque during instability. OSC Specifically:

[0047] Step 1-4-1, Analyze the time-domain response T of the compensation torque. comp (k) Perform convergence / divergence analysis:

[0048] a. When β < 1, that is: J t / Js When the value is less than 2, the base of the exponential part is less than 1, and the system converges once every k0+1 steps and tends to 0, indicating that the system is stable.

[0049] b. When β = 1, that is: J t / J s When the value is 2, the base of the exponential part is equal to 1, resulting in constant amplitude oscillations and system instability.

[0050] c. When β > 1, that is: J t / J s When the value is greater than 2, the base of the exponential part is greater than 1, and it diverges gradually every k0+1 steps, causing the system to oscillate and become unstable.

[0051] Step 1-4-2: Based on the above convergence and divergence analysis, extract the time delay order k0 and the oscillation period T of the compensation torque. OSC The defining relationship is:

[0052] T OSC =2<(k0+1)cT

[0053] Step 1-4-3: Obtain the oscillation period T of the compensation torque during instability from the communication delay τ = k0·T. OSC The deterministic relationship between τ(T) and time delay τ OSC )for:

[0054] τ(T OSC )=(T OSC -2·T) / 2

[0055] Step 2: Construct a fractional time delay model D(s) and obtain a continuous transmission chain model W(s) containing time delays, specifically:

[0056] Step 2-1: Considering the case where the delay is an integer multiple of the non-control period, construct a fractional delay model D(s):

[0057]

[0058] Where k0 is the integer order of the delay obtained in step 1; λ is the fractional order of the delay; and T is the system control period.

[0059] Step 2-2: Based on the fractional time delay model D(s), determine the continuous transmission chain model W(s) with time delay:

[0060]

[0061] Steps 2-3: Performing an extended z-transform on the continuous transmission chain model W(s) with time delay yields the corresponding discrete transformation process:

[0062]

[0063] Among them, J s Let be the moment of inertia of the test bench; m = 1 - λ, 0 ≤ m < 1 are the parameters of the extended z-transform F(z, m).

[0064] Step 3: Construct the discretized model Φ0(z) of the equivalent transmission chain of the wind turbine test bench with fractional time delay, specifically as follows:

[0065] Considering the influence of the inertia compensation circuit, with the unbalanced torque ΔT as the input and the compensation torque T as the input... comp For the output, determine the transfer function of the equivalent drive train model containing fractional time delay:

[0066]

[0067] In the formula, β=(J t -J s ) / J s J t This represents the actual moment of inertia of the wind turbine.

[0068] Step 4: Stability analysis of the discretized model of the equivalent transmission chain, specifically:

[0069] Step 4-1: Determine the characteristic equation F(z) of the transfer function of the equivalent transmission chain model with fractional time delay:

[0070]

[0071] Step 4-2: Analyze the critical stability state of the wind turbine test bench system and determine β at critical stability. crit :

[0072] Let the characteristic equation F(z) = 0, and its corresponding characteristic roots lie on the unit circle when it is critically stable, z = e iθ Where 0 ≤ θ < 2π, we get:

[0073]

[0074] β crit =(J t_crit -J s ) / J s =N crit -1

[0075] Where z = e iθ Where 0 ≤ θ < 2π, J t_crit To ensure the critical stability of the wind turbine test bench system, the moment of inertia of the simulated turbine, β crit N is a real number; crit This is the simulation multiple under critical conditions.

[0076] When the rotational inertia J of the simulated object on the wind turbine test bench t>J t_crit The simulator system became unstable.

[0077] Conversely, when the moment of inertia J of the simulated object on the wind turbine test bench... t <J t_crit The simulator system is stable.

[0078] Step 5: Based on the analysis results of Step 4, determine the time delay τ and the simulation multiple N under critical conditions. crit The relationship between them is as follows:

[0079] By adjusting the inertia simulation factor N, the wind turbine test bench was brought to a critical instability state, and the simulation factor N for critical stability of the wind turbine test bench system was determined. crit β = β at critical instability crit Thus, the time delay parameter m is determined:

[0080]

[0081] Step 5-2: Determine the fractional delay λT based on the delay parameter m:

[0082] λ=1-m

[0083] Thus, the time delay and the simulation multiplier N under critical conditions are obtained. crit The relationship between them.

[0084] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0085] Example

[0086] As a specific example, in one embodiment, the delay identification method and system for wind turbine test bench based on fractional delay model of the present invention are further verified and explained.

[0087] This embodiment uses the 600kW CART3 test turbine provided by the National Renewable Energy Laboratory (NREL) of the U.S. Department of Energy as the simulation object of the wind turbine test bench. The main parameters of the wind turbine test bench with a capacity of 15kW are shown in Table 1 below.

[0088] Table 1 Parameters of the wind turbine test bench

[0089]

[0090] First, obtain the integer order of the test bench delay, k0 = 2;

[0091] Finally, time delay identification was performed using a 15kW wind turbine test bench. Digital filters were added to two sets of experiments, such as... Figure 4 As shown, the effectiveness and accuracy of the method are verified by using an equivalent transmission chain structure, and the identification results of the time delay fraction are compared.

