A method and device for dividing a wind turbine fault ride-through interval

Abstract

The application provides a kind of wind turbine fault ride-through interval division method and device, determine the time when wind turbine implements control after power grid fault occurs, the time when wind turbine implements control after power grid fault is cleared, the time when wind turbine control is stable after power grid fault occurs and the time when wind turbine control is stable after power grid fault is cleared;Wind turbine fault ride-through interval is divided based on the above four time, wind turbine fault ride-through interval can be automatically divided, and it is strong in universality, and the control response characteristics of simulation model can be verified.The application is based on wind turbine fault ride-through measured data, not only can effectively solve the problem that the automatic interval division error is larger and the universality is poorer caused by the fixed time length of standard, but also can avoid artificial correction workload, and can be applied to high voltage ride-through and low voltage ride-through test at the same time.

Description

Technical Field

[0001] The present invention relates to the field of new energy technology, and in particular to a method and device for dividing a fault ride-through interval of a wind turbine generator set. Background Art

[0002] Fault ride-through simulation research for power systems incorporating wind power requires that the wind turbine electrical simulation model accurately simulate the fault ride-through response characteristics of actual wind turbines. Power system control decisions based on erroneous simulation results can lead to system disturbances, equipment damage, and even large-scale grid failures. Therefore, before conducting wind power grid integration simulation analysis, the accuracy of the fault ride-through response characteristics of the wind turbine electrical simulation model should be verified.

[0003] Properly demarcating the transient and steady-state intervals of a wind turbine's fault ride-through process is fundamental to model validation. In particular, the demarcation of the transient interval during the grid fault's duration significantly impacts model validation results. However, due to the lack of simple and effective methods for analyzing wind turbine fault ride-through characteristics and identifying states, model validation can only employ coarse demarcation methods or empirically define a fixed time duration. For example, Chinese and German standards define a fault transient interval of 100ms, with a 20ms delay for entering the 10% bandwidth of the target value as an auxiliary method for demarcating the transient and steady-state intervals. Spanish and IEC standards, however, define the fault transient interval as 150ms and 140ms, respectively, under all circumstances. Therefore, prior art demarcation of wind turbine fault ride-through intervals typically requires manual correction through test data analysis, resulting in limited versatility. Furthermore, during the transient interval of a wind turbine's fault ride-through process, the wind turbine will only implement corresponding control after a certain response delay. Because prior art cannot effectively identify the moment of control implementation, it fails to further demarcate the transient interval, making it difficult to verify the accuracy of the control response characteristics of the simulation model. Summary of the Invention

[0004] In order to overcome the shortcomings of the above-mentioned prior art in that the interval division has poor versatility and it is difficult to verify the control response characteristics of the simulation model, the present invention provides a method and device for dividing the fault ride-through interval of a wind turbine generator set, which brings the measured data of the output end of the wind turbine generator set into a pre-built artificial neural network model to determine the time when the wind turbine generator set implements control after the grid fault occurs, the time when the wind turbine generator set implements control after the grid fault is cleared, the time when the wind turbine generator set control is stable after the grid fault occurs, and the time when the wind turbine generator set control is stable after the grid fault occurs; the wind turbine generator set fault ride-through interval is divided based on the measured data of the output end of the wind turbine generator set, the time when the wind turbine generator set implements control after the grid fault occurs, the time when the wind turbine generator set implements control after the grid fault is cleared, the time when the wind turbine generator set control is stable after the grid fault occurs, and the time when the wind turbine generator set control is stable after the grid fault occurs; the measured data includes the fundamental positive sequence value of the low-voltage side bus voltage, the active power value and the three-phase current value of the generator set output end, which can realize automatic division of the fault ride-through interval of the wind turbine generator set, has strong versatility, and can verify the control response characteristics of the simulation model.

[0005] In order to achieve the above-mentioned object of the invention, the present invention adopts the following technical solutions:

[0006] In one aspect, the present invention provides a method for dividing a wind turbine generator fault ride-through interval, comprising:

[0007] The measured data at the output of the wind turbine generator set is fed into a pre-built artificial neural network model to determine the wind turbine generator set control implementation time after a grid fault occurs, the wind turbine generator set control implementation time after the grid fault is cleared, the wind turbine generator set control stabilization time after a grid fault occurs, and the wind turbine generator set control stabilization time after the grid fault is cleared;

[0008] The fault ride-through interval of the wind turbine is divided based on the measured data at the output end of the wind turbine, the time when the wind turbine implements control after the grid fault occurs, the time when the wind turbine implements control after the grid fault is cleared, the time when the wind turbine control is stable after the grid fault occurs, and the time when the wind turbine control is stable after the grid fault is cleared.

[0009] The measured data include the fundamental positive sequence value of the low-voltage side bus voltage at the output end of the generator group, the active power value and the three-phase current value.

[0010] The construction of the artificial neural network model includes:

[0011] The three-phase current at the output of the wind turbine and the transfer entropy of the three-phase current are set as neurons in the input layer;

[0012] Based on the application scenario, the neurons in the hidden layer are determined using the error back propagation algorithm;

[0013] The wind turbine control implementation moment after the grid fault occurs, the wind turbine control implementation moment after the grid fault is cleared, the wind turbine control stabilization moment after the grid fault occurs, and the wind turbine control stabilization moment after the grid fault is cleared are set as neurons in the output layer.

