Data-driven fan grid-connected power transmission system short-circuit current interval prediction method

Abstract

The invention provides a short-circuit current interval prediction method for a data-driven fan grid-connected power transmission system, which belongs to the technical field of power systems and comprises the following steps: S1, acquiring operation parameters of the fan grid-connected power transmission system; s2, extracting operation characteristics of the operation parameters, inputting the extracted operation characteristics into a trained BP neural network, and outputting a short-circuit current when a short-circuit fault occurs in the current operation parameters of the fan grid-connected power transmission system by the BP neural network; and S3, inputting the short-circuit current into a short-circuit current interval prediction model, and outputting a prediction interval of the short-circuit current by the short-circuit current interval prediction model. According to the method, the short-circuit current of the fan grid-connected power transmission system can be accurately predicted based on the BP neural network, and the prediction interval of the short-circuit current is obtained through the short-circuit current interval prediction model, so that the confidence coefficient of the short-circuit current prediction result is given, and the reliability of the short-circuit current prediction result is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power systems, and in particular to a data-driven short-circuit current interval prediction method for a wind turbine grid-connected power transmission system. Background Art

[0002] With the massive commissioning of large-capacity generators and substation equipment, the continuous increase in power load, and the access to new energy sources such as wind power and nuclear power, my country's power grid has entered a new period of rapid development. The emergence of large power plants in load centers and the interconnection between large power systems have made the entire power grid present large-scale, complex and heterogeneous characteristics. At the same time, various problems have also emerged. One of the most prominent problems is that the short-circuit current level of the system has irregular changes in size and distribution with the development of the power grid. Therefore, according to the increase in system power load and changes in natural conditions, it is increasingly urgent to calculate and predict short-circuit current in real time and quickly. However, after a high proportion of new energy is connected to the grid, the traditional short-circuit current calculation method gradually becomes inaccurate, and the short-circuit current calculation method based on simulation is relatively time-consuming, and the existing short-circuit current prediction method cannot give the confidence of the result, making the reliability of the prediction result low. Summary of the invention

[0003] In view of this, the present invention provides a data-driven short-circuit current interval prediction method for a wind turbine grid-connected power transmission system to provide the confidence of the short-circuit current prediction result and improve the reliability of the short-circuit current prediction.

[0004] The technical solution adopted by the present invention to solve its technical problem is:

[0005] A data-driven short-circuit current interval prediction method for a wind turbine grid-connected power transmission system comprises the following steps:

[0006] S1. Obtaining the operating parameters of the wind turbine grid-connected power transmission system;

[0007] S2, extracting the operating characteristics of the operating parameters, inputting the extracted operating characteristics into the trained BP neural network, and the BP neural network outputs the short-circuit current when a short-circuit fault occurs in the current operating parameters of the wind turbine grid-connected power transmission system;

[0008] S3. Input the short-circuit current into a short-circuit current interval prediction model, and the short-circuit current interval prediction model outputs a prediction interval of the short-circuit current.

[0009] Preferably, before step S1, the method further comprises:

[0010] The wind turbine grid-connected power transmission system includes a doubly-fed wind turbine generator system equipped with a rotor crowbar circuit. The response phase of the doubly-fed wind turbine generator system when a short-circuit fault occurs is analyzed, and the short-circuit current calculation formula in the response phase is determined.

[0011] Establish a simulation model of the wind turbine grid-connected power transmission system, set different system operation modes to generate a large number of simulation samples;

[0012] Establishing a BP neural network, dividing the simulation samples and historical data into a training set, a validation set and a test set in proportion;

[0013] Iteratively training the BP neural network using the training set, and verifying the network performance using the verification set after each iteration;

[0014] When the iterative training is completed and the network performance is optimized, the accuracy of the BP neural network calculation is tested through the test set to obtain the trained BP neural network.

[0015] Preferably, the analysis of the response phase of the doubly-fed wind turbine power generation system when a short circuit fault occurs includes:

[0016] The response of the doubly-fed wind turbine generator system when a short-circuit fault occurs includes two stages: the first stage is the sub-transient operation stage, and the second stage is the Crowbar protection system triggering stage.