[0092] Using a constant wind speed of 5 m / s as the input to the wind turbine test bench, multiple identification experiments were conducted based on the simulator test bench. First, the initial value of the simulation factor N of the wind turbine simulator was set to 2. The wind turbine simulator system was stable. The critical instability state was found by adjusting the simulation factor of inertia or adjusting the filter coefficient. The identification results are shown in Table 2.

[0093] Table 2 Statistical analysis of time delay identification results

[0094]

[0095] According to the statistical results in Table 2, the identification results after changing the equivalent transmission chain structure are still basically consistent with the conventional identification, with only a difference at the microsecond level. The time delay fraction is about 28ms, which shows that the identification method is still effective even after the model structure is changed, and maintains high accuracy. It can achieve identification of non-control cycle integer multiple time delays.

[0096] In summary, compared with general time delay measurement or identification methods, the present invention can achieve accurate time delay identification at low cost without relying on any measurement equipment or intervening in the communication circuit of the wind turbine test bench system.

[0097] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A method for time delay identification of a wind turbine test bench based on a fractional time delay model, characterized in that, Includes the following steps: Step 1: Obtain the integer order of the time delay of the wind turbine test bench system. ; Step 2: Construct a fractional delay model And obtain a continuous model of the transmission chain with time delay. : Step 2-1: Considering the case where the delay is an integer multiple of the non-control period, construct a fractional delay model. : ； in, The time delay is an integer order. The time delay is a fractional order; T is the system control period; Step 2-2: Based on the fractional delay model Determine the continuous model of the transmission chain with time delay. : ； Steps 2-3: For the continuous model of the transmission chain with time delay Perform the extended z-transform: ； in, The moment of inertia of the test bench; , For extended z-transform The parameters; Step 3: Construct a discretized model of the equivalent transmission chain of the wind turbine test bench with fractional time delay. : Considering the influence of the inertia compensation circuit, with unbalanced torque As input, to compensate for torque For the output, determine the transfer function of the equivalent drive train model containing fractional time delay: ； In the formula, , This represents the actual moment of inertia of the wind turbine. Step 4: Stability analysis of the discretized model of the equivalent transmission chain; Step 5: Determine the time delay based on the analysis results of Step 4. Simulation multiples under critical conditions The relationship between them.

2. The wind turbine test bench time delay identification method based on a fractional time delay model according to claim 1, characterized in that, The stability analysis of the discretized model of the equivalent transmission chain in step 4 is as follows: Step 4-1: Determine the characteristic equation of the transfer function of the equivalent transmission chain model with fractional time delay. : ； Step 4-2: Analyze the critical stability state of the wind turbine test bench system and determine the critical stability time. : ； ； ； in, , To ensure the critical stability of the wind turbine test bench system, the moment of inertia of the simulated turbine is calculated. It is a real number; This is the simulation multiple under critical conditions.

3. The wind turbine test bench time delay identification method based on a fractional time delay model according to claim 2, characterized in that, Determining the delay in step 5 Simulation multiples under critical conditions The relationship between them is as follows: Step 5-1: Determine the critical stability condition Determine the time delay parameter m: By adjusting the inertia simulation factor N, the wind turbine test bench was brought to a critical instability state, and the simulation factor for the critical stability of the wind turbine test bench system was determined. Critical instability Thus, the time delay parameter m is determined: ； Step 5-2: Determine the fractional delay based on the delay parameter m. : 。 4. A time delay identification system for a wind turbine test bench based on a fractional time delay model, characterized in that, Includes the following modules: Integer delay acquisition module: Used to acquire the integer order of the delay of the wind turbine test bench system. ; Model building module: used to build fractional time-delay models And obtain a continuous model of the transmission chain with time delay. Construct a discretized model of the equivalent transmission chain of a wind turbine test bench with fractional time delay. ; The construction of fractional delay model And obtain a continuous model of the transmission chain with time delay. Specifically: Considering the case where the delay is an integer multiple of the non-control period, construct a fractional delay model. : ； in, The time delay is an integer order. The time delay is a fractional order; T is the system control period; According to the fractional delay model Determine the continuous model of the transmission chain with time delay. : ； For a continuous model of a transmission chain with time delay Perform the extended z-transform: ； in, The moment of inertia of the test bench; , For extended z-transform The parameters; The equivalent transmission chain discretization model of the wind turbine test bench with fractional time delay is constructed. Specifically: Considering the influence of the inertia compensation circuit, with unbalanced torque As input, to compensate for torque For the output, determine the transfer function of the equivalent drive train model containing fractional time delay: ； In the formula, , This represents the actual moment of inertia of the wind turbine. Analysis module: Used for stability analysis of the discretized model of the equivalent transmission chain; Fractional delay determination module: Used to determine the delay based on the results of the analysis module. Simulation multiples under critical conditions The relationship between these factors is used to determine the fractional delay.

5. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-3.

6. A computer-storable medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-3.

Images