[0014] The construction of the artificial neural network model also includes:

[0015] Acquire historical data, the historical data including three-phase current values ​​at the output end of the wind turbine generator set, time series values ​​of transfer entropy of the three-phase current, the moment when the wind turbine generator set implements control after a grid fault occurs, the moment when the wind turbine generator set implements control after the grid fault is cleared, the moment when the wind turbine generator set control stabilizes after the grid fault occurs, and the moment when the wind turbine generator set control stabilizes after the grid fault is cleared;

[0016] Dividing the historical data into training data and test data;

[0017] Training the artificial neural network model based on the training data;

[0018] The trained artificial neural network model is tested based on the test data.

[0019] The dividing of the wind turbine generator fault ride-through interval based on the wind turbine generator control implementation moment and the wind turbine generator control stable moment includes:

[0020] Based on the grid fault occurrence time and the grid fault clearing time, the fault ride-through interval is divided into the pre-fault interval, the fault occurrence interval and the post-fault clearing interval;

[0021] Dividing the fault occurrence interval into a fault occurrence transient interval and a fault occurrence steady-state interval based on the wind turbine control stabilization moment after the grid fault occurs, and dividing the fault occurrence transient interval into a fault occurrence transition transient interval and a fault occurrence control transient interval based on the wind turbine control implementation moment after the grid fault occurs;

[0022] Based on the moment when the wind turbine control stabilizes after the grid fault is cleared, the post-fault clearing interval is divided into a post-fault clearing transient interval and a post-fault clearing steady-state interval. Based on the moment when the wind turbine implements control after the grid fault is cleared, the post-fault clearing transient interval is divided into a fault clearing transition transient interval and a fault clearing control transient interval. Based on the moment when the active power is restored after the grid fault is cleared, the post-fault clearing steady-state interval is divided into a fault clearing control steady-state interval and a fault clearing recovery steady-state interval.

[0023] The determination of the grid fault occurrence time and the grid fault recovery time includes:

[0024] The time when the fundamental positive sequence voltage value of the low-voltage side busbar at the output end of the wind turbine generator set is less than or equal to 11 times the rated voltage value of the low-voltage side busbar at the output end of the wind turbine generator set is taken as the time when the low-voltage fault ride-through occurs;

[0025] The time when the fundamental positive sequence voltage value of the low-voltage side busbar at the output end of the wind turbine generator set becomes greater than or equal to 11 times the rated voltage value of the low-voltage side busbar at the output end of the wind turbine generator set is taken as the low voltage fault ride-through clearing time;

[0026] The time when the fundamental positive sequence voltage value of the low-voltage busbar at the output end of the wind turbine generator set is greater than or equal to 12 times the rated voltage value of the low-voltage busbar at the output end of the wind turbine generator set is taken as the time when the high voltage fault ride-through occurs;

[0027] The time when the fundamental positive sequence voltage value of the low-voltage busbar at the output end of the wind turbine generator set is less than or equal to 12 times the rated voltage value of the low-voltage busbar at the output end of the wind turbine generator set is taken as the high voltage fault ride-through clearing time;

[0028] Wherein, l1 represents the low voltage fault occurrence judgment coefficient, l2 represents the high voltage fault occurrence judgment coefficient, and l1<1, l2>1.

[0029] Determining the active power recovery time after the grid fault is cleared includes:

[0030] The time corresponding to the active power value at the output of the wind turbine generator set after the grid fault is cleared is greater than or equal to 13 times the active power value at the output of the wind turbine generator set before the grid fault is cleared is taken as the active power recovery time after the grid fault is cleared;

[0031] l3 represents the active power recovery judgment coefficient, and l3<1.

[0032] The transfer entropy time series value of the three-phase current is determined by the following formula:

[0033]

[0034] Where J and I represent any two different phases, J = A, B, C, I = A, B, C; T J→I represents the time series value of the transfer entropy from the J-phase current to the I-phase current; k represents the order of the current Markov property; j n Indicates the J-phase current value at time n, i n+1 Indicates the I-phase current value at time n+1, Represents the I-phase current value at time n and k-1 moments before time n; Indicates i n+1 、 j n The joint probability of express and j nUnder the condition i n+1 The probability of express Under the condition i n+1 probability.

[0035] On the other hand, the present invention further provides a device for dividing a wind turbine generator fault ride-through interval, comprising:

[0036] The first determination module is used to bring the measured data of the wind turbine output end into a pre-built artificial neural network model to determine the wind turbine control implementation time after the grid fault occurs, the wind turbine control implementation time after the grid fault is cleared, the wind turbine control stabilization time after the grid fault occurs, and the wind turbine control stabilization time after the grid fault is cleared;

[0037] a division module for dividing the wind turbine fault ride-through interval based on the wind turbine control implementation time after the grid fault occurs, the wind turbine control implementation time after the grid fault is cleared, the wind turbine control stabilization time after the grid fault occurs, and the wind turbine control stabilization time after the grid fault is cleared;

[0038] The measured data includes the fundamental positive sequence value of the low-voltage side bus voltage at the output end of the motor group, the active power value and the three-phase current value.