[0017] Preferably, the short-circuit current calculation formula for determining the response stage includes:

[0018] The first stage is the sub-transient operation stage. The doubly-fed wind turbine power generation system is regarded as an induction generator. The model of the induction generator is represented by a transient voltage source followed by a series stator transient reactance. The specific expression is as follows:

[0019]

[0020] In the formula, the subscripts {s, r} represent the electrical quantities of the stator winding and rotor winding of the generator, respectively; the subscripts {d, q} represent the d-axis electrical quantities and q-axis electrical quantities in the dq synchronous coordinate system, respectively; L ls and L lr are stator leakage inductance and rotor leakage inductance respectively, L m is the stator-rotor mutual inductance;

[0021] The stator short-circuit current of an induction generator can be obtained by the following formula:

[0022]

[0023] Where:

[0024]

[0025] The second stage is the Crowbar protection system triggering stage. If a short circuit occurs at the terminals of the doubly fed wind turbine generator system, the maximum short circuit current is approximately:

[0026]

[0027] In the formula, R cb is the resistance value of the Crowbar resistor.

[0028] Preferably, different system operation modes are set to generate a large number of simulation samples, including:

[0029] Select the wind turbine grid-connected power transmission system balancing synchronous machine, select the random switching line, set the basic operation mode of the wind turbine grid-connected power transmission system and set the total number of simulation samples N generated, where the active output of the synchronous machine is P Gi,base (i=1,2,…,m), the load consumes active power P loadj,base (j=1,2,…,n), the inductive reactive power absorbed by the load is Q loadj,base (j=1,2,…,n);

[0030] Randomly generate the active output of the kth group of synchronous machines and the load power consumption, where k≤N;

[0031] Randomly generate the kth group of wind speeds and set the response wind speed scenario;

[0032] Randomly generate the kth group of line switching conditions;

[0033] Set the kth group of operation modes and lines in the simulation system and initialize the power flow;

[0034] Set the fan to be off and calculate the short-circuit current I through simulation f0 ;

[0035] Set the fan input and calculate the short-circuit current I through simulation f ;

[0036] Repeat the above steps until N groups of simulation samples are obtained.

[0037] Preferably, the kth group of operation modes is set in the simulation system as follows:

[0038] P Gi,k =k Gi,k ·P Gi ,i=1,2,...,n

[0039]

[0040] Where P Gi,k is the active power output of the ith synchronous machine under the kth operation, P loadj,k is the active power consumed by the jth load under the kth operation, Q loadj,k is the inductive reactive power absorbed by the jth load under the kth operation; k Gi,k , k PLj,k , kQLj,k They are all independent random quantities in the range of [0.8,1.2], and the general distribution is uniform, so that all types of operation modes have probabilities and the possibility of occurrence is the same.

[0041] Preferably, the accuracy of the BP neural network calculation is tested by a test set, including:

[0042] The accuracy of BP neural network calculation is tested by the mean absolute percentage error (MAPE), which is defined as:

[0043]

[0044] Among them, y i is the actual result in the i-th sample, that is, the short-circuit current value obtained by simulation; is the output value predicted by the machine learning method when the i-th group of input is used, that is, the short-circuit current prediction value; N is the total number of test samples, that is, the total number of samples used to test the machine learning model.

[0045] Preferably, the construction of the short-circuit current interval prediction model includes:

[0046] The upper and lower limits of the short-circuit current interval prediction model are parameterized by BPNN, and the ensemble learner is trained to construct the short-circuit current interval prediction model.

[0047] Preferably, the short-circuit current interval prediction model is constructed by the following formula:

[0048]

[0049] Where β is the weight of the ensemble learner, c is the significance level, 100×(1-c)% is the confidence level, PICP is the prediction interval coverage probability (PI coverage probability), PINAW is the prediction interval normalized average width (PInormalized average width), and AWD is the accumulated width deviation (Accumulated width deviation).

[0050] Preferably, PICP is calculated using the following formula:

[0051]

[0052] Where: κ i is a Boolean variable;

[0053] PINAW is calculated using the following formula:

[0054]

[0055] Where: Z is the normalization factor;

[0056] AWD is calculated using the following formula:

[0057]

[0058] Where: Z is the normalization factor.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] The present invention can accurately predict the short-circuit current of the wind turbine grid-connected power transmission system based on the BP neural network, and then obtain the prediction interval of the short-circuit current through the short-circuit current interval prediction model, thereby providing the confidence of the short-circuit current prediction result and improving the reliability of the short-circuit current prediction result. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a flow chart of the data-driven wind turbine grid-connected power transmission system short-circuit current interval prediction method of the present invention.

[0062] Figure 2 Schematic diagram of the data-driven short-circuit current prediction application framework.

[0063] Figure 3 Schematic diagram of a doubly-fed wind turbine power generation system equipped with a rotor Crowbar circuit.

[0064] Figure 4 Generate a flow chart for the simulation sample.