[0039] The apparatus also includes a building block for constructing an artificial neural network model;

[0040] The building blocks include:

[0041] A first determining unit is configured to set the three-phase current at the output end of the wind turbine generator set and the transfer entropy of the three-phase current as neurons in the input layer;

[0042] A second determining unit is used to determine the neurons of the hidden layer based on the application scenario and using an error back propagation algorithm;

[0043] The third determining unit is used to set the wind turbine generator set control implementation time after the grid fault occurs, the wind turbine generator set control implementation time after the grid fault is cleared, the wind turbine generator set control stabilization time after the grid fault occurs, and the wind turbine generator set control stabilization time after the grid fault is cleared as neurons of the output layer.

[0044] The building blocks also include:

[0045] An acquisition unit is configured to acquire historical data, wherein the historical data includes three-phase current values ​​at the output end of the wind turbine generator set, time series values ​​of transfer entropy of the three-phase current, a moment when the wind turbine generator set implements control after a power grid fault occurs, a moment when the wind turbine generator set implements control after the power grid fault is cleared, a moment when the wind turbine generator set control stabilizes after a power grid fault occurs, and a moment when the wind turbine generator set control stabilizes after the power grid fault is cleared;

[0046] A division unit, configured to divide the historical data into training data and test data;

[0047] A training unit, used for training the artificial neural network model based on training data;

[0048] The testing unit is used to test the trained artificial neural network model based on the test data.

[0049] The division module includes:

[0050] A first dividing unit is configured to divide the fault-crossing interval into a pre-fault interval, a fault-occurrence interval, and a post-fault-clearing interval based on a power grid fault occurrence time and a power grid fault clearing time;

[0051] a second dividing unit, configured to divide the fault occurrence interval into a fault occurrence transient interval and a fault occurrence steady-state interval based on a control stabilization moment of the wind turbine generator set after the grid fault occurs, and to divide the fault occurrence transient interval into a fault occurrence transition transient interval and a fault occurrence control transient interval based on a control implementation moment of the wind turbine generator set after the grid fault occurs;

[0052] The third dividing unit is used to divide the post-fault-clearing interval into a post-fault-clearing transient interval and a post-fault-clearing steady-state interval based on the wind turbine control stabilization moment after the grid fault is cleared, and to divide the post-fault-clearing transient interval into a fault-clearing transition transient interval and a fault-clearing control transient interval based on the wind turbine control implementation moment after the grid fault is cleared, and to divide the post-fault-clearing steady-state interval into a fault-clearing control steady-state interval and a fault-clearing recovery steady-state interval based on the active power recovery moment after the grid fault is cleared.

[0053] The second determining module is specifically configured to:

[0054] The time when the fundamental positive sequence voltage value of the low-voltage side busbar at the output end of the wind turbine generator set is less than or equal to 11 times the rated voltage value of the low-voltage side busbar at the output end of the wind turbine generator set is taken as the time when the low-voltage fault ride-through occurs;

[0055] The time when the fundamental positive sequence voltage value of the low-voltage side busbar at the output end of the wind turbine generator set becomes greater than or equal to 11 times the rated voltage value of the low-voltage side busbar at the output end of the wind turbine generator set is taken as the low voltage fault ride-through clearing time;

[0056] The time when the fundamental positive sequence voltage value of the low-voltage busbar at the output end of the wind turbine generator set is greater than or equal to 12 times the rated voltage value of the low-voltage busbar at the output end of the wind turbine generator set is taken as the time when the high voltage fault ride-through occurs;

[0057] The time when the fundamental positive sequence voltage value of the low-voltage busbar at the output end of the wind turbine generator set is less than or equal to 12 times the rated voltage value of the low-voltage busbar at the output end of the wind turbine generator set is taken as the high voltage fault ride-through clearing time;

[0058] Wherein, l1 represents the low voltage fault occurrence judgment coefficient, l2 represents the high voltage fault occurrence judgment coefficient, and l1<1, l2>1.

[0059] The second determining module is specifically configured to:

[0060] The time corresponding to the active power value at the output of the wind turbine generator set after the grid fault is cleared is greater than or equal to 13 times the active power value at the output of the wind turbine generator set before the grid fault is cleared is taken as the active power recovery time after the grid fault is cleared;

[0061] l3 represents the active power recovery judgment coefficient, and l3<1.

[0062] The second determination module determines the time series value of the transfer entropy of the three-phase current according to the following formula:

[0063]

[0064] Where J and I represent any two different phases, J = A, B, C, I = A, B, C; T J→I represents the time series value of the transfer entropy from the J-phase current to the I-phase current; k represents the order of the current Markov property; j n Indicates the J-phase current value at time n, i n+1 Indicates the I-phase current value at time n+1, Represents the I-phase current value at time n and k-1 moments before time n; Indicates i n+1 、 j n The joint probability of express and j n Under the condition i n+1 The probability of express Under the condition i n+1 probability.

[0065] Compared with the closest existing technology, the technical solution provided by the present invention has the following beneficial effects:

[0066] The method for dividing the fault ride-through interval of a wind turbine provided by the present invention brings the measured data of the output end of the wind turbine into a pre-built artificial neural network model to determine the time when the wind turbine implements control after the grid fault occurs, the time when the wind turbine implements control after the grid fault is cleared, the time when the wind turbine control is stable after the grid fault occurs, and the time when the wind turbine control is stable after the grid fault is cleared; the wind turbine fault ride-through interval is divided based on the above four moments, and the wind turbine fault ride-through interval can be automatically divided, and has strong versatility, and can verify the control response characteristics of the simulation model;

[0067] Based on measured data from wind turbine fault ride-through tests, the present invention not only effectively addresses the issues of large errors and poor versatility in automatic interval division caused by the standard's use of fixed time lengths, but also avoids the workload of manual correction and can be applied to both high voltage ride-through and low voltage ride-through tests.

[0068] The present invention helps to improve the accuracy and simulation performance of the electrical simulation model of wind turbines, and facilitates the planning and design of wind power grid connection and the safe and stable operation of the power grid, thereby reducing safety hazards of the power system and improving economic efficiency.

[0069] The interval division of the present invention is clear and the results are accurate, meeting the requirements of engineering applications;

[0070] The present invention can be applied to electrical model verification in actual engineering, as well as to fields such as power system relay protection and equipment operation status monitoring, and has broad application prospects in the field of wind power grid-connected simulation technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 is a flow chart of a method for dividing a wind turbine generator set fault ride-through interval in an embodiment of the present invention;

[0072] Figure 2 This is a schematic diagram of on-site measurement signal points in an embodiment of the present invention;

[0073] Figure 3 Schematic diagram of an artificial neural network model according to an embodiment of the present invention;

[0074] Figure 4 Schematic diagram of fault ride-through interval division in an embodiment of the present invention. DETAILED DESCRIPTION

[0075] The present invention will be described in further detail below with reference to the accompanying drawings.

[0076] Example 1

[0077] Embodiment 1 of the present invention provides a method for dividing a wind turbine generator set fault ride-through interval, and the specific flow chart is as follows: Figure 1 The specific process is as follows:

[0078] S101: bringing the measured data of the wind turbine output end into a pre-built artificial neural network model to determine the wind turbine control implementation time after the grid fault occurs, the wind turbine control implementation time after the grid fault is cleared, the wind turbine control stabilization time after the grid fault occurs, and the wind turbine control stabilization time after the grid fault is cleared;

[0079] S102: dividing the wind turbine fault ride-through interval based on the wind turbine control implementation time after the grid fault occurs, the wind turbine control implementation time after the grid fault is cleared, the wind turbine control stabilization time after the grid fault occurs, and the wind turbine control stabilization time after the grid fault is cleared;

[0080] The measured data include the fundamental positive sequence value of the low-voltage side bus voltage at the output end of the generator group, the active power value and the three-phase current value.

[0081] The construction of the above artificial neural network model includes:

[0082] The three-phase current (I mea-a , I mea-b , I mea-c ) and the transfer entropy of the three-phase current (T A→B 、T A→C 、T B→A 、T B→C 、T C→A 、T C→B ) is set as the neuron of the input layer;

[0083] Based on the application scenario, the neurons in the hidden layer are determined using the error back propagation algorithm;

[0084] The wind turbine control implementation moment after the grid fault occurs, the wind turbine control implementation moment after the grid fault is cleared, the wind turbine control stabilization moment after the grid fault occurs, and the wind turbine control stabilization moment after the grid fault is cleared are set as neurons in the output layer.

[0085] The construction of artificial neural network model also includes:

[0086] Acquire historical data, including three-phase current values ​​at the output end of the wind turbine generator set, time series values ​​of transfer entropy of the three-phase current, the moment when the wind turbine generator set implements control after a power grid fault occurs, the moment when the wind turbine generator set implements control after the power grid fault is cleared, the moment when the wind turbine generator set control stabilizes after the power grid fault occurs, and the moment when the wind turbine generator set control stabilizes after the power grid fault is cleared;

[0087] Divide historical data into training data and test data;

[0088] Training the artificial neural network model based on the training data;

[0089] The trained artificial neural network model is tested based on the test data.

[0090] In the above S102, the wind turbine fault ride-through interval is divided based on the wind turbine control implementation time after the grid fault occurs, the wind turbine control implementation time after the grid fault is cleared, the wind turbine control stable time after the grid fault occurs, and the wind turbine control stable time after the grid fault is cleared, specifically as follows: Figure 4 As shown, t fa is the time when the fault occurs, t fb is the fault clearing time, t c1 is the time when the wind turbine is controlled after the grid fault occurs, t c2 is the time when the wind turbine is controlled after the grid fault is cleared, t s1 is the wind turbine control stabilization time after the grid fault occurs, t s2 is the wind turbine control stabilization moment after the grid fault is cleared, t pr The active power recovery time after the grid fault is cleared. The specific process of S102 is as follows:

[0091] Based on the grid fault occurrence time and the grid fault clearing time, the fault ride-through interval is divided into the pre-fault interval, the fault occurrence interval and the post-fault clearing interval;

[0092] Dividing the fault occurrence interval into a fault occurrence transient interval and a fault occurrence steady-state interval based on the wind turbine control stabilization moment after the grid fault occurs, and dividing the fault occurrence transient interval into a fault occurrence transition transient interval and a fault occurrence control transient interval based on the wind turbine control implementation moment after the grid fault occurs;

[0093] Based on the moment when the wind turbine control stabilizes after the grid fault is cleared, the post-fault clearing interval is divided into a post-fault clearing transient interval and a post-fault clearing steady-state interval. Based on the moment when the wind turbine implements control after the grid fault is cleared, the post-fault clearing transient interval is divided into a fault clearing transition transient interval and a fault clearing control transient interval. Based on the moment when the active power is restored after the grid fault is cleared, the post-fault clearing steady-state interval is divided into a fault clearing control steady-state interval and a fault clearing recovery steady-state interval.