[0065] Figure 5 Schematic diagram of the improved topology of the New England 39-node system.

[0066] Figure 6 The performance of BPNN, SVM model and EL method on the first 10 groups of samples in the test set.

[0067] Figure 7 Envelope curves of EBPNN-based PIs, RL-based PIs, and SVM-based PIs. DETAILED DESCRIPTION

[0068] The technical solutions and technical effects of the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings of the present invention.

[0069] Please see Figure 1 , a data-driven short-circuit current interval prediction method for a wind turbine grid-connected power transmission system, comprising the following steps:

[0070] S1. Obtaining the operating parameters of the wind turbine grid-connected power transmission system;

[0071] S2, extracting the operating characteristics of the operating parameters, inputting the extracted operating characteristics into the trained BP neural network, and the BP neural network outputs the short-circuit current when a short-circuit fault occurs in the current operating parameters of the wind turbine grid-connected power transmission system;

[0072] S3. Input the short-circuit current into a short-circuit current interval prediction model, and the short-circuit current interval prediction model outputs a prediction interval of the short-circuit current.

[0073] BP neural network (BPNN, Back Propagation Neural Network) learns and simulates complex electrical behavior patterns by training simulation samples and historical data. The present invention can accurately predict the short-circuit current of the wind turbine grid-connected transmission system based on the BP neural network, and then obtain the prediction interval of the short-circuit current through the short-circuit current interval prediction model, thereby giving the confidence of the short-circuit current prediction result and improving the reliability of the short-circuit current prediction result.

[0074] Furthermore, before step S1, the method further includes:

[0075] The wind turbine grid-connected power transmission system includes a doubly-fed wind turbine generator system equipped with a rotor crowbar circuit. The response phase of the doubly-fed wind turbine generator system when a short-circuit fault occurs is analyzed, and the short-circuit current calculation formula in the response phase is determined.

[0076] Establish a simulation model of the wind turbine grid-connected power transmission system, set different system operation modes to generate a large number of simulation samples;

[0077] Establishing a BP neural network, dividing the simulation samples and historical data into a training set, a validation set and a test set in proportion;

[0078] Iteratively training the BP neural network using the training set, and verifying the network performance using the verification set after each iteration;

[0079] When the iterative training is completed and the network performance is optimized, the accuracy of the BP neural network calculation is tested through the test set to obtain the trained BP neural network.

[0080] There is a mapping relationship between the short-circuit current level and the power system unit combination, operation mode and topology. The prediction of short-circuit current in power systems is a regression problem. Artificial neural networks have been widely used to solve such complex regression problems and are suitable for rapid prediction of short-circuit current. Among them, using BP neural networks to predict short-circuit current in power systems has multiple advantages. BP neural networks are very effective in dealing with complex nonlinear relationships inherent in power system dynamics. They can learn and simulate complex electrical behavior patterns by training on historical data to accurately predict short-circuit current. And the adaptability of BPNN ensures that they can continuously improve predictions as more data is obtained.

[0081] Please see Figure 2 The present invention is based on BP neural network and adopts Figure 2 The short-circuit current prediction framework of offline learning + online application is shown in the figure. In the offline training stage, BPNN trains the neural network parameters based on simulation samples and historical data as training data. After the training data is input, the training data is divided into training set, validation set and test set in proportion. The training set is the training sample for training the calculation model. The validation set is used to verify whether the performance of the network is improving. After each iteration, the validation set is used to verify the network performance. If the network performance is not optimized after several consecutive iterations, the training stops. The role of the validation set is to prevent the neural network from overfitting. The test set is used to test the accuracy of the neural network calculation. After that, the data is normalized, that is, all data is mapped to a unified scale to prevent the training results from being dominated by one or several sample features. The structure, parameters and training methods of the network will have a great impact on the training results. Therefore, it is necessary to select appropriate neural network structures and parameters according to the characteristics of the samples, and try to use multiple training methods to train and test the effects. In the online application stage, the trained BP neural network quickly predicts the short-circuit current based on the operating parameters of the wind turbine grid-connected transmission system.

[0082] Furthermore, the wind turbine grid-connected transmission system includes a doubly-fed wind turbine power generation system equipped with a rotor Crowbar circuit. In the doubly-fed wind turbine power generation system, due to the magnetic coupling between the stator and the rotor, when a short circuit fault occurs in the system, a large current will pass through the rotor loop and through the generator-side converter. Therefore, it is necessary to protect the generator-side converter during the short circuit. The protection system may be based on an external resistor connected to the rotor circuit, which is called a Crowbar resistor. The goal of the protection system is to allow the current to pass through the Crowbar resistor instead of the converter. Among them, the doubly-fed wind turbine power generation system equipped with a rotor Crowbar circuit is as follows Figure 3 shown.