[0094] The determination of the above-mentioned grid fault occurrence time and grid fault recovery time includes:

[0095] The time when the fundamental positive sequence voltage value of the low-voltage side busbar at the output end of the wind turbine generator set is less than or equal to 11 times the rated voltage value of the low-voltage side busbar at the output end of the wind turbine generator set is taken as the time when the low-voltage fault ride-through occurs;

[0096] The time when the fundamental positive sequence voltage value of the low-voltage side busbar at the output end of the wind turbine generator set becomes greater than or equal to 11 times the rated voltage value of the low-voltage side busbar at the output end of the wind turbine generator set is taken as the low voltage fault ride-through clearing time;

[0097] The time when the fundamental positive sequence voltage value of the low-voltage busbar at the output end of the wind turbine generator set is greater than or equal to 12 times the rated voltage value of the low-voltage busbar at the output end of the wind turbine generator set is taken as the time when the high voltage fault ride-through occurs;

[0098] The time when the fundamental positive sequence voltage value of the low-voltage busbar at the output end of the wind turbine generator set is less than or equal to 12 times the rated voltage value of the low-voltage busbar at the output end of the wind turbine generator set is taken as the high voltage fault ride-through clearing time;

[0099] Wherein, l1 represents a low voltage fault occurrence judgment coefficient, l2 represents a high voltage fault occurrence judgment coefficient, and l1<1, l2>1. In embodiment 1 of the present invention, l1 is 0.85, and l2 is 1.15.

[0100] The determination of the active power restoration time after the above-mentioned grid fault is cleared includes:

[0101] The time corresponding to the active power value at the output of the wind turbine generator set after the grid fault is cleared is greater than or equal to 13 times the active power value at the output of the wind turbine generator set before the grid fault is cleared is taken as the active power recovery time after the grid fault is cleared;

[0102] l3 represents the active power recovery judgment coefficient, and l3<1. In embodiment 1 of the present invention, l3 is set to 0.9.

[0103] The time series value of the transfer entropy of the above three-phase current is determined by the following formula:

[0104]

[0105] Where J and I represent any two different phases, J = A, B, C, I = A, B, C; T J→I represents the time series value of the transfer entropy from the J-phase current to the I-phase current; k represents the order of the current Markov property; j n Indicates the J-phase current value at time n, i n+1 Indicates the I-phase current value at time n+1, Represents the I-phase current value at time n and k-1 moments before time n; Indicates i n+1 、 j n The joint probability of express and j n Under the condition i n+1 The probability of express Under the condition i n+1 probability.

[0106] Example 2

[0107] Based on the same inventive concept, embodiment 2 of the present invention further provides a device for dividing a wind turbine generator fault ride-through interval. The functions of the above modules are described in detail below:

[0108] The first determination module is used to bring the measured data of the wind turbine output end into a pre-built artificial neural network model to determine the wind turbine control implementation time after the grid fault occurs, the wind turbine control implementation time after the grid fault is cleared, the wind turbine control stabilization time after the grid fault occurs, and the wind turbine control stabilization time after the grid fault is cleared;

[0109] The division module is used to divide the wind turbine fault ride-through interval based on the wind turbine control implementation time after the grid fault occurs, the wind turbine control implementation time after the grid fault is cleared, the wind turbine control stabilization time after the grid fault occurs, and the wind turbine control stabilization time after the grid fault is cleared.

[0110] The measured data include the fundamental positive sequence value of the low-voltage side bus voltage at the output end of the generator group, the active power value and the three-phase current value.

[0111] The apparatus provided in embodiment 2 of the present invention further includes a construction module for constructing an artificial neural network model;

[0112] The building blocks include:

[0113] A first determining unit is configured to set the three-phase current at the output end of the wind turbine generator set and the transfer entropy of the three-phase current as neurons in the input layer;

[0114] A second determining unit is used to determine the neurons of the hidden layer based on the application scenario and using an error back propagation algorithm;

[0115] The third determining unit is used to set the wind turbine generator set control implementation time after the grid fault occurs, the wind turbine generator set control implementation time after the grid fault is cleared, the wind turbine generator set control stabilization time after the grid fault occurs, and the wind turbine generator set control stabilization time after the grid fault is cleared as neurons of the output layer.

[0116] The aforementioned building blocks also include:

[0117] An acquisition unit is configured to acquire historical data, wherein the historical data includes three-phase current values ​​at the output end of the wind turbine generator set, time series values ​​of transfer entropy of the three-phase current, a moment when the wind turbine generator set implements control after a power grid fault occurs, a moment when the wind turbine generator set implements control after the power grid fault is cleared, a moment when the wind turbine generator set control stabilizes after a power grid fault occurs, and a moment when the wind turbine generator set control stabilizes after the power grid fault is cleared;

[0118] A partitioning unit, used to divide historical data into training data and test data;

[0119] A training unit, used for training the artificial neural network model based on training data;

[0120] The testing unit is used to test the trained artificial neural network model based on the test data.

[0121] The division modules include:

[0122] A first dividing unit is configured to divide the fault-crossing interval into a pre-fault interval, a fault-occurrence interval, and a post-fault-clearing interval based on a power grid fault occurrence time and a power grid fault clearing time;

[0123] a second dividing unit, configured to divide the fault occurrence interval into a fault occurrence transient interval and a fault occurrence steady-state interval based on a control stabilization moment of the wind turbine generator set after the grid fault occurs, and to divide the fault occurrence transient interval into a fault occurrence transition transient interval and a fault occurrence control transient interval based on a control implementation moment of the wind turbine generator set after the grid fault occurs;

[0124] The third dividing unit is used to divide the post-fault-clearing interval into a post-fault-clearing transient interval and a post-fault-clearing steady-state interval based on the wind turbine control stabilization moment after the grid fault is cleared, and to divide the post-fault-clearing transient interval into a fault-clearing transition transient interval and a fault-clearing control transient interval based on the wind turbine control implementation moment after the grid fault is cleared, and to divide the post-fault-clearing steady-state interval into a fault-clearing control steady-state interval and a fault-clearing recovery steady-state interval based on the active power recovery moment after the grid fault is cleared.

[0125] The second determining module is specifically configured to:

[0126] The time when the fundamental positive sequence voltage value of the low-voltage side busbar at the output end of the wind turbine generator set is less than or equal to 11 times the rated voltage value of the low-voltage side busbar at the output end of the wind turbine generator set is taken as the time when the low-voltage fault ride-through occurs;

[0127] The time when the fundamental positive sequence voltage value of the low-voltage side busbar at the output end of the wind turbine generator set becomes greater than or equal to 11 times the rated voltage value of the low-voltage side busbar at the output end of the wind turbine generator set is taken as the low voltage fault ride-through clearing time;

[0128] The time when the fundamental positive sequence voltage value of the low-voltage busbar at the output end of the wind turbine generator set is greater than or equal to 12 times the rated voltage value of the low-voltage busbar at the output end of the wind turbine generator set is taken as the time when the high voltage fault ride-through occurs;

[0129] The time when the fundamental positive sequence voltage value of the low-voltage busbar at the output end of the wind turbine generator set is less than or equal to 12 times the rated voltage value of the low-voltage busbar at the output end of the wind turbine generator set is taken as the high voltage fault ride-through clearing time;

[0130] Wherein, l1 represents the low voltage fault occurrence judgment coefficient, l2 represents the high voltage fault occurrence judgment coefficient, and l1<1, l2>1.

[0131] The second determining module is specifically configured to:

[0132] The time corresponding to the active power value at the output of the wind turbine generator set after the grid fault is cleared is greater than or equal to 13 times the active power value at the output of the wind turbine generator set before the grid fault is cleared is taken as the active power recovery time after the grid fault is cleared;

[0133] l3 represents the active power recovery judgment coefficient, and l3<1.

[0134] The second determination module determines the time series value of the transfer entropy of the three-phase current according to the following formula:

[0135]

[0136] Where J and I represent any two different phases, J = A, B, C, I = A, B, C; T J→I represents the time series value of the transfer entropy from the J-phase current to the I-phase current; k represents the order of the current Markov property; j n Indicates the J-phase current value at time n, i n+1 Indicates the I-phase current value at time n+1, Represents the I-phase current value at time n and k-1 moments before time n; Indicates i n+1 、 j n The joint probability of express and j n Under the condition i n+1 The probability of express Under the condition i n+1 probability.

[0137] For the convenience of description, the various parts of the above-mentioned device are divided into various modules or units according to their functions and described separately. Of course, when implementing this application, the functions of each module or unit can be implemented in the same or multiple software or hardware.

[0138] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0139] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0140] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0141] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0142] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Ordinary technicians in the relevant field can still modify or replace the specific implementation methods of the present invention with equivalents by referring to the above embodiments. Any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention are within the scope of protection of the claims of the present invention to be approved.