[0083] Furthermore, when a short circuit occurs, the response of the doubly-fed wind turbine generator system equipped with a rotor Crowbar circuit can be divided into two stages: the first stage is the sub-transient operation stage (lasting about 2-3 cycles), at which time the doubly-fed wind turbine generator system can be regarded as an induction generator operation; the second stage is when the Crowbar protection system is triggered, the transient of the rotor flux begins to decay, and the rotor current is limited. The following is a brief introduction to the doubly-fed wind turbine model in the two stages.

[0084] The first stage is the sub-transient operation stage. The doubly-fed wind turbine power generation system is regarded as an induction generator. The model of the induction generator is represented by a transient voltage source followed by a series stator transient reactance. The specific expression is as follows:

[0085]

[0086] In the formula, the subscripts {s, r} represent the electrical quantities of the stator winding and rotor winding of the generator, respectively; the subscripts {d, q} represent the d-axis electrical quantities and q-axis electrical quantities in the dq synchronous coordinate system, respectively; L ls and L lr are stator leakage inductance and rotor leakage inductance respectively, L m is the stator-rotor mutual inductance;

[0087] The stator short-circuit current of an induction generator can be obtained by the following formula:

[0088]

[0089] Where:

[0090]

[0091] Among them, T s ' is the stator transient time constant, T r ' is the rotor transient time constant, L' s is the stator transient inductance, L' r is the transient inductance of the rotor. Although the current vector cannot reach its maximum value exactly at t = T / 2, the current half a cycle later can be well approximated to the maximum current. Therefore, the maximum current can be replaced by the current at t = T / 2. If the purpose of the analysis is to calculate the maximum value of the root mean square (rms) value of the AC current component, the stator voltage u before the fault occurs is used. s (0 - ) and stator transient reactance X′ s However, if the purpose of the analysis is to estimate the peak value of the fault current, then X s ′ / C(T / 2), not X′ s .

[0092] The second stage is the Crowbar protection system triggering stage. If a short circuit occurs at the terminals of the doubly fed wind turbine generator system, the maximum short circuit current is approximately:

[0093]

[0094] In the formula, R cb is the resistance value of the Crowbar resistor.

[0095] Thus, the maximum short-circuit current of the doubly-fed wind turbine generator can be well calculated by simply using the transient reactance of the corresponding induction motor.

[0096] For further information, see Figure 4 , set different system operation modes to generate a large number of simulation samples using Figure 4The process shown in the figure specifically selects a wind turbine grid-connected power transmission system balancing synchronous machine, selects a random switching line, sets the basic operation mode of the wind turbine grid-connected power transmission system and sets the total number of generated simulation samples N, wherein the active output of the synchronous machine is P Gi,base (i=1,2,…,m), the load consumes active power P loadj,base (j=1,2,…,n), the inductive reactive power absorbed by the load is Q loadj,base (j=1,2,…,n); randomly generate the active output of the kth group of synchronous machines and the power consumption of the load, where k≤N; randomly generate the kth group of wind speeds and set the response wind speed scenario; randomly generate the kth group of line switching conditions; set the kth group of operation mode and line in the simulation system, and initialize the power flow; set the wind turbine not to be put into use, and calculate the short-circuit current I through simulation f0 ; Set the fan input and calculate the short-circuit current I through simulation f ; Repeat the above steps until N groups of simulation samples are obtained.

[0097] Before generating samples, it is first necessary to determine the sample labels and sample features. In the power system, the main factors affecting the size of the short-circuit current are the topological structure of the power grid, the short-circuit parameters of the power equipment and the operation mode of the power system. The topological structure of the power system is relatively fixed after construction, and the power equipment update cycle is relatively long. Therefore, in terms of the topological structure and equipment parameters of the power system, the present invention only considers the switching of some lines when the topological structure and fixed equipment are determined. The operation mode of the power system will be adjusted according to changes in factors such as user demand and the supply of primary energy. Especially after the new energy is connected to the grid, the operation mode of the entire power system also has a certain randomness due to the randomness of its primary energy. Therefore, the present invention mainly considers obtaining training samples by changing the operation mode of the power system.