Claims

1. A method for dividing a wind turbine fault ride-through interval, characterized in that: include: The measured data at the output of the wind turbine generator set is fed into a pre-built artificial neural network model to determine the wind turbine generator set control implementation time after a grid fault occurs, the wind turbine generator set control implementation time after the grid fault is cleared, the wind turbine generator set control stabilization time after a grid fault occurs, and the wind turbine generator set control stabilization time after the grid fault is cleared; The fault ride-through interval of the wind turbine is divided based on the measured data at the output end of the wind turbine, the time when the wind turbine implements control after the grid fault occurs, the time when the wind turbine implements control after the grid fault is cleared, the time when the wind turbine control is stable after the grid fault occurs, and the time when the wind turbine control is stable after the grid fault is cleared; The measured data include the fundamental positive sequence value of the low-voltage side bus voltage, the active power value and the three-phase current value at the output end of the motor group; The construction of the artificial neural network model includes: The three-phase current at the output of the wind turbine and the transfer entropy of the three-phase current are set as neurons in the input layer; Based on the application scenario, the neurons in the hidden layer are determined using the error back propagation algorithm; The wind turbine control implementation time after the grid fault occurs, the wind turbine control implementation time after the grid fault is cleared, the wind turbine control stabilization time after the grid fault occurs, and the wind turbine control stabilization time after the grid fault is cleared are set as neurons in the output layer; The fault ride-through interval of wind turbines is divided based on the wind turbine control implementation time and wind turbine control stability time, including: Based on the grid fault occurrence time and the grid fault clearing time, the fault ride-through interval is divided into the pre-fault interval, the fault occurrence interval and the post-fault clearing interval; Dividing the fault occurrence interval into a fault occurrence transient interval and a fault occurrence steady-state interval based on the wind turbine control stabilization moment after the grid fault occurs, and dividing the fault occurrence transient interval into a fault occurrence transition transient interval and a fault occurrence control transient interval based on the wind turbine control implementation moment after the grid fault occurs; Based on the moment when the wind turbine control stabilizes after the grid fault is cleared, the post-fault clearing interval is divided into a post-fault clearing transient interval and a post-fault clearing steady-state interval. Based on the moment when the wind turbine implements control after the grid fault is cleared, the post-fault clearing transient interval is divided into a fault clearing transition transient interval and a fault clearing control transient interval. Based on the moment when the active power is restored after the grid fault is cleared, the post-fault clearing steady-state interval is divided into a fault clearing control steady-state interval and a fault clearing recovery steady-state interval.

2. The method for dividing the fault ride-through interval of a wind turbine generator set according to claim 1, characterized in that: The construction of the artificial neural network model also includes: Acquire historical data, the historical data including three-phase current values ​​at the output end of the wind turbine generator set, time series values ​​of transfer entropy of the three-phase current, a time instant when the wind turbine generator set implements control after a power grid fault occurs, a time instant when the wind turbine generator set implements control after the power grid fault is cleared, a time instant when the wind turbine generator set control stabilizes after the power grid fault occurs, and a time instant when the wind turbine generator set control stabilizes after the power grid fault is cleared; Dividing the historical data into training data and test data; Using training data to train the artificial neural network model; The trained artificial neural network model is tested based on the test data.

3. The method for dividing the fault ride-through interval of a wind turbine generator set according to claim 1, characterized in that: The determination of the grid fault occurrence time and the grid fault recovery time includes: The time when the fundamental positive sequence voltage value of the low-voltage side busbar at the output end of the wind turbine generator set is less than or equal to 11 times the rated voltage value of the low-voltage side busbar at the output end of the wind turbine generator set is taken as the time when the low-voltage fault ride-through occurs; The time when the fundamental positive sequence voltage value of the low-voltage side busbar at the output end of the wind turbine generator set becomes greater than or equal to 11 times the rated voltage value of the low-voltage side busbar at the output end of the wind turbine generator set is taken as the low voltage fault ride-through clearing time; The time when the fundamental positive sequence voltage value of the low-voltage busbar at the output end of the wind turbine generator set is greater than or equal to 12 times the rated voltage value of the low-voltage busbar at the output end of the wind turbine generator set is taken as the time when the high voltage fault ride-through occurs; The time when the fundamental positive sequence voltage value of the low-voltage busbar at the output end of the wind turbine generator set is less than or equal to 12 times the rated voltage value of the low-voltage busbar at the output end of the wind turbine generator set is taken as the high voltage fault ride-through clearing time; Wherein, l1 represents the low voltage fault occurrence judgment coefficient, l2 represents the high voltage fault occurrence judgment coefficient, and l1<1, l2>1.

4. The method for dividing the fault ride-through interval of a wind turbine generator set according to claim 1, characterized in that: Determining the active power recovery time after the grid fault is cleared includes: The time corresponding to the active power value at the output of the wind turbine generator set after the grid fault is cleared is greater than or equal to 13 times the active power value at the output of the wind turbine generator set before the grid fault is cleared is taken as the active power recovery time after the grid fault is cleared; l3 represents the active power recovery judgment coefficient, and l3<1.

5. The method for dividing the fault ride-through interval of a wind turbine generator set according to claim 2, characterized in that: The transfer entropy time series value of the three-phase current is determined by the following formula: Where J and I represent any two different phases, J = A, B, C, I = A, B, C; T J→I represents the time series value of the transfer entropy from the J-phase current to the I-phase current; k represents the order of the current Markov property; j n Indicates the J-phase current value at time n, i n+1 Indicates the I-phase current value at time n+1, Represents the I-phase current value at time n and k-1 moments before time n; Indicates i n+1 、 j n The joint probability of express and j n Under the condition i n+1 The probability of express Under the condition i n+1 probability.

6. A device for dividing fault ride-through intervals of a wind turbine generator set, characterized in that: include: The first determination module is used to bring the measured data of the wind turbine output end into a pre-built artificial neural network model to determine the wind turbine control implementation time after the grid fault occurs, the wind turbine control implementation time after the grid fault is cleared, the wind turbine control stabilization time after the grid fault occurs, and the wind turbine control stabilization time after the grid fault is cleared; a partitioning module for partitioning the fault ride-through interval of the wind turbine generator set based on measured data at the output end of the wind turbine generator set, the time when the wind turbine generator set implements control after a power grid fault occurs, the time when the wind turbine generator set implements control after the power grid fault is cleared, the time when the wind turbine generator set control is stable after a power grid fault occurs, and the time when the wind turbine generator set control is stable after the power grid fault is cleared; The measured data include the fundamental positive sequence value of the low-voltage side bus voltage, the active power value and the three-phase current value at the output end of the motor group; The apparatus also includes a building block for constructing an artificial neural network model; The building blocks include: A first determining unit is configured to set the three-phase current at the output end of the wind turbine generator set and the transfer entropy of the three-phase current as neurons in the input layer; A second determining unit is used to determine the neurons of the hidden layer based on the application scenario and using an error back propagation algorithm; a third determining unit, configured to set the wind turbine generator control implementation time after a power grid fault occurs, the wind turbine generator control implementation time after a power grid fault is cleared, the wind turbine generator control stabilization time after a power grid fault occurs, and the wind turbine generator control stabilization time after a power grid fault is cleared as neurons in the output layer; The division module includes: A first dividing unit is configured to divide the fault-crossing interval into a pre-fault interval, a fault-occurrence interval, and a post-fault-clearing interval based on a power grid fault occurrence time and a power grid fault clearing time; a second dividing unit, configured to divide the fault occurrence interval into a fault occurrence transient interval and a fault occurrence steady-state interval based on a control stabilization moment of the wind turbine generator set after the grid fault occurs, and to divide the fault occurrence transient interval into a fault occurrence transition transient interval and a fault occurrence control transient interval based on a control implementation moment of the wind turbine generator set after the grid fault occurs; The third dividing unit is used to divide the post-fault-clearing interval into a post-fault-clearing transient interval and a post-fault-clearing steady-state interval based on the wind turbine control stabilization moment after the grid fault is cleared, and to divide the post-fault-clearing transient interval into a fault-clearing transition transient interval and a fault-clearing control transient interval based on the wind turbine control implementation moment after the grid fault is cleared, and to divide the post-fault-clearing steady-state interval into a fault-clearing control steady-state interval and a fault-clearing recovery steady-state interval based on the active power recovery moment after the grid fault is cleared.

7. The device for dividing the fault ride-through interval of a wind turbine generator set according to claim 6, characterized in that: The building blocks also include: an acquisition unit, configured to acquire historical data, the historical data including three-phase current values ​​at the output end of the wind turbine generator set, a time series value of transfer entropy of the three-phase current, a moment when the wind turbine generator set implements control after a power grid fault occurs, a moment when the wind turbine generator set implements control after the power grid fault is cleared, a moment when the wind turbine generator set stabilizes control after a power grid fault occurs, and a moment when the wind turbine generator set stabilizes control after the power grid fault is cleared; A division unit, configured to divide the historical data into training data and test data; A training unit, used for training the artificial neural network model using training data; The testing unit is used to test the trained artificial neural network model based on the test data.

8. The device for dividing the fault ride-through interval of a wind turbine generator set according to claim 6, characterized in that: The determination of the grid fault occurrence time and the grid fault recovery time in the division module includes: The time when the fundamental positive sequence voltage value of the low-voltage side busbar at the output end of the wind turbine generator set is less than or equal to 11 times the rated voltage value of the low-voltage side busbar at the output end of the wind turbine generator set is taken as the time when the low-voltage fault ride-through occurs; The time when the fundamental positive sequence voltage value of the low-voltage side busbar at the output end of the wind turbine generator set becomes greater than or equal to 11 times the rated voltage value of the low-voltage side busbar at the output end of the wind turbine generator set is taken as the low voltage fault ride-through clearing time; The time when the fundamental positive sequence voltage value of the low-voltage busbar at the output end of the wind turbine generator set is greater than or equal to 12 times the rated voltage value of the low-voltage busbar at the output end of the wind turbine generator set is taken as the time when the high voltage fault ride-through occurs; The time when the fundamental positive sequence voltage value of the low-voltage busbar at the output end of the wind turbine generator set is less than or equal to 12 times the rated voltage value of the low-voltage busbar at the output end of the wind turbine generator set is taken as the high voltage fault ride-through clearing time; Wherein, l1 represents the low voltage fault occurrence judgment coefficient, l2 represents the high voltage fault occurrence judgment coefficient, and l1<1, l2>1.

9. The device for dividing the fault ride-through interval of a wind turbine generator set according to claim 6, characterized in that: Determining the active power recovery time after the grid fault is cleared in the third division unit includes: The time corresponding to the active power value at the output of the wind turbine generator set after the grid fault is cleared is greater than or equal to 13 times the active power value at the output of the wind turbine generator set before the grid fault is cleared is taken as the active power recovery time after the grid fault is cleared; l3 represents the active power recovery judgment coefficient, and l3<1.

10. The device for dividing the fault ride-through interval of a wind turbine generator set according to claim 7, characterized in that: The acquisition unit determines the time series value of the transfer entropy of the three-phase current according to the following formula: Where J and I represent any two different phases, J = A, B, C, I = A, B, C; T J→I represents the time series value of the transfer entropy from the J-phase current to the I-phase current; k represents the order of the current Markov property; j n Indicates the J-phase current value at time n, i n+1 Indicates the I-phase current value at time n+1, Represents the I-phase current value at time n and k-1 moments before time n; Indicates i n+1 、 j n The joint probability of express and j n Under the condition i n+1 The probability of express Under the condition i n+1 probability.

Images