[0098] Among them, there are many features that describe the operation mode of the power system, including the power and power loss transmitted by each line, the voltage amplitude and phase angle of each node, the output of the generator, the load power, etc. For large power systems, the huge number of features that describe the operation mode is very unfavorable to artificial intelligence methods, and not all operation characteristics have a significant impact on the short-circuit current level, so it is not necessary to use all features as inputs to the machine learning model. Considering that the system reactive power involved in the present invention is sufficient and the generator voltage can be maintained stable under different operation modes, the load power and the generator output are used as sample features. In addition, since the traditional method of calculating the short-circuit current of the power system that only includes the rotating power source can achieve relatively accurate and fast calculation, the present invention considers taking the system short-circuit current when the doubly fed wind turbine is not put into operation as a sample feature. This sample feature reflects the operation mode of the system to a certain extent, which greatly improves the training efficiency and accuracy of the machine learning method.

[0099] Based on the above considerations, for a power system with (m+1) synchronous machines (one of which is a balancing node), n loads (constant PQ model) and l wind farms, a system operation mode is set as the basic operation mode. The generator output under this operation mode is P Gi,base (i=1,2,…,m), the active power of the load is P loadj,base (j=1,2,…,n), reactive power Q of the load loadj,base (j=1,2,…,n). Based on the basic operation mode, adjust the values ​​of synchronous motor output, wind speed and load power consumption of unbalanced units, limit the fluctuation of synchronous motor output, load active power and reactive power consumption to 0.8-1.2 times of the reference value, and let the wind speed V wind Also fluctuates randomly. Set the kth operation mode of the system in the simulation system:

[0100] P Gi,k =k Gi,k ·P Gi ,i=1,2,...,n

[0101]

[0102] Where P Gi,k is the active power output of the ith synchronous machine under the kth operation, P loadj,k is the active power consumed by the jth load under the kth operation, Q loadj,k is the inductive reactive power absorbed by the jth load under the kth operation; k Gi,k , k PLj,k , k QLj,k They are all independent random quantities in the range of [0.8,1.2], and the general distribution is uniform, so that all types of operation modes have probabilities and the possibility of occurrence is the same.

[0103] At the same time, in the wind turbine grid-connected power transmission system with double-fed wind turbines, the most common reason for the change of operation mode is the fluctuation of wind speed, so it is necessary to make the wind speed also random. According to statistics, the probability distribution of wind speed is generally normal distribution of load, so in the present invention, the normal distribution model is used to describe the wind speed. In each operation, the fixed wind speed is V wind , whose distribution load mean is 12 and variance is 4 (if the wind speed is less than 0, the wind speed is set to 0).

[0104] In addition to the change in the system's operating mode, a simple change in the system topology is considered, that is, controlling the switching of some lines. Some lines (a total of l branches) are selected as switchable lines. The selection is based on the fact that the normal operation of the system will not be affected after the line is cut off, and there will be no islands in the system after all the lines are cut off, and no generators and loads will be disconnected from the grid. Set the line state variable matrix Line = [L1, L2, ..., L l ], if the element in the matrix is ​​0, it means the corresponding line is disconnected, on the contrary, if the element in the matrix is ​​1, it means the line is put into operation. The elements in the state variable matrix Line are 0 or 1, and their probability conforms to the 0-1 distribution, and the probability of being 1 is 0.7.

[0105] Through Figure 4 The flowchart shown can obtain the simulation samples required by the BP neural network. Each sample contains complete sample features and sample labels. The specific composition of each sample is shown in the following table:

[0106] Table 1 Sample composition

[0107]

[0108] Among them, when the BP neural network is iteratively trained using the training set, the input of the BP neural network is the characteristics of each sample, and the output is the sample label, so that in the application process, the operating parameters of the wind turbine grid-connected power transmission system are obtained; the operating characteristics of the operating parameters are extracted, including the active output of the synchronous machine, wind speed, active power consumed by the load, inductive reactive power absorbed by the load, line switching state vector, and short-circuit current when the wind turbine is not put into use; the extracted operating characteristics are input into the trained BP neural network, and the short-circuit current when a short-circuit fault occurs in the current operating parameters of the wind turbine grid-connected power transmission system output by the BP neural network can be obtained.

[0109] Furthermore, the accuracy of BP neural network calculation is tested through the test set, including:

[0110] The accuracy of BP neural network calculation is tested by the mean absolute percentage error (MAPE), which is defined as:

[0111]

[0112] Among them, y i is the actual result in the i-th sample, that is, the short-circuit current value obtained by simulation; is the output value predicted by the machine learning method when the i-th group of input is used, that is, the predicted value of short-circuit current; N is the total number of test samples, that is, the total number of samples used to test the machine learning model. The smaller the MAPE, the more accurate the short-circuit current predicted by the BP neural network.

[0113] Furthermore, the prediction interval (PIs) technology shows the best interpretability due to its interpretable loss function, and PIs also has excellent scalability and the advantages of easy implementation. Therefore, the present invention obtains the prediction interval of the short-circuit current through the short-circuit current interval prediction model to improve the reliability of the short-circuit current predicted based on the BP neural network.

[0114] Since the information accuracy provided by PIs constructed by a single learner is limited, the present invention uses ensemble learning (EL) to improve the robustness and quality of PIs, and provides a PIs construction method based on ensemble backpropagation neural networks (EBPNN). Specifically, EBPNN first uses BPNN to parameterize the upper and lower limits of PIs, and then trains EL. The goal of EBPNN is mainly to construct PIs with the best quality under given c. Therefore, PIs construction is theoretically a multi-objective optimization problem, in which the decision variables are the weights of the ensemble learner, defined as β, (Where 2M represents the sum of the base DBNs involved in PIs construction). Therefore, from an optimization perspective, PIs construction can be expressed as follows:

[0115]

[0116] Where c is the significance level, 100×(1-c)% is the confidence level, PICP is the prediction interval coverage probability (PI coverage probability), PINAW is the prediction interval normalized average width (PI normalized average width), and AWD is the accumulated width deviation (Accumulated width deviation).

[0117] In order to simplify the solution process, the present invention introduces the importance factor v i c , i=1,2,3. The above formula is further transformed into a single-objective optimization problem, as shown below:

[0118]

[0119] The present invention uses particle swarm optimization (PSO) to solve the above formula. Based on this method, a short-circuit current interval prediction model based on ensemble learning can be constructed.

[0120] In some embodiments, a typical PIs refers to a predicted interval range that the actual target value will fall into with a certain probability (i.e., confidence level). PIs consist of a lower limit A cap and a specific probability 100×(1-c)%, where c is the significance level and 100×(1-c)% is the confidence level. t The training sample set of samples Then in Ψ t The mathematical definition of PI trained on θ is as follows:

[0121]

[0122] Furthermore, the PICP indicator indicates the probability that the target of the training sample set is covered by the prediction interval, which is calculated using the following formula:

[0123]

[0124] Where: k i is a Boolean variable; its definition is shown in the following formula. A higher PICP value indicates a higher quality of PI.

[0125]

[0126] Where: and Represent the i-th group of sample vectors of the input feature set and the target feature set respectively. For the convenience of description, the present invention only considers a single target in the target set, that is,

[0127] By selecting a prediction interval with an upper limit of positive infinity and a lower limit of negative infinity, the PICP and AWD values ​​can be optimized. However, this PI cannot provide any effective information. Therefore, in order to construct a PI with high information, PINAW needs to be introduced. The smaller the PINAW value, the higher the PI quality. Among them, PINAW is calculated using the following formula:

[0128]

[0129] Where: Z is the normalization factor;

[0130] AWD is used to quantify the degree to which the training target deviates from the upper or lower limit of PI and can be calculated using the following formula:

[0131]

[0132] Where: Z is the normalization factor.

[0133] Among them, the lower the AWD, the higher the quality of PI.

[0134] Example

[0135] The data-driven wind turbine grid-connected power transmission system short-circuit current interval prediction method of the present invention is verified based on the improved New England 39-node system. Compared with the standard 39-node system, the improved system replaces the grid-connected synchronous motor on node 38 with a grid-connected wind turbine of the same capacity. Among them, the topological structure of the improved New England 39-node system is as follows: Figure 5 shown.

[0136] First, the effectiveness of the BP neural network, that is, the short-circuit current prediction model based on BPNN, is verified. The fault setting is a three-phase short-circuit fault on bus2, the fault lasts for 0.1s, and the grounding impedance is 0.01Ω. Three switchable lines are set, including the connecting line of nodes 26-29, the connecting line of nodes 22-23, and the connecting line of nodes 4-14. The basic operation mode of the system is the operation mode under the initial power flow. According to the above method of setting the operation mode, 1500 groups of operation modes are generated, and 1500 groups of simulation samples are constructed as training sets through electromagnetic transient simulation. In addition, 100 groups of simulation samples are retained as test sets and do not participate in model training. BPNN contains two layers of hidden layers, both of which contain 24 neurons, and the training function is based on the Powell / Beale restarted conjugate gradient algorithm. The comparison algorithms are the Support Vector Machines (SVM) model and the Ensemble Learning (EL) method. The SVM is based on the quadratic kernel function, and the EL method is a random forest. The performance of each method on the first 10 groups of samples in the test set is shown in the figure. Figure 6 shown.

[0137] from Figure 6 It can be seen that BPNN has a good prediction effect with a MAPE less than 0.5% except for the sample numbered #3. The other two machine learning algorithms have a large gap compared with BPNN. The average effect and calculation speed of the three methods on the test set are shown in Table 2. In order to reflect the calculation efficiency, the time domain simulation (TDS) method is introduced for comparison.

[0138] Table 2 Comparison of short-circuit current prediction effect and calculation efficiency of different methods

[0139]

[0140] As can be seen from Table 2, although BPNN requires longer training time and computing time than RL and SVM, its average percentage error is only 27.24% and 22.16% of the other two methods, respectively. Compared with TDS, the BPNN method improves computing efficiency by more than 200 times on average, reaching the millisecond level, which can meet real-time computing needs.

[0141] Next, the effectiveness of the short-circuit current interval prediction model based on EBPNN is verified. In addition to the PIs based on EBPNN, PIs based on RL and SVM will also be used for comparative experiments. In particular, in order to ensure unbiased testing, the learning methods of the four PIs will be trained and tested based on the same training sample set and test sample set. The training algorithm of each PI is executed 10 times, and the average value of the PIs quality index is taken to evaluate the performance of each algorithm. The results after testing on the test sample set are shown in Figure 2. Figure 7 As shown. Figure 7 It can be seen that the PIs based on EBPNN are more accurate than those based on RL and SVM.

[0142] The quantitative evaluation indicators of each PIs are shown in Table 3. It should be noted that in Table 3, in addition to the AWD and PINAW indicators used in the present invention, the average coverage error (ACE) indicator is also used to evaluate the quality of PIs, and its calculation formula is as follows:

[0143]

[0144] Table 3 Quantitative evaluation indicators of PIs constructed by each method

[0145] method SVM RL EBPNN ACE 6.7e-3 5.1e-3 3.2e-3 PINAW 0.6823 0.6476 0.6191 AWD 0.0128 0.0084 0.0093

[0146] As can be seen from Table 3, the PIs based on SVM has the worst performance, and its three performance indicators ACE, PINAW and AWD values ​​are 6.7e-3, 0.6823 and 0.0128 respectively. The PIs driven by EBPNN in the present invention has the best performance, and its three performance indicators are 3.2e-3, 0.6191 and 0.0093 respectively.

[0147] What is disclosed above is only a preferred embodiment of the present invention, which certainly cannot be used to limit the scope of rights of the present invention. A person skilled in the art can understand that all or part of the processes of the above embodiments and equivalent changes made according to the claims of the present invention still fall within the scope of the invention.

Claims

1. A data-driven short-circuit current interval prediction method for a wind turbine grid-connected power transmission system, characterized in that: The following steps are involved: S1. Obtaining the operating parameters of the wind turbine grid-connected power transmission system; S2, extracting the operating characteristics of the operating parameters, inputting the extracted operating characteristics into the trained BP neural network, and the BP neural network outputs the short-circuit current when a short-circuit fault occurs in the current operating parameters of the wind turbine grid-connected power transmission system; S3. Input the short-circuit current into a short-circuit current interval prediction model, and the short-circuit current interval prediction model outputs a prediction interval of the short-circuit current.

2. The data-driven short-circuit current interval prediction method for wind turbine grid-connected power transmission system according to claim 1 is characterized in that: Before step S1, the method further includes: The wind turbine grid-connected power transmission system includes a doubly-fed wind turbine generator system equipped with a rotor crowbar circuit. The response phase of the doubly-fed wind turbine generator system when a short-circuit fault occurs is analyzed, and the short-circuit current calculation formula in the response phase is determined. Establish a simulation model of the wind turbine grid-connected power transmission system, set different system operation modes to generate a large number of simulation samples; establish a BP neural network, and divide the simulation samples and historical data into a training set, a validation set, and a test set in proportion; Iteratively training the BP neural network using the training set, and verifying the network performance using the verification set after each iteration; When the iterative training is completed and the network performance is optimized, the accuracy of the BP neural network calculation is tested through the test set to obtain the trained BP neural network.

3. The data-driven short-circuit current interval prediction method for wind turbine grid-connected power transmission system according to claim 2 is characterized in that: The analysis of the response stages of the doubly-fed wind turbine generator system when a short circuit fault occurs includes: the response of the doubly-fed wind turbine generator system when a short circuit fault occurs includes two stages, the first stage is the sub-transient operation stage, and the second stage is the Crowbar protection system triggering stage.

4. The data-driven short-circuit current interval prediction method for wind turbine grid-connected power transmission system according to claim 3 is characterized in that: The short-circuit current calculation formula for determining the response stage includes: The first stage is the sub-transient operation stage. The doubly-fed wind turbine power generation system is regarded as an induction generator. The model of the induction generator is represented by a transient voltage source followed by a series stator transient reactance. The specific expression is as follows: In the formula, the subscripts {s, r} represent the electrical quantities of the stator winding and rotor winding of the generator, respectively; the subscripts {d, q} represent the d-axis electrical quantities and q-axis electrical quantities in the dq synchronous coordinate system, respectively; L ls and L lr are stator leakage inductance and rotor leakage inductance respectively, L m is the stator-rotor mutual inductance; The stator short-circuit current of an induction generator can be obtained by the following formula: Where: The second stage is the Crowbar protection system triggering stage. If a short circuit occurs at the terminals of the doubly fed wind turbine generator system, the maximum short circuit current is approximately: In the formula, R cb is the resistance value of the Crowbar resistor.

5. The data-driven short-circuit current interval prediction method for wind turbine grid-connected power transmission system according to claim 2 is characterized in that: Set up different system operation modes to generate a large number of simulation samples, including: Select the wind turbine grid-connected power transmission system balancing synchronous machine, select the random switching line, set the basic operation mode of the wind turbine grid-connected power transmission system and set the total number of simulation samples N generated, where the active output of the synchronous machine is P Gi,base (i=1,2,…,m), the load consumes active power P loadj,base (j=1,2,…,n), the inductive reactive power absorbed by the load is Q loadj,base (j=1,2,…,n); Randomly generate the active output of the kth group of synchronous machines and the load power consumption, where k≤N; Randomly generate the kth group of wind speeds and set the response wind speed scenario; Randomly generate the kth group of line switching conditions; Set the kth group of operation modes and lines in the simulation system and initialize the power flow; Set the fan to be off and calculate the short-circuit current I through simulation f0 ; Set the fan input and calculate the short-circuit current I through simulation f ; Repeat the above steps until N groups of simulation samples are obtained.

6. The data-driven short-circuit current interval prediction method for wind turbine grid-connected power transmission system according to claim 5 is characterized in that: The operation mode of the kth group is set in the simulation system as follows: P Gi,k =k Gi,k ·P Gi ,i=1,2,...,n Where P Gi,k is the active power output of the ith synchronous machine under the kth operation, P loadj,k is the active power consumed by the jth load under the kth operation, Q loadj,k is the inductive reactive power absorbed by the jth load under the kth operation; k Gi,k , k PLj,k , k QLj,k They are all independent random quantities in the range of [0.8,1.2], and the general distribution is uniform, so that all types of operation modes have probabilities and the possibility of occurrence is the same.

7. The data-driven short-circuit current interval prediction method for wind turbine grid-connected power transmission system according to claim 2 is characterized in that: The accuracy of BP neural network calculation is tested through the test set, including: The accuracy of BP neural network calculation is tested by the mean absolute percentage error (MAPE), which is defined as: Among them, y i is the actual result in the i-th sample, that is, the short-circuit current value obtained by simulation; is the output value predicted by the machine learning method when the i-th group of input is used, that is, the short-circuit current prediction value; N is the total number of test samples, that is, the total number of samples used to test the machine learning model.

8. The data-driven short-circuit current interval prediction method for wind turbine grid-connected power transmission system according to claim 1 is characterized in that: The construction of the short-circuit current interval prediction model includes: The upper and lower limits of the short-circuit current interval prediction model are parameterized by BPNN, and the ensemble learner is trained to construct the short-circuit current interval prediction model.

9. The data-driven short-circuit current interval prediction method for wind turbine grid-connected power transmission system according to claim 8, characterized in that: The short-circuit current interval prediction model is constructed as follows: Where β is the weight of the ensemble learner, c is the significance level, 100×(1-c)% is the confidence level, PICP is the prediction interval coverage probability (PI coverage probability), PINAW is the prediction interval normalized average width (PInormalized average width), and AWD is the accumulated width deviation (Accumulated width deviation).

10. The data-driven short-circuit current interval prediction method for wind turbine grid-connected power transmission system according to claim 1, characterized in that: PICP is calculated using the following formula: Where: κ i is a Boolean variable; PINAW is calculated using the following formula: Where: Z is the normalization factor; AWD is calculated using the following formula: Where: Z is the normalization factor.

